very cool!
@clefourrier\n\n\t \n"}},{"time":"2024-05-14T21:26:12.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"356171965298722","content":[{"type":"text","value":"What are you excited about from Google I/O?","raw":"What are you excited about from Google I/O?"}],"rawContent":"What are you excited about from Google I/O?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👀","users":["victor","julien-c","ramiam"],"count":3}],"publishedAt":"2024-05-14T18:28:01.000Z","updatedAt":"2024-05-15T20:18:33.170Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1641697287903-5eb8ddf64e876668a0c377c0.png","fullname":"Max Woolf","name":"minimaxir","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/50fec2cd5a1ab2a17d10014cf85b7bcf.svg","fullname":"Jayeeta P","name":"JayeetaP","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/1631fbb3c64d98f082144706100d5bc8.svg","fullname":"Niccolo minetti","name":"nhminetz","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/db0615d2ffeb12fac1d3a797672511fb.svg","fullname":"FS_Hri","name":"hrishbhdalal","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/356171965298722","totalUniqueImpressions":1560,"numComments":9},"comment":{"_id":"6643d6f434fff844a50973e1","raw":"https://huggingface.co/posts/merve/375349782904361","html":"
view reply
view reply
view post
view post
view reply
view post
view reply
view reply
view post
view reply
view reply
view reply
view post
view reply
view reply
view reply
view reply
view reply
view reply
view post
view post
view reply
view post
view post
view post
view reply
view reply
view post
view reply
view reply
view reply
view reply
view reply
view reply
view reply
view reply
view reply
view post
view post
view post
view reply
view post
view post
view reply
view reply
view post
view reply
view post
view reply
https://huggingface.co/posts/merve/375349782904361
\n"}},{"time":"2024-05-14T18:44:58.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"356171965298722","content":[{"type":"text","value":"What are you excited about from Google I/O?","raw":"What are you excited about from Google I/O?"}],"rawContent":"What are you excited about from Google I/O?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👀","users":["victor","julien-c","ramiam"],"count":3}],"publishedAt":"2024-05-14T18:28:01.000Z","updatedAt":"2024-05-15T20:18:33.170Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1641697287903-5eb8ddf64e876668a0c377c0.png","fullname":"Max Woolf","name":"minimaxir","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/50fec2cd5a1ab2a17d10014cf85b7bcf.svg","fullname":"Jayeeta P","name":"JayeetaP","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/1631fbb3c64d98f082144706100d5bc8.svg","fullname":"Niccolo minetti","name":"nhminetz","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/db0615d2ffeb12fac1d3a797672511fb.svg","fullname":"FS_Hri","name":"hrishbhdalal","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/356171965298722","totalUniqueImpressions":1560,"numComments":9},"comment":{"_id":"6643b12a31f4ea1730a3210b","raw":"any info on when it's going to be released though?","html":"any info on when it's going to be released though?
\n"}},{"time":"2024-05-14T18:32:11.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"356171965298722","content":[{"type":"text","value":"What are you excited about from Google I/O?","raw":"What are you excited about from Google I/O?"}],"rawContent":"What are you excited about from Google I/O?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👀","users":["victor","julien-c","ramiam"],"count":3}],"publishedAt":"2024-05-14T18:28:01.000Z","updatedAt":"2024-05-15T20:18:33.170Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1641697287903-5eb8ddf64e876668a0c377c0.png","fullname":"Max Woolf","name":"minimaxir","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/50fec2cd5a1ab2a17d10014cf85b7bcf.svg","fullname":"Jayeeta P","name":"JayeetaP","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/1631fbb3c64d98f082144706100d5bc8.svg","fullname":"Niccolo minetti","name":"nhminetz","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/db0615d2ffeb12fac1d3a797672511fb.svg","fullname":"FS_Hri","name":"hrishbhdalal","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/356171965298722","totalUniqueImpressions":1560,"numComments":9},"comment":{"_id":"6643ae2bfb769dad650c7783","raw":"do you have a link?","html":"do you have a link?
\n"}},{"time":"2024-05-14T18:28:01.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"356171965298722","content":[{"type":"text","value":"What are you excited about from Google I/O?","raw":"What are you excited about from Google I/O?"}],"rawContent":"What are you excited about from Google I/O?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👀","users":["victor","julien-c","ramiam"],"count":3}],"publishedAt":"2024-05-14T18:28:01.000Z","updatedAt":"2024-05-15T20:18:33.170Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1641697287903-5eb8ddf64e876668a0c377c0.png","fullname":"Max Woolf","name":"minimaxir","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/50fec2cd5a1ab2a17d10014cf85b7bcf.svg","fullname":"Jayeeta P","name":"JayeetaP","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/1631fbb3c64d98f082144706100d5bc8.svg","fullname":"Niccolo minetti","name":"nhminetz","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/db0615d2ffeb12fac1d3a797672511fb.svg","fullname":"FS_Hri","name":"hrishbhdalal","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/356171965298722","totalUniqueImpressions":1560,"numComments":9}},{"time":"2024-05-13T23:28:38.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"373085132561453","content":[{"type":"text","value":"Hey everyone,","raw":"Hey everyone,"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Just wanted to shout out a massive thank you to all 2000 of you who've followed me on Hugging Face! 🎉 It's incredible to have such an awesome crew backing me up as I dive into all these LLM experiments.","raw":"Just wanted to shout out a massive thank you to all 2000 of you who've followed me on Hugging Face! 🎉 It's incredible to have such an awesome crew backing me up as I dive into all these LLM experiments."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Even though not all my models turn out perfect, I've found some real gems and methods along the way 💎. It's like digging for treasure – sometimes you found nothing, but sometimes you find a pearl, and sometimes you find a new method to try.","raw":"Even though not all my models turn out perfect, I've found some real gems and methods along the way 💎. It's like digging for treasure – sometimes you found nothing, but sometimes you find a pearl, and sometimes you find a new method to try."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Your support and encouragement mean the world to me, and I'm really stoked to keep experimenting and learning. If you told me some years ago I would have so much people following me for what I do, I wouldn't have believed it. Here's to more discoveries and adventures ahead! 🚀","raw":"Your support and encouragement mean the world to me, and I'm really stoked to keep experimenting and learning. If you told me some years ago I would have so much people following me for what I do, I wouldn't have believed it. Here's to more discoveries and adventures ahead! 🚀"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Also, big thanks once again, and a huge shoutout to ","raw":"Also, big thanks once again, and a huge shoutout to "},{"type":"mention","user":"IkariDev","raw":"@IkariDev"},{"type":"text","value":" for being there through this journey and supporting me. I'm excited for our future work together and hope we will continue to make people happy! 👏","raw":" for being there through this journey and supporting me. I'm excited for our future work together and hope we will continue to make people happy! 👏"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"I want to thank ","raw":"I want to thank "},{"type":"mention","user":"Gryphe","raw":"@Gryphe"},{"type":"text","value":" too, since my early work was heavily inspired from MythoMax and the RP/ERP vibe of it. If I'm here today it's probably because of you 😂","raw":" too, since my early work was heavily inspired from MythoMax and the RP/ERP vibe of it. If I'm here today it's probably because of you 😂"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"I was so close to forget ","raw":"I was so close to forget "},{"type":"mention","user":"chargoddard","raw":"@chargoddard"},{"type":"text","value":" and his amazing tool too! What will we do without mergekit in our life? Thank you! 🙏","raw":" and his amazing tool too! What will we do without mergekit in our life? Thank you! 🙏"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"See y'all at 3k!","raw":"See y'all at 3k!"}],"rawContent":"Hey everyone,\n\nJust wanted to shout out a massive thank you to all 2000 of you who've followed me on Hugging Face! 🎉 It's incredible to have such an awesome crew backing me up as I dive into all these LLM experiments.\n\nEven though not all my models turn out perfect, I've found some real gems and methods along the way 💎. It's like digging for treasure – sometimes you found nothing, but sometimes you find a pearl, and sometimes you find a new method to try.\n\nYour support and encouragement mean the world to me, and I'm really stoked to keep experimenting and learning. If you told me some years ago I would have so much people following me for what I do, I wouldn't have believed it. Here's to more discoveries and adventures ahead! 🚀\n\nAlso, big thanks once again, and a huge shoutout to @IkariDev for being there through this journey and supporting me. I'm excited for our future work together and hope we will continue to make people happy! 👏\n\nI want to thank @Gryphe too, since my early work was heavily inspired from MythoMax and the RP/ERP vibe of it. If I'm here today it's probably because of you 😂\n\nI was so close to forget @chargoddard and his amazing tool too! What will we do without mergekit in our life? Thank you! 🙏\n\nSee y'all at 3k!","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ab1241ad514ca8d1430003/d-43TcOxG-zqAbzrH2m7H.png","fullname":"Undi","name":"Undi95","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/630495b0ce6b12280b193c25/bT61kBtQPhDYk00AI0o0g.png","fullname":"Charles Goddard","name":"chargoddard","type":"user","isPro":true,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64ae4107ad6218d51a2a7d0c/3dcor68aYBKEcTlOUHJpK.png","fullname":"Gryphe Padar","name":"Gryphe","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1667690511692-630dfb008df86f1e5becadc3.png","fullname":"IkariDev","name":"IkariDev","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"👍","users":["thedarktrumpet","Gryphe","victor","EloyOn","nisten","ajibawa-2023","PowerBot","taufiqdp","samiraeli","ZootZootTesla","Anarchist","osanseviero","EdsonT","samusenps","PopHorn1956","AuraRuby","Novelo","DataSoul","kramp","SkySyrup","lysandre","DutchMjr","clem","ineap09","Winnougan","sn2234","SerialKicked","FuturisticVibes","Josephgflowers","nikodemos71","coffeedean","HalleyStarbun","Eilian","Salekeen","woolfog","dashfunnydashdash","SnakyMcSnekFace","wunderschnitzel","MateoTeo","kamiao","jce1971"],"count":41},{"reaction":"❤️","users":["Skydea","victor","Severian","ZootZootTesla","Anarchist","osanseviero","Novelo","SkySyrup","lysandre","OrangeApples","morgul","clem","FallenMerick","gghfez","Meggido","FuturisticVibes","spuun","HalleyStarbun","Ripter001","xpgx1","Fortuze","MateoTeo","kamiao"],"count":23},{"reaction":"🤗","users":["FOX18067890","lysandre","clem","Novelo","FuturisticVibes","AtonMountlook","HalleyStarbun","xpgx1","SicariusSicariiStuff"],"count":9},{"reaction":"🔥","users":["fluffyone"],"count":1}],"publishedAt":"2024-05-12T10:55:57.000Z","updatedAt":"2024-05-15T19:06:41.037Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1667690511692-630dfb008df86f1e5becadc3.png","fullname":"IkariDev","name":"IkariDev","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64b55879f2858e648d5460fb/ZrkS_BmET_WIXZCS_q6Bv.png","fullname":"OlO","name":"010O11","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64ae4107ad6218d51a2a7d0c/3dcor68aYBKEcTlOUHJpK.png","fullname":"Gryphe Padar","name":"Gryphe","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/660432d2d2e59abb3fd40b8c/TrvNnR8wHDh9lPHm81JfQ.png","fullname":"David Meriwether","name":"BigHuggyD","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/Undi95/373085132561453","totalUniqueImpressions":16384,"numComments":5},"comment":{"_id":"6642a226355e6a18fd429535","raw":"congrats!","html":"congrats!
\n"}},{"time":"2024-05-13T23:27:57.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"223614613791481","content":[{"type":"new_line","raw":"\n"},{"type":"text","value":"Hey everyone! Just wanted to share this awesome dataset that features over 1 million tokens specifically for the Egyptian dialect. Check it ","raw":"Hey everyone! Just wanted to share this awesome dataset that features over 1 million tokens specifically for the Egyptian dialect. Check it "},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"dataset","id":"HeshamHaroon/1milion_token_EGY_songs"},"url":"https://huggingface.co/datasets/HeshamHaroon/1milion_token_EGY_songs","raw":"https://huggingface.co/datasets/HeshamHaroon/1milion_token_EGY_songs"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"}],"rawContent":"\nHey everyone! Just wanted to share this awesome dataset that features over 1 million tokens specifically for the Egyptian dialect. Check it \nhttps://huggingface.co/datasets/HeshamHaroon/1milion_token_EGY_songs\n\n","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63881c1f3143f4706311c2d0/JeZnhiigCg3tBmEuK83QO.jpeg","fullname":"Hesham Haroon","name":"HeshamHaroon","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👍","users":["victor","osanseviero","MrArray22","bodaaa","clem","rahim707"],"count":6},{"reaction":"🤝","users":["victor","clem"],"count":2}],"publishedAt":"2024-05-13T12:51:26.000Z","updatedAt":"2024-05-13T23:27:57.348Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/HeshamHaroon/223614613791481","totalUniqueImpressions":1756,"numComments":1},"comment":{"_id":"6642a1fdafb424f68ec60941","raw":"very cool, thanks for sharing!","html":"very cool, thanks for sharing!
\n"}},{"time":"2024-05-13T23:27:00.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"431728663194168","content":[{"type":"text","value":"Is GPT-4o everything you expected? 🤔","raw":"Is GPT-4o everything you expected? 🤔"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"mention","user":"OpenAI","raw":"@OpenAI"},{"type":"text","value":" has gone omni (GPT-4\"o\" 🌐), a multimodal LLM, it accepts as input any combination of text, audio, and image and generates any combination of text, audio, and image outputs. 🎤📸✏️","raw":" has gone omni (GPT-4\"o\" 🌐), a multimodal LLM, it accepts as input any combination of text, audio, and image and generates any combination of text, audio, and image outputs. 🎤📸✏️"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ Based on the examples seen:","raw":"1️⃣ Based on the examples seen:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Inputs possible are Text ✏️, Text + Image 📝🖼️, Text + Audio 📝🎧, Text + Video 📝🎥, Audio 🎧","raw":"Inputs possible are Text ✏️, Text + Image 📝🖼️, Text + Audio 📝🎧, Text + Video 📝🎥, Audio 🎧"},{"type":"new_line","raw":"\n"},{"type":"text","value":"and outputs possible are Image 🖼️, Image + Text 🖼️📝, Text 📝, Audio 🎧","raw":"and outputs possible are Image 🖼️, Image + Text 🖼️📝, Text 📝, Audio 🎧"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"2️⃣ 88.7% on MMLU 🏆; 90.2% on HumanEval (best in class) 🥇","raw":"2️⃣ 88.7% on MMLU 🏆; 90.2% on HumanEval (best in class) 🥇"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"3️⃣ Up to 50% cheaper 💸 and 2x faster ⚡ than GPT-4 Turbo","raw":"3️⃣ Up to 50% cheaper 💸 and 2x faster ⚡ than GPT-4 Turbo"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"4️⃣ GPT-4o will be available in the free tier of ChatGPT 🎉","raw":"4️⃣ GPT-4o will be available in the free tier of ChatGPT 🎉"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"5️⃣ Near real-time audio with 320ms on average, similar to human conversation 🗣️**","raw":"5️⃣ Near real-time audio with 320ms on average, similar to human conversation 🗣️**"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"6️⃣ New tokenizer with a 200k token vocabulary 📚 (previously 100k vocabulary) leading to 1.1x - 4.4x fewer tokens needed across 20 languages 🌍","raw":"6️⃣ New tokenizer with a 200k token vocabulary 📚 (previously 100k vocabulary) leading to 1.1x - 4.4x fewer tokens needed across 20 languages 🌍"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"7️⃣ Tokenizer compression and more efficient across non-English languages (3-5 times fewer tokens for major Indian languages 🇮🇳)","raw":"7️⃣ Tokenizer compression and more efficient across non-English languages (3-5 times fewer tokens for major Indian languages 🇮🇳)"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"👐Open questions:","raw":"👐Open questions:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- What is the context length? ❓","raw":"- What is the context length? ❓"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Why does GPT-4 still exist, if GPT-4o is better, faster, and cheaper? 🤨","raw":"- Why does GPT-4 still exist, if GPT-4o is better, faster, and cheaper? 🤨"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Blog: ","raw":"Blog: "},{"type":"link","href":"https://openai.com/index/hello-gpt-4o/","raw":"https://openai.com/index/hello-gpt-4o/"},{"type":"text","value":" 🌐","raw":" 🌐"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Available today:https://chatgpt.com/ 🚀","raw":"Available today:https://chatgpt.com/ 🚀"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"I just wanted it to be cheaper, and more accessible! 🙌","raw":"I just wanted it to be cheaper, and more accessible! 🙌"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Still not open source, but a price reduction is welcome! 💰","raw":"Still not open source, but a price reduction is welcome! 💰"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Also, something fun happened, for the first 10-15 mins all search engines were correcting GPT-4o to GPT-4 😂","raw":"Also, something fun happened, for the first 10-15 mins all search engines were correcting GPT-4o to GPT-4 😂"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Also, also, GPT-4o is the model which was powering the GPT2 chatbot in the LMsys arena (ELO 1310 vs. 1253 for GPT-4 Turbo) 🏅","raw":"Also, also, GPT-4o is the model which was powering the GPT2 chatbot in the LMsys arena (ELO 1310 vs. 1253 for GPT-4 Turbo) 🏅"},{"type":"new_line","raw":"\n"}],"rawContent":"Is GPT-4o everything you expected? 🤔\n\n@OpenAI has gone omni (GPT-4\"o\" 🌐), a multimodal LLM, it accepts as input any combination of text, audio, and image and generates any combination of text, audio, and image outputs. 🎤📸✏️\n\n1️⃣ Based on the examples seen:\nInputs possible are Text ✏️, Text + Image 📝🖼️, Text + Audio 📝🎧, Text + Video 📝🎥, Audio 🎧\nand outputs possible are Image 🖼️, Image + Text 🖼️📝, Text 📝, Audio 🎧\n\n2️⃣ 88.7% on MMLU 🏆; 90.2% on HumanEval (best in class) 🥇\n\n3️⃣ Up to 50% cheaper 💸 and 2x faster ⚡ than GPT-4 Turbo\n\n4️⃣ GPT-4o will be available in the free tier of ChatGPT 🎉\n\n5️⃣ Near real-time audio with 320ms on average, similar to human conversation 🗣️**\n\n6️⃣ New tokenizer with a 200k token vocabulary 📚 (previously 100k vocabulary) leading to 1.1x - 4.4x fewer tokens needed across 20 languages 🌍\n\n7️⃣ Tokenizer compression and more efficient across non-English languages (3-5 times fewer tokens for major Indian languages 🇮🇳)\n\n👐Open questions:\n- What is the context length? ❓\n- Why does GPT-4 still exist, if GPT-4o is better, faster, and cheaper? 🤨\n\nBlog: https://openai.com/index/hello-gpt-4o/ 🌐\nAvailable today:https://chatgpt.com/ 🚀\n\nI just wanted it to be cheaper, and more accessible! 🙌\n\nStill not open source, but a price reduction is welcome! 💰\n\nAlso, something fun happened, for the first 10-15 mins all search engines were correcting GPT-4o to GPT-4 😂\n\nAlso, also, GPT-4o is the model which was powering the GPT2 chatbot in the LMsys arena (ELO 1310 vs. 1253 for GPT-4 Turbo) 🏅\n","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/662bf5bfe93bb73804ef9344/WXYLnjjJ4SROkoveIi7If.png","fullname":"Kuldeep Singh Sidhu","name":"singhsidhukuldeep","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/662bf5bfe93bb73804ef9344/5F48aMGYG_BDy_KojJodY.png"}],"mentions":[],"reactions":[{"reaction":"🔥","users":["julesbsz","vihangsharma","KingNish","Taylor658","victor","Felladrin"],"count":6},{"reaction":"👀","users":["clem"],"count":1}],"publishedAt":"2024-05-13T20:21:41.000Z","updatedAt":"2024-05-25T18:45:42.197Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6612aedf09f16e7347dfa7e1/bPYjBXCedY_1fSIPjoBTY.jpeg","fullname":"Nishith Jain","name":"KingNish","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/9fbbb8eaaa7b19752b336cf228d4679e.svg","fullname":"lucasjin","name":"lucasjin","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64d71ab4089bc502ceb44d29/nnacD7gbRSMbxCYBqkRYX.png","fullname":"Eugenio Schiavoni","name":"Kukedlc","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/63a75476f74d55cf00a0b12a26464bcc.svg","fullname":"adamsmith","name":"adamsmith887","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/6a1edd15ada4d2013fd0754b636c0c22.svg","fullname":"A","name":"Kilaril","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/singhsidhukuldeep/431728663194168","totalUniqueImpressions":3581,"numComments":7},"comment":{"_id":"6642a1c4ff3f8cfaf6966dd2","raw":"Interesting update! They can open-source GPT4 now haha","html":"Interesting update! They can open-source GPT4 now haha
\n"}},{"time":"2024-04-26T19:56:18.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"125379785468349","content":[{"type":"text","value":"Yay we got 500K+ monthly HF downloads on our Unsloth HF repo! :) Super appreciate everyone in the OSS community - and thanks for using Unsloth!!","raw":"Yay we got 500K+ monthly HF downloads on our Unsloth HF repo! :) Super appreciate everyone in the OSS community - and thanks for using Unsloth!!"}],"rawContent":"Yay we got 500K+ monthly HF downloads on our Unsloth HF repo! :) Super appreciate everyone in the OSS community - and thanks for using Unsloth!!","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62ecdc18b72a69615d6bd857/ixLCk0TwaCVyL_nAfrgEs.png","fullname":"Daniel Han-Chen","name":"danielhanchen","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/62ecdc18b72a69615d6bd857/Cwnm4lq7DtNq_LbdJUpF3.jpeg"}],"mentions":[],"reactions":[{"reaction":"🚀","users":["clem","giux78","lunarflu","nbroad","Ali-C137","Kukedlc","Oscar007","Dlbk","Chunte","catastropiyush","AtAndDev","julien-c","duxx","damerajee","feeltheAGI","beomi"],"count":16},{"reaction":"❤️","users":["clem","lunarflu","fredericmenezes","samusenps","nbroad","Ali-C137","Kukedlc","Chunte","AtAndDev","f0ster","julien-c","louisbrulenaudet","damerajee","aloobun","beomi"],"count":15},{"reaction":"🤗","users":["Kukedlc","Oscar007","AtAndDev","damerajee","Ji-Xiang","beomi"],"count":6}],"publishedAt":"2024-04-26T18:44:49.000Z","updatedAt":"2024-04-27T17:30:31.925Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1639773384591-5f353bb37e58354338621655.jpeg","fullname":"Nicholas Broad","name":"nbroad","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62ecdc18b72a69615d6bd857/ixLCk0TwaCVyL_nAfrgEs.png","fullname":"Daniel Han-Chen","name":"danielhanchen","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/danielhanchen/125379785468349","totalUniqueImpressions":3155,"numComments":4},"comment":{"_id":"662c06e20158a8f30d581451","raw":"congratulations! well deserved!","html":"congratulations! well deserved!
\n"}},{"time":"2024-04-24T22:59:01.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"890709532691598","content":[{"type":"text","value":"It's been only a week since I joined 🤗 and the community has released a constant flow of content! ","raw":"It's been only a week since I joined 🤗 and the community has released a constant flow of content! "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Notable models:","raw":"Notable models:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Apple OpenELM ","raw":"- Apple OpenELM "},{"type":"resource","resource":{"type":"collection","id":"apple/openelm-instruct-models-6619ad295d7ae9f868b759ca"},"url":"https://huggingface.co/collections/apple/openelm-instruct-models-6619ad295d7ae9f868b759ca","raw":"https://huggingface.co/collections/apple/openelm-instruct-models-6619ad295d7ae9f868b759ca"},{"type":"text","value":" + ","raw":" + "},{"type":"resource","resource":{"type":"collection","id":"apple/openelm-pretrained-models-6619ac6ca12a10bd0d0df89e"},"url":"https://huggingface.co/collections/apple/openelm-pretrained-models-6619ac6ca12a10bd0d0df89e","raw":"https://huggingface.co/collections/apple/openelm-pretrained-models-6619ac6ca12a10bd0d0df89e"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- HuggingFaceM4 Idefics2 ","raw":"- HuggingFaceM4 Idefics2 "},{"type":"resource","resource":{"type":"model","id":"HuggingFaceM4/idefics2-8b"},"url":"https://huggingface.co/HuggingFaceM4/idefics2-8b","raw":"https://huggingface.co/HuggingFaceM4/idefics2-8b"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Meta Llama 3 ","raw":"- Meta Llama 3 "},{"type":"resource","resource":{"type":"collection","id":"meta-llama/meta-llama-3-66214712577ca38149ebb2b6"},"url":"https://huggingface.co/collections/meta-llama/meta-llama-3-66214712577ca38149ebb2b6","raw":"https://huggingface.co/collections/meta-llama/meta-llama-3-66214712577ca38149ebb2b6"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Microsoft Phi-3 ","raw":"- Microsoft Phi-3 "},{"type":"resource","resource":{"type":"collection","id":"microsoft/phi-3-6626e15e9585a200d2d761e3"},"url":"https://huggingface.co/collections/microsoft/phi-3-6626e15e9585a200d2d761e3","raw":"https://huggingface.co/collections/microsoft/phi-3-6626e15e9585a200d2d761e3"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Snowflake Arctic ","raw":"- Snowflake Arctic "},{"type":"resource","resource":{"type":"collection","id":"Snowflake/arctic-66290090abe542894a5ac520"},"url":"https://huggingface.co/collections/Snowflake/arctic-66290090abe542894a5ac520","raw":"https://huggingface.co/collections/Snowflake/arctic-66290090abe542894a5ac520"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Great datasets:","raw":"Great datasets:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- HuggingFaceFW FineWeb ","raw":"- HuggingFaceFW FineWeb "},{"type":"resource","resource":{"type":"dataset","id":"HuggingFaceFW/fineweb"},"url":"https://huggingface.co/datasets/HuggingFaceFW/fineweb","raw":"https://huggingface.co/datasets/HuggingFaceFW/fineweb"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- HuggingFaceM4/the_cauldron ","raw":"- HuggingFaceM4/the_cauldron "},{"type":"resource","resource":{"type":"dataset","id":"HuggingFaceM4/the_cauldron"},"url":"https://huggingface.co/datasets/HuggingFaceM4/the_cauldron","raw":"https://huggingface.co/datasets/HuggingFaceM4/the_cauldron"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- PleIAs/YouTube-Commons ","raw":"- PleIAs/YouTube-Commons "},{"type":"resource","resource":{"type":"dataset","id":"PleIAs/YouTube-Commons"},"url":"https://huggingface.co/datasets/PleIAs/YouTube-Commons","raw":"https://huggingface.co/datasets/PleIAs/YouTube-Commons"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Fascinating Spaces","raw":"Fascinating Spaces"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- InstantMesh ","raw":"- InstantMesh "},{"type":"resource","resource":{"type":"space","id":"TencentARC/InstantMesh"},"url":"https://huggingface.co/spaces/TencentARC/InstantMesh","raw":"https://huggingface.co/spaces/TencentARC/InstantMesh"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Chat with Llama 3 8B ","raw":"- Chat with Llama 3 8B "},{"type":"resource","resource":{"type":"space","id":"ysharma/Chat_with_Meta_llama3_8b"},"url":"https://huggingface.co/spaces/ysharma/Chat_with_Meta_llama3_8b","raw":"https://huggingface.co/spaces/ysharma/Chat_with_Meta_llama3_8b"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Parler-TTS ","raw":"- Parler-TTS "},{"type":"link","href":"https://huggingface.co/spaces/parler-tts/parler_tts_mini","raw":"https://huggingface.co/spaces/parler-tts/parler_tts_mini"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- AI Jukebox ","raw":"- AI Jukebox "},{"type":"resource","resource":{"type":"space","id":"enzostvs/ai-jukebox"},"url":"https://huggingface.co/spaces/enzostvs/ai-jukebox","raw":"https://huggingface.co/spaces/enzostvs/ai-jukebox"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- CosXL ","raw":"- CosXL "},{"type":"resource","resource":{"type":"space","id":"multimodalart/cosxl"},"url":"https://huggingface.co/spaces/multimodalart/cosxl","raw":"https://huggingface.co/spaces/multimodalart/cosxl"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Singing songstarter ","raw":"- Singing songstarter "},{"type":"resource","resource":{"type":"space","id":"nateraw/singing-songstarter"},"url":"https://huggingface.co/spaces/nateraw/singing-songstarter","raw":"https://huggingface.co/spaces/nateraw/singing-songstarter"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Play with Idefics2 8B ","raw":"- Play with Idefics2 8B "},{"type":"resource","resource":{"type":"space","id":"HuggingFaceM4/idefics-8b"},"url":"https://huggingface.co/spaces/HuggingFaceM4/idefics-8b","raw":"https://huggingface.co/spaces/HuggingFaceM4/idefics-8b"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- CodeQwen1.5-7B-Chat Bot👾","raw":"- CodeQwen1.5-7B-Chat Bot👾"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"space","id":"Qwen/CodeQwen1.5-7b-Chat-demo"},"url":"https://huggingface.co/spaces/Qwen/CodeQwen1.5-7b-Chat-demo","raw":"https://huggingface.co/spaces/Qwen/CodeQwen1.5-7b-Chat-demo"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"I expected to be at the center of AI development. I'm not disappointed! ","raw":"I expected to be at the center of AI development. I'm not disappointed! "}],"rawContent":"It's been only a week since I joined 🤗 and the community has released a constant flow of content! \n\nNotable models:\n- Apple OpenELM https://huggingface.co/collections/apple/openelm-instruct-models-6619ad295d7ae9f868b759ca + https://huggingface.co/collections/apple/openelm-pretrained-models-6619ac6ca12a10bd0d0df89e\n- HuggingFaceM4 Idefics2 https://huggingface.co/HuggingFaceM4/idefics2-8b\n- Meta Llama 3 https://huggingface.co/collections/meta-llama/meta-llama-3-66214712577ca38149ebb2b6\n- Microsoft Phi-3 https://huggingface.co/collections/microsoft/phi-3-6626e15e9585a200d2d761e3\n- Snowflake Arctic https://huggingface.co/collections/Snowflake/arctic-66290090abe542894a5ac520\n\nGreat datasets:\n- HuggingFaceFW FineWeb https://huggingface.co/datasets/HuggingFaceFW/fineweb\n- HuggingFaceM4/the_cauldron https://huggingface.co/datasets/HuggingFaceM4/the_cauldron\n- PleIAs/YouTube-Commons https://huggingface.co/datasets/PleIAs/YouTube-Commons\n\nFascinating Spaces\n- InstantMesh https://huggingface.co/spaces/TencentARC/InstantMesh\n- Chat with Llama 3 8B https://huggingface.co/spaces/ysharma/Chat_with_Meta_llama3_8b\n- Parler-TTS https://huggingface.co/spaces/parler-tts/parler_tts_mini\n- AI Jukebox https://huggingface.co/spaces/enzostvs/ai-jukebox\n- CosXL https://huggingface.co/spaces/multimodalart/cosxl\n- Singing songstarter https://huggingface.co/spaces/nateraw/singing-songstarter\n- Play with Idefics2 8B https://huggingface.co/spaces/HuggingFaceM4/idefics-8b\n- CodeQwen1.5-7B-Chat Bot👾\nhttps://huggingface.co/spaces/Qwen/CodeQwen1.5-7b-Chat-demo\n\nI expected to be at the center of AI development. I'm not disappointed! ","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/647f36a8454af0237bd49574/jshkqBUTY-GZL8As8y6Aq.jpeg","fullname":"Florent Daudens","name":"fdaudens","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"🔥","users":["BrigitteTousi","clem","zealota","radames","daniel4999","tomaarsen","Dlbk","andrewrreed","vikas","danielus","Flashyflash","evijit","lunarflu","letheviet","4eyez","VictorSanh"],"count":16},{"reaction":"❤️","users":["BrigitteTousi","clem","zealota","radames","tomaarsen","lunarflu","letheviet","4eyez"],"count":8},{"reaction":"🚀","users":["BrigitteTousi","clem","zealota","lunarflu","letheviet","4eyez"],"count":6},{"reaction":"🤗","users":["BrigitteTousi","clem","zealota","Midgardsormr","lunarflu","4eyez"],"count":6}],"publishedAt":"2024-04-24T19:51:20.000Z","updatedAt":"2024-04-25T14:00:33.316Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64517300c5d273f9548172e7/p6dcLVprk51Jf2yNSwLpU.jpeg","fullname":"Icin","name":"swizzcheeze","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6317233cc92fd6fee317e030/cJHSvvimr1kqgQfHOjO5n.png","fullname":"Tom Aarsen","name":"tomaarsen","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/647f36a8454af0237bd49574/jshkqBUTY-GZL8As8y6Aq.jpeg","fullname":"Florent Daudens","name":"fdaudens","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/fdaudens/890709532691598","totalUniqueImpressions":2676,"numComments":4},"comment":{"_id":"66298eb59b9492f9cb2d25e8","raw":"gotta catch them all!","html":"gotta catch them all!
\n"}},{"time":"2024-04-24T22:57:18.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"471234622044561","content":[{"type":"text","value":"I'm super happy to co-organize the (Mechanistic) Interpretability social at #ICLR2024 with ","raw":"I'm super happy to co-organize the (Mechanistic) Interpretability social at #ICLR2024 with "},{"type":"mention","user":"nikhil07prakash","raw":"@nikhil07prakash"},{"type":"text","value":" ! 🔍","raw":" ! 🔍"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"If you plan to attend, help us make this meetup awesome by filling the form below! 😄 ","raw":"If you plan to attend, help us make this meetup awesome by filling the form below! 😄 "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"📅 Wed, May 8, 12:45-2:15 PM","raw":"📅 Wed, May 8, 12:45-2:15 PM"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔗 RSVP & share your ideas here: ","raw":"🔗 RSVP & share your ideas here: "},{"type":"link","href":"https://forms.gle/FWap4KW2ikdntjfb8","raw":"https://forms.gle/FWap4KW2ikdntjfb8"}],"rawContent":"I'm super happy to co-organize the (Mechanistic) Interpretability social at #ICLR2024 with @nikhil07prakash ! 🔍\n\nIf you plan to attend, help us make this meetup awesome by filling the form below! 😄 \n\n📅 Wed, May 8, 12:45-2:15 PM\n🔗 RSVP & share your ideas here: https://forms.gle/FWap4KW2ikdntjfb8","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670231290373-5e7749883d77a72421292d07.jpeg","fullname":"Gabriele Sarti","name":"gsarti","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"/avatars/1cc6c8ced4d455115bbc3a5126a4b8aa.svg","fullname":"Nikhil Prakash","name":"nikhil07prakash","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"🔥","users":["DmitryRyumin","DeepMount00","ajibawa-2023","emaballarin"],"count":4},{"reaction":"👀","users":["Tonic"],"count":1}],"publishedAt":"2024-04-23T06:34:23.000Z","updatedAt":"2024-04-25T14:46:35.591Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64aea8ff67511bd3d965697b/Jxn52EmDF5RApJh8antxn.jpeg","fullname":"Feynman Innovations","name":"ajibawa-2023","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670231290373-5e7749883d77a72421292d07.jpeg","fullname":"Gabriele Sarti","name":"gsarti","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/gsarti/471234622044561","totalUniqueImpressions":1767,"numComments":5},"comment":{"_id":"66298e4e21e410e67c29b167","raw":"you should create an org on HF for it","html":"you should create an org on HF for it
\n"}},{"time":"2024-04-22T22:17:36.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"736759191735161","content":[{"type":"text","value":"Great in-depth Llama-3 tests from ","raw":"Great in-depth Llama-3 tests from "},{"type":"mention","user":"wolfram","raw":"@wolfram"},{"type":"text","value":", of the models from Meta of course but also ","raw":", of the models from Meta of course but also "},{"type":"mention","user":"MaziyarPanahi","raw":"@MaziyarPanahi"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"emozilla","raw":"@emozilla"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"turboderp","raw":"@turboderp"},{"type":"text","value":" : ","raw":" : "},{"type":"link","href":"https://huggingface.co/blog/wolfram/llm-comparison-test-llama-3","raw":"https://huggingface.co/blog/wolfram/llm-comparison-test-llama-3"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Spotted by ","raw":"Spotted by "},{"type":"mention","user":"jack-kumar","raw":"@jack-kumar"}],"rawContent":"Great in-depth Llama-3 tests from @wolfram, of the models from Meta of course but also @MaziyarPanahi @emozilla @turboderp : https://huggingface.co/blog/wolfram/llm-comparison-test-llama-3\n\nSpotted by @jack-kumar","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1676652577978-630581db99870e13d3e0006f.jpeg","fullname":"Jeffrey Quesnelle","name":"emozilla","type":"user","isPro":true,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/66214e4185446629f393638f/25wbY0H9u1KekuFwiK3OI.png","fullname":"Jack Kumar","name":"jack-kumar","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5fd5e18a90b6dc4633f6d292/gZXHW5dd9R86AV9LMZ--y.png","fullname":"Maziyar Panahi","name":"MaziyarPanahi","type":"user","isPro":true,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6383dc174c48969dcf1b4fce/4N-GY7jVvdk08kp2B8DLh.jpeg","fullname":"turboderp","name":"turboderp","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6303ca537373aacccd85d8a7/JZqLjXZVGWXJdWUNI99db.jpeg","fullname":"Wolfram Ravenwolf","name":"wolfram","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"🤗","users":["wolfram","fdaudens","raincandy-u","MaziyarPanahi","julien-c","SaylorTwift"],"count":6}],"publishedAt":"2024-04-22T22:17:36.000Z","updatedAt":"2024-05-08T17:39:28.862Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5fd5e18a90b6dc4633f6d292/gZXHW5dd9R86AV9LMZ--y.png","fullname":"Maziyar Panahi","name":"MaziyarPanahi","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64b1619f4c3cc95a751e6c41/oSUtmp1Gw0I-ve_1wFlNW.jpeg","fullname":"Michal Zebrowski (Enki)","name":"M1cler","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/736759191735161","totalUniqueImpressions":2516,"numComments":2}},{"time":"2024-04-22T14:27:58.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"189843726787210","content":[{"type":"text","value":"Already almost 1,000 llama3 model variations have been shared publicly on HF (many more in private use at companies): ","raw":"Already almost 1,000 llama3 model variations have been shared publicly on HF (many more in private use at companies): "},{"type":"link","href":"https://huggingface.co/models?p=5&sort=trending&search=llama3","raw":"https://huggingface.co/models?p=5&sort=trending&search=llama3"},{"type":"text","value":". ","raw":". "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Everyone should fine-tune their own models for their use-cases, languages, industry, infra constraints,... ","raw":"Everyone should fine-tune their own models for their use-cases, languages, industry, infra constraints,... "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"10,000 llama3 variants by the end of next week?","raw":"10,000 llama3 variants by the end of next week?"}],"rawContent":"Already almost 1,000 llama3 model variations have been shared publicly on HF (many more in private use at companies): https://huggingface.co/models?p=5&sort=trending&search=llama3. \n\nEveryone should fine-tune their own models for their use-cases, languages, industry, infra constraints,... \n\n10,000 llama3 variants by the end of next week?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👍","users":["ijohn07","qnixsynapse","Aeio","DmitryRyumin","Kathakali","tomaarsen","qubvel-hf","rksiitd","victor","Jaward","radames","shiv2050","praveenpankaj","Tonic","rasonyang"],"count":15},{"reaction":"❤️","users":["yo","DmitryRyumin","Ali-C137","sepal","tomaarsen","fbjr","raincandy-u","radames","Tonic","louisbrulenaudet"],"count":10},{"reaction":"🤯","users":["DmitryRyumin","abidlabs","Tonic"],"count":3}],"publishedAt":"2024-04-21T14:38:03.000Z","updatedAt":"2024-04-25T05:40:05.174Z","commentators":[{"avatarUrl":"/avatars/0cba7c22c135f9c8fbe19398ba408923.svg","fullname":"Fred Bliss","name":"fbjr","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/657eb5b256c9c67605a6e8b5/RPblnGJX57oiIcASEz_S8.png","fullname":"raincandy_U","name":"raincandy-u","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1675590698085-noauth.jpeg","fullname":"Sai","name":"saishf","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/189843726787210","totalUniqueImpressions":2895,"numComments":4},"comment":{"_id":"662673ee2c0d5257395c8709","raw":"let's fix it: https://twitter.com/ClementDelangue/status/1782065141200073122","html":"let's fix it: https://twitter.com/ClementDelangue/status/1782065141200073122
\n"}},{"time":"2024-04-21T14:38:03.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"189843726787210","content":[{"type":"text","value":"Already almost 1,000 llama3 model variations have been shared publicly on HF (many more in private use at companies): ","raw":"Already almost 1,000 llama3 model variations have been shared publicly on HF (many more in private use at companies): "},{"type":"link","href":"https://huggingface.co/models?p=5&sort=trending&search=llama3","raw":"https://huggingface.co/models?p=5&sort=trending&search=llama3"},{"type":"text","value":". ","raw":". "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Everyone should fine-tune their own models for their use-cases, languages, industry, infra constraints,... ","raw":"Everyone should fine-tune their own models for their use-cases, languages, industry, infra constraints,... "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"10,000 llama3 variants by the end of next week?","raw":"10,000 llama3 variants by the end of next week?"}],"rawContent":"Already almost 1,000 llama3 model variations have been shared publicly on HF (many more in private use at companies): https://huggingface.co/models?p=5&sort=trending&search=llama3. \n\nEveryone should fine-tune their own models for their use-cases, languages, industry, infra constraints,... \n\n10,000 llama3 variants by the end of next week?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👍","users":["ijohn07","qnixsynapse","Aeio","DmitryRyumin","Kathakali","tomaarsen","qubvel-hf","rksiitd","victor","Jaward","radames","shiv2050","praveenpankaj","Tonic","rasonyang"],"count":15},{"reaction":"❤️","users":["yo","DmitryRyumin","Ali-C137","sepal","tomaarsen","fbjr","raincandy-u","radames","Tonic","louisbrulenaudet"],"count":10},{"reaction":"🤯","users":["DmitryRyumin","abidlabs","Tonic"],"count":3}],"publishedAt":"2024-04-21T14:38:03.000Z","updatedAt":"2024-04-25T05:40:05.174Z","commentators":[{"avatarUrl":"/avatars/0cba7c22c135f9c8fbe19398ba408923.svg","fullname":"Fred Bliss","name":"fbjr","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/657eb5b256c9c67605a6e8b5/RPblnGJX57oiIcASEz_S8.png","fullname":"raincandy_U","name":"raincandy-u","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1675590698085-noauth.jpeg","fullname":"Sai","name":"saishf","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/189843726787210","totalUniqueImpressions":2895,"numComments":4}},{"time":"2024-04-17T01:08:41.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"722159008619961","content":[{"type":"text","value":"Yesterday, xAI announced Grok-1.5 Vision - ","raw":"Yesterday, xAI announced Grok-1.5 Vision - "},{"type":"link","href":"https://x.ai/blog/grok-1.5v","raw":"https://x.ai/blog/grok-1.5v"},{"type":"text","value":". But more importantly, they also released a new VLM benchmark dataset - RealWorldQA. The only problem was that they released it as a ZIP archive. I fixed that! Now you can use it in your evaluations as a regular HF dataset: ","raw":". But more importantly, they also released a new VLM benchmark dataset - RealWorldQA. The only problem was that they released it as a ZIP archive. I fixed that! Now you can use it in your evaluations as a regular HF dataset: "},{"type":"resource","resource":{"type":"dataset","id":"visheratin/realworldqa"},"url":"https://huggingface.co/datasets/visheratin/realworldqa","raw":"https://huggingface.co/datasets/visheratin/realworldqa"}],"rawContent":"Yesterday, xAI announced Grok-1.5 Vision - https://x.ai/blog/grok-1.5v. But more importantly, they also released a new VLM benchmark dataset - RealWorldQA. The only problem was that they released it as a ZIP archive. I fixed that! Now you can use it in your evaluations as a regular HF dataset: https://huggingface.co/datasets/visheratin/realworldqa","author":{"avatarUrl":"/avatars/b892a3d50b2f8ead0a5f7108564e45d0.svg","fullname":"Alexander Visheratin","name":"visheratin","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👍","users":["musfiqdehan","fynnkroeger","osanseviero","Bruno","ajibawa-2023","VictorSanh","clem","not-lain","ShihHong","Ali-C137","dinhanhx"],"count":11}],"publishedAt":"2024-04-13T13:15:12.000Z","updatedAt":"2024-04-17T01:08:41.477Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/visheratin/722159008619961","totalUniqueImpressions":3047,"numComments":1},"comment":{"_id":"661f211942c9bf3876e33cfd","raw":"Thank you! You should tweet it mentioning @elonmuskceo!","html":"Thank you! You should tweet it mentioning \n\n@elonmuskceo\n\n\t!
\n"}},{"time":"2024-04-16T21:44:18.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"583952430077847","content":[{"type":"text","value":"We noticed that all the open-source models and datasets from ","raw":"We noticed that all the open-source models and datasets from "},{"type":"link","href":"https://huggingface.co/WizardLM","raw":"https://huggingface.co/WizardLM"},{"type":"text","value":" in their personal Hugging Face account & in the Microsoft Hugging Face organization (","raw":" in their personal Hugging Face account & in the Microsoft Hugging Face organization ("},{"type":"link","href":"https://huggingface.co/microsoft","raw":"https://huggingface.co/microsoft"},{"type":"text","value":") have been made private by the author, which will lead some demos to fail (these models were collectively downloaded over a hundred thousand times a month). ","raw":") have been made private by the author, which will lead some demos to fail (these models were collectively downloaded over a hundred thousand times a month). "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"This is the explanation that ","raw":"This is the explanation that "},{"type":"mention","user":"WizardLM","raw":"@WizardLM"},{"type":"text","value":" communicated a few hours ago: ","raw":" communicated a few hours ago: "},{"type":"link","href":"https://huggingface.co/posts/WizardLM/329547800484476#661e0d17bca1a6038b60503e","raw":"https://huggingface.co/posts/WizardLM/329547800484476#661e0d17bca1a6038b60503e"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"We apologize for the inconvenience & are trying to get in touch with the author & Microsoft in order to try to find a good resolution for community members. Let us know if you have any questions!","raw":"We apologize for the inconvenience & are trying to get in touch with the author & Microsoft in order to try to find a good resolution for community members. Let us know if you have any questions!"}],"rawContent":"We noticed that all the open-source models and datasets from https://huggingface.co/WizardLM in their personal Hugging Face account & in the Microsoft Hugging Face organization (https://huggingface.co/microsoft) have been made private by the author, which will lead some demos to fail (these models were collectively downloaded over a hundred thousand times a month). \n\nThis is the explanation that @WizardLM communicated a few hours ago: https://huggingface.co/posts/WizardLM/329547800484476#661e0d17bca1a6038b60503e\n\nWe apologize for the inconvenience & are trying to get in touch with the author & Microsoft in order to try to find a good resolution for community members. Let us know if you have any questions!","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/627b9f3f4d0858f0034efbb9/2Qnattrzv6qvqiZVVfV5x.png","fullname":"WizardLM","name":"WizardLM","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"👀","users":["osanseviero","fdaudens","ijohn07","pierrci","BrigitteTousi","lazarustda","samusenps","clefourrier","lunarflu","Diz187","guilherme-am","m-ric","acamilogg88"],"count":13},{"reaction":"😔","users":["osanseviero","pierrci","Mehrdadgpt","lunarflu","fynnkroeger","m-ric"],"count":6}],"publishedAt":"2024-04-16T21:44:18.000Z","updatedAt":"2024-04-17T22:41:48.267Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/654db98e1b922725af7ec5ec/gChOe5aQjFnQxnksbD2AY.jpeg","fullname":"Diz","name":"Diz187","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/583952430077847","totalUniqueImpressions":2671,"numComments":1}},{"time":"2024-04-15T07:15:36.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"132951187064518","content":[{"type":"text","value":"Fun dataset added last week by ","raw":"Fun dataset added last week by "},{"type":"mention","user":"esind","raw":"@esind"},{"type":"text","value":" from ","raw":" from "},{"type":"link","href":"https://huggingface.co/Anthropic","raw":"https://huggingface.co/Anthropic"},{"type":"text","value":" to compare persuasiveness between AI and human outputs: ","raw":" to compare persuasiveness between AI and human outputs: "},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"dataset","id":"Anthropic/persuasion"},"url":"https://huggingface.co/datasets/Anthropic/persuasion","raw":"https://huggingface.co/datasets/Anthropic/persuasion"}],"rawContent":"Fun dataset added last week by @esind from https://huggingface.co/Anthropic to compare persuasiveness between AI and human outputs: \nhttps://huggingface.co/datasets/Anthropic/persuasion","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/77UIqMwhFodTRi35sy5EP.png"}],"mentions":[{"avatarUrl":"/avatars/c515e502a45156b28d845af6033ad38c.svg","fullname":"Esin Durmus","name":"esind","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"🚀","users":["Jaward","kgourgou","osanseviero","not-lain","KvrParaskevi","InferenceIllusionist","nogully"],"count":7},{"reaction":"👍","users":["pierrci","not-lain"],"count":2}],"publishedAt":"2024-04-15T07:15:36.000Z","updatedAt":"2024-04-15T10:50:25.232Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6438a9027de34e8ea7e4b257/vib8QSd1AWMr_bR9ig_xJ.jpeg","fullname":"Jaward Sesay","name":"Jaward","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/clem/132951187064518","totalUniqueImpressions":2427,"numComments":2}},{"time":"2024-04-04T17:35:37.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"731102107007171","content":[{"type":"text","value":"Introducing ","raw":"Introducing "},{"type":"resource","resource":{"type":"dataset","id":"gretelai/synthetic_text_to_sql"},"url":"https://huggingface.co/datasets/gretelai/synthetic_text_to_sql","raw":"https://huggingface.co/datasets/gretelai/synthetic_text_to_sql"},{"type":"text","value":" by ","raw":" by "},{"type":"link","href":"https://huggingface.co/gretelai","raw":"https://huggingface.co/gretelai"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"It stands as the largest and most diverse synthetic Text-to-SQL dataset available to-date.","raw":"It stands as the largest and most diverse synthetic Text-to-SQL dataset available to-date."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"The dataset includes:","raw":"The dataset includes:"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- 105,851 records partitioned into 100,000 train and 5,851 test records","raw":"- 105,851 records partitioned into 100,000 train and 5,851 test records"},{"type":"new_line","raw":"\n"},{"type":"text","value":"~23M total tokens, including ~12M SQL tokens","raw":"~23M total tokens, including ~12M SQL tokens"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Coverage across 100 distinct domains/verticals","raw":"- Coverage across 100 distinct domains/verticals"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Comprehensive array of SQL tasks: data definition, retrieval, manipulation, analytics & reporting","raw":"- Comprehensive array of SQL tasks: data definition, retrieval, manipulation, analytics & reporting"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Wide range of SQL complexity levels, including subqueries, single joins, multiple joins, aggregations, window functions, set operations","raw":"- Wide range of SQL complexity levels, including subqueries, single joins, multiple joins, aggregations, window functions, set operations"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Database context, including table and view create statements","raw":"- Database context, including table and view create statements"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Natural language explanations of what the SQL query is doing","raw":"- Natural language explanations of what the SQL query is doing"},{"type":"new_line","raw":"\n"},{"type":"text","value":"- Contextual tags to optimize model training","raw":"- Contextual tags to optimize model training"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Blogpost: ","raw":"Blogpost: "},{"type":"link","href":"https://gretel.ai/blog/synthetic-text-to-sql-dataset","raw":"https://gretel.ai/blog/synthetic-text-to-sql-dataset"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Dataset: ","raw":"Dataset: "},{"type":"resource","resource":{"type":"dataset","id":"gretelai/synthetic_text_to_sql"},"url":"https://huggingface.co/datasets/gretelai/synthetic_text_to_sql","raw":"https://huggingface.co/datasets/gretelai/synthetic_text_to_sql"},{"type":"text","value":" ","raw":" "}],"rawContent":"Introducing https://huggingface.co/datasets/gretelai/synthetic_text_to_sql by https://huggingface.co/gretelai\n\nIt stands as the largest and most diverse synthetic Text-to-SQL dataset available to-date.\n\nThe dataset includes:\n\n- 105,851 records partitioned into 100,000 train and 5,851 test records\n~23M total tokens, including ~12M SQL tokens\n- Coverage across 100 distinct domains/verticals\n- Comprehensive array of SQL tasks: data definition, retrieval, manipulation, analytics & reporting\n- Wide range of SQL complexity levels, including subqueries, single joins, multiple joins, aggregations, window functions, set operations\n- Database context, including table and view create statements\n- Natural language explanations of what the SQL query is doing\n- Contextual tags to optimize model training\n\nBlogpost: https://gretel.ai/blog/synthetic-text-to-sql-dataset\nDataset: https://huggingface.co/datasets/gretelai/synthetic_text_to_sql ","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/LReH7TiCXyLumal3deBBn.webp"}],"mentions":[],"reactions":[{"reaction":"🤗","users":["ichigoberry","mvaloatto","Svngoku","nikgr","thomwolf","lhoestq","brhnsbn"],"count":7},{"reaction":"❤️","users":["samusenps","thomwolf","lhoestq"],"count":3},{"reaction":"🔥","users":["Jensen-holm"],"count":1}],"publishedAt":"2024-04-04T17:35:37.000Z","updatedAt":"2024-04-04T18:54:59.472Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6389d3c61e8755d777902366/xR50wHQW4ZsFxI8lolk7a.jpeg","fullname":"Thomas Schranz","name":"ichigoberry","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/731102107007171","totalUniqueImpressions":2518,"numComments":1}},{"time":"2024-04-03T18:40:07.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"299776234518647","content":[{"type":"text","value":"👋 👋 👋 In a recent AI Time Journal Interview, I was asked: How has HackerNoon leveraged AI to enhance its platform? and what lessons have you learned about AI’s role in spam creation and prevention? ","raw":"👋 👋 👋 In a recent AI Time Journal Interview, I was asked: How has HackerNoon leveraged AI to enhance its platform? and what lessons have you learned about AI’s role in spam creation and prevention? "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"link","href":"https://HackerNoon.com","raw":"https://HackerNoon.com"},{"type":"text","value":" uses AI and machine learning to improve grammar, translate stories, generate featured images, suggest possible headlines, (and note, I would love to hear what else you think we should use AI for!) . Maybe Clippy was ahead of its time? Writers have and will continue to have smart AI assistants. AI can not and should not replace the writer, but it should and can assist the writer and editor throughout the publishing process. We are documenting our approach to editorial at ","raw":" uses AI and machine learning to improve grammar, translate stories, generate featured images, suggest possible headlines, (and note, I would love to hear what else you think we should use AI for!) . Maybe Clippy was ahead of its time? Writers have and will continue to have smart AI assistants. AI can not and should not replace the writer, but it should and can assist the writer and editor throughout the publishing process. We are documenting our approach to editorial at "},{"type":"link","href":"https://EditingProtocol.com","raw":"https://EditingProtocol.com"},{"type":"text","value":". With the rise of LLMs, we’ve seen an influx of AI story spam submissions. Other blogging platforms are hosting so many of these types of machine-generated stories right now. We are actively taking steps not to become a wasteland of AI’s musings. Because every story is reviewed before publication (with about half being rejected), I think our average story quality is higher than other blogging platforms. Every story submission automatically enters into plagiarism and AI writing detection. We highlight what sections are likely, possibly, and unlikely to be written by AI. It’s all about confidence levels. Ultimately, human editors make the final call in interpreting the machines’ reports.","raw":". With the rise of LLMs, we’ve seen an influx of AI story spam submissions. Other blogging platforms are hosting so many of these types of machine-generated stories right now. We are actively taking steps not to become a wasteland of AI’s musings. Because every story is reviewed before publication (with about half being rejected), I think our average story quality is higher than other blogging platforms. Every story submission automatically enters into plagiarism and AI writing detection. We highlight what sections are likely, possibly, and unlikely to be written by AI. It’s all about confidence levels. Ultimately, human editors make the final call in interpreting the machines’ reports."}],"rawContent":"👋 👋 👋 In a recent AI Time Journal Interview, I was asked: How has HackerNoon leveraged AI to enhance its platform? and what lessons have you learned about AI’s role in spam creation and prevention? \n\nhttps://HackerNoon.com uses AI and machine learning to improve grammar, translate stories, generate featured images, suggest possible headlines, (and note, I would love to hear what else you think we should use AI for!) . Maybe Clippy was ahead of its time? Writers have and will continue to have smart AI assistants. AI can not and should not replace the writer, but it should and can assist the writer and editor throughout the publishing process. We are documenting our approach to editorial at https://EditingProtocol.com. With the rise of LLMs, we’ve seen an influx of AI story spam submissions. Other blogging platforms are hosting so many of these types of machine-generated stories right now. We are actively taking steps not to become a wasteland of AI’s musings. Because every story is reviewed before publication (with about half being rejected), I think our average story quality is higher than other blogging platforms. Every story submission automatically enters into plagiarism and AI writing detection. We highlight what sections are likely, possibly, and unlikely to be written by AI. It’s all about confidence levels. Ultimately, human editors make the final call in interpreting the machines’ reports.","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64862a25cf5ad5e1f0482ef2/61qPUtw9jIl7zpPYmi0VW.jpeg","fullname":"David Smooke","name":"Smooke","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"🔥","users":["julien-c","clem","osanseviero","merve"],"count":4},{"reaction":"❤️","users":["samusenps","kramp","osanseviero","merve"],"count":4},{"reaction":"➕","users":["samusenps","clem","merve"],"count":3}],"publishedAt":"2024-04-03T16:08:05.000Z","updatedAt":"2024-04-03T18:54:11.800Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64862a25cf5ad5e1f0482ef2/61qPUtw9jIl7zpPYmi0VW.jpeg","fullname":"David Smooke","name":"Smooke","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6538119803519fddb4a17e10/ffJMkdx-rM7VvLTCM6ri_.jpeg","fullname":"samusenps","name":"samusenps","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/Smooke/299776234518647","totalUniqueImpressions":1930,"numComments":4},"comment":{"_id":"660da2871813bad9008f0484","raw":"Thanks for sharing!","html":"Thanks for sharing!
\n"}},{"time":"2024-03-28T16:48:32.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"236311186560364","content":[{"type":"text","value":"Very glad to welcome ","raw":"Very glad to welcome "},{"type":"mention","user":"josefprusa","raw":"@josefprusa"},{"type":"text","value":", pioneer of 3D printing and open source hardware, founder of ","raw":", pioneer of 3D printing and open source hardware, founder of "},{"type":"link","href":"https://www.prusa3d.com/","raw":"https://www.prusa3d.com/"},{"type":"text","value":", to the HF Hub 👋","raw":", to the HF Hub 👋"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"AI applied to 3D printing could be big.","raw":"AI applied to 3D printing could be big."}],"rawContent":"Very glad to welcome @josefprusa, pioneer of 3D printing and open source hardware, founder of https://www.prusa3d.com/, to the HF Hub 👋\n\nAI applied to 3D printing could be big.","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/NQtzmrDdbG0H8qkZvRyGk.jpeg","fullname":"Julien Chaumond","name":"julien-c","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5dd96eb166059660ed1ee413/Buyqz5oR6wuj5-C9iFUkE.png"}],"mentions":[{"avatarUrl":"/avatars/533162c0048b18e983a1b220b82230e1.svg","fullname":"Josef Průša","name":"josefprusa","type":"user","isPro":true,"isHf":false,"isMod":false}],"reactions":[{"reaction":"🔥","users":["victor","London12345","satpalsr","diwank","ZennyKenny","radames","adamelliotfields","VictorSanh","mmhamdy","samusenps","pcuenq","kramp","osanseviero","stattmone","mdouglas","clem","Skier8402","jarvisx17"],"count":18},{"reaction":"👍","users":["Aurelien-Morgan","imomayiz","mmhamdy","samusenps","clem"],"count":5},{"reaction":"❤️","users":["samusenps","osanseviero","clem"],"count":3},{"reaction":"👀","users":["samusenps","clem"],"count":2},{"reaction":"➕","users":["samusenps","clem"],"count":2}],"publishedAt":"2024-03-26T16:25:55.000Z","updatedAt":"2024-03-28T16:48:32.476Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/julien-c/236311186560364","totalUniqueImpressions":3154,"numComments":1},"comment":{"_id":"66059f60b3d0e949eed8b17c","raw":"Welcome @josefprusa!","html":"Welcome \n\n@josefprusa\n\n\t!
\n"}},{"time":"2024-02-26T22:04:16.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"132841805412297","content":[{"type":"text","value":"Terribly excited about open-source + on-device AI these days! Great to see ","raw":"Terribly excited about open-source + on-device AI these days! Great to see "},{"type":"mention","user":"qualcomm","raw":"@qualcomm"},{"type":"text","value":" release 80+ models optimized and curated for their devices and chips on HF: ","raw":" release 80+ models optimized and curated for their devices and chips on HF: "},{"type":"link","href":"https://huggingface.co/qualcomm","raw":"https://huggingface.co/qualcomm"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"}],"rawContent":"Terribly excited about open-source + on-device AI these days! Great to see @qualcomm release 80+ models optimized and curated for their devices and chips on HF: https://huggingface.co/qualcomm\n\n","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/6-FF9UzxJjRv6LEZPOkYs.png"}],"mentions":[],"reactions":[{"reaction":"❤️","users":["DmitryRyumin","vladbogo","harpreetsahota","adamelliotfields","banghua","mvaloatto","lunarflu","osanseviero","marceloxp","ajibawa-2023","terpjwu1","danielhanchen","yjernite","samusenps","realtamimuddin"],"count":15},{"reaction":"🤯","users":["lunarflu","osanseviero","danielus","ehristoforu","siyengfeng","yjernite","realtamimuddin"],"count":7}],"publishedAt":"2024-02-26T22:04:16.000Z","updatedAt":"2024-02-28T03:55:41.302Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64aea8ff67511bd3d965697b/Jxn52EmDF5RApJh8antxn.jpeg","fullname":"Feynman Innovations","name":"ajibawa-2023","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/132841805412297","totalUniqueImpressions":491,"numComments":1}},{"time":"2024-02-26T22:03:18.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"297458605697772","content":[{"type":"text","value":"🚀🧙🏼♂️Introducing OpenHermesPreferences: the largest open AI feedback dataset for RLHF & DPO","raw":"🚀🧙🏼♂️Introducing OpenHermesPreferences: the largest open AI feedback dataset for RLHF & DPO"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"> Using LLMs to improve other LLMs, at scale!","raw":"> Using LLMs to improve other LLMs, at scale!"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Built in collaboration with the H4 Hugging Face team, it's a 1M preferences dataset on top of the amazing ","raw":"Built in collaboration with the H4 Hugging Face team, it's a 1M preferences dataset on top of the amazing "},{"type":"mention","user":"teknium","raw":"@teknium"},{"type":"text","value":" 's dataset.","raw":" 's dataset."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Dataset:","raw":"Dataset:"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"dataset","id":"argilla/OpenHermesPreferences"},"url":"https://huggingface.co/datasets/argilla/OpenHermesPreferences","raw":"https://huggingface.co/datasets/argilla/OpenHermesPreferences"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"The dataset is another example of open collaboration:","raw":"The dataset is another example of open collaboration:"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"> The H4 team created responses with Mixtral using llm-swarm","raw":"> The H4 team created responses with Mixtral using llm-swarm"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"> Argilla created responses with NousResearch Hermes-2-Yi-34B using distilabel","raw":"> Argilla created responses with NousResearch Hermes-2-Yi-34B using distilabel"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"> The H4 ranked these responses + original response with PairRM from AllenAI, University of Southern California, Zhejiang University (","raw":"> The H4 ranked these responses + original response with PairRM from AllenAI, University of Southern California, Zhejiang University ("},{"type":"mention","user":"yuchenlin","raw":"@yuchenlin"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"DongfuTingle","raw":"@DongfuTingle"},{"type":"text","value":" and colleagues)","raw":" and colleagues)"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"We hope this dataset will help the community's research efforts towards understanding the role of AI feedback for LLM alignment.","raw":"We hope this dataset will help the community's research efforts towards understanding the role of AI feedback for LLM alignment."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"We're particularly excited about the ability of filtering specific subsets to improve LLM skills like math or reasoning.","raw":"We're particularly excited about the ability of filtering specific subsets to improve LLM skills like math or reasoning."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Here's how easy it is to filter by subset:","raw":"Here's how easy it is to filter by subset:"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"code_fence","code":"ds = load_dataset(\"HuggingFaceH4/OpenHermesPreferences\", split=\"train\")\n\n# Get the categories of the source dataset\n# ['airoboros2.2', 'CamelAI', 'caseus_custom', ...]\nsources = ds.unique(\"source\")\n\n# Filter for a subset\nds_filtered = ds.filter(lambda x : x[\"source\"] in [\"metamath\", \"EvolInstruct_70k\"], num_proc=6)","raw":"```\nds = load_dataset(\"HuggingFaceH4/OpenHermesPreferences\", split=\"train\")\n\n# Get the categories of the source dataset\n# ['airoboros2.2', 'CamelAI', 'caseus_custom', ...]\nsources = ds.unique(\"source\")\n\n# Filter for a subset\nds_filtered = ds.filter(lambda x : x[\"source\"] in [\"metamath\", \"EvolInstruct_70k\"], num_proc=6)\n```"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"As usual, all the scripts to reproduce this work are available and open to the community!","raw":"As usual, all the scripts to reproduce this work are available and open to the community!"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"dataset","id":"argilla/OpenHermesPreferences"},"url":"https://huggingface.co/datasets/argilla/OpenHermesPreferences/tree/main","raw":"https://huggingface.co/datasets/argilla/OpenHermesPreferences/tree/main"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"So fun collab between ","raw":"So fun collab between "},{"type":"mention","user":"vwxyzjn","raw":"@vwxyzjn"},{"type":"text","value":" , ","raw":" , "},{"type":"mention","user":"plaguss","raw":"@plaguss"},{"type":"text","value":", ","raw":", "},{"type":"mention","user":"kashif","raw":"@kashif"},{"type":"text","value":", ","raw":", "},{"type":"mention","user":"philschmid","raw":"@philschmid"},{"type":"text","value":" & ","raw":" & "},{"type":"mention","user":"lewtun","raw":"@lewtun"},{"type":"text","value":"!","raw":"!"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Open Source AI FTW! ","raw":"Open Source AI FTW! "},{"type":"new_line","raw":"\n"}],"rawContent":"🚀🧙🏼♂️Introducing OpenHermesPreferences: the largest open AI feedback dataset for RLHF & DPO\n\n> Using LLMs to improve other LLMs, at scale!\n\nBuilt in collaboration with the H4 Hugging Face team, it's a 1M preferences dataset on top of the amazing @teknium 's dataset.\n\nDataset:\nhttps://huggingface.co/datasets/argilla/OpenHermesPreferences\n\nThe dataset is another example of open collaboration:\n\n> The H4 team created responses with Mixtral using llm-swarm\n\n> Argilla created responses with NousResearch Hermes-2-Yi-34B using distilabel\n\n> The H4 ranked these responses + original response with PairRM from AllenAI, University of Southern California, Zhejiang University (@yuchenlin @DongfuTingle and colleagues)\n\nWe hope this dataset will help the community's research efforts towards understanding the role of AI feedback for LLM alignment.\n\nWe're particularly excited about the ability of filtering specific subsets to improve LLM skills like math or reasoning.\n\nHere's how easy it is to filter by subset:\n\n```\nds = load_dataset(\"HuggingFaceH4/OpenHermesPreferences\", split=\"train\")\n\n# Get the categories of the source dataset\n# ['airoboros2.2', 'CamelAI', 'caseus_custom', ...]\nsources = ds.unique(\"source\")\n\n# Filter for a subset\nds_filtered = ds.filter(lambda x : x[\"source\"] in [\"metamath\", \"EvolInstruct_70k\"], num_proc=6)\n```\n\nAs usual, all the scripts to reproduce this work are available and open to the community!\n\nhttps://huggingface.co/datasets/argilla/OpenHermesPreferences/tree/main\n\nSo fun collab between @vwxyzjn , @plaguss, @kashif, @philschmid & @lewtun!\n\nOpen Source AI FTW! \n","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60420dccc15e823a685f2b03/Dn7QTyy9SZ7jKN6xpufVD.png","fullname":"Daniel Vila","name":"dvilasuero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/60420dccc15e823a685f2b03/ZsuNNR_QglUktaU5vjooi.png"}],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62567c86d444a9b5a0ec51c1/SUgRd1glmXO9Z7x_h0yMn.jpeg","fullname":"Dongfu Jiang","name":"DongfuJiang","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1669189789447-629f3b18ee05727ce328ccbe.jpeg","fullname":"Kashif Rasul","name":"kashif","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1594651707950-noauth.jpeg","fullname":"Lewis Tunstall","name":"lewtun","type":"user","isPro":true,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1624629516652-5ff5d596f244529b3ec0fb89.png","fullname":"Philipp Schmid","name":"philschmid","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6435d564a4bd75c62cc03701/7P2G_wVNB6MISp2Phh427.jpeg","fullname":"Agustín Piqueres Lajarín","name":"plaguss","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6317aade83d8d2fd903192d9/erOwgMXc_CZih3uMoyTAp.jpeg","fullname":"Teknium","name":"teknium","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62bfc80b54db40ca617ce53a/IZxafMqoYJNHwTuFt8AqF.png","fullname":"Shengyi Costa Huang","name":"vwxyzjn","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/607f666a4ad99100d63ce35c/QxhxnvfeV6efkxwUFHwjI.png","fullname":"Bill Yuchen Lin","name":"yuchenlin","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"❤️","users":["davanstrien","plaguss","philschmid","osanseviero","Ali-C137","Rybens","lewtun","xiaotianhan","gabrielmbmb","clem","vladbogo","koutch","kramp","thivav","Nindaleth"],"count":15},{"reaction":"🤯","users":["philschmid","osanseviero","Ali-C137","clem"],"count":4},{"reaction":"🤗","users":["Ali-C137","clem"],"count":2}],"publishedAt":"2024-02-26T17:17:58.000Z","updatedAt":"2024-02-27T13:27:25.726Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1627505688463-60107b385ac3e86b3ea4fc34.jpeg","fullname":"Daniel van Strien","name":"davanstrien","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1594651707950-noauth.jpeg","fullname":"Lewis Tunstall","name":"lewtun","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/dvilasuero/297458605697772","totalUniqueImpressions":268,"numComments":4},"comment":{"_id":"65dd0aa667c3363787a5f251","raw":"Unpopular opinion: this is the most impactful release of the day (because open)!","html":"Unpopular opinion: this is the most impactful release of the day (because open)!
\n"}},{"time":"2024-02-25T19:28:59.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"891608574268336","content":[{"type":"text","value":"🌟✨ Exciting Announcement: NVIDIA AI Foundation Models ✨🌟","raw":"🌟✨ Exciting Announcement: NVIDIA AI Foundation Models ✨🌟"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🚀 Interact effortlessly with the latest SOTA AI model APIs, all optimized on the powerful NVIDIA accelerated computing stack-right from your browser! 💻⚡","raw":"🚀 Interact effortlessly with the latest SOTA AI model APIs, all optimized on the powerful NVIDIA accelerated computing stack-right from your browser! 💻⚡"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔗 Web Page: ","raw":"🔗 Web Page: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/ai-foundation-models","raw":"https://catalog.ngc.nvidia.com/ai-foundation-models"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🌟🎯 Favorites:","raw":"🌟🎯 Favorites:"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Code Generation:","raw":"🔹 Code Generation:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ Code Llama 70B 📝🔥: ","raw":"1️⃣ Code Llama 70B 📝🔥: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/codellama-70b","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/codellama-70b"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Model 🤖: ","raw":"Model 🤖: "},{"type":"resource","resource":{"type":"model","id":"codellama/CodeLlama-70b-hf"},"url":"https://huggingface.co/codellama/CodeLlama-70b-hf","raw":"https://huggingface.co/codellama/CodeLlama-70b-hf"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Text and Code Generation:","raw":"🔹 Text and Code Generation:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ Gemma 7B 💬💻: ","raw":"1️⃣ Gemma 7B 💬💻: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/gemma-7b","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/gemma-7b"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Model 🤖: ","raw":"Model 🤖: "},{"type":"resource","resource":{"type":"model","id":"google/gemma-7b"},"url":"https://huggingface.co/google/gemma-7b","raw":"https://huggingface.co/google/gemma-7b"},{"type":"new_line","raw":"\n"},{"type":"text","value":"2️⃣ Yi-34B 📚💡: ","raw":"2️⃣ Yi-34B 📚💡: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/yi-34b","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/yi-34b"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Model 🤖: ","raw":"Model 🤖: "},{"type":"resource","resource":{"type":"model","id":"01-ai/Yi-34B"},"url":"https://huggingface.co/01-ai/Yi-34B","raw":"https://huggingface.co/01-ai/Yi-34B"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Text Generation:","raw":"🔹 Text Generation:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ Mamba-Chat 💬🐍: ","raw":"1️⃣ Mamba-Chat 💬🐍: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/mamba-chat","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/mamba-chat"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Model 🤖: ","raw":"Model 🤖: "},{"type":"resource","resource":{"type":"model","id":"havenhq/mamba-chat"},"url":"https://huggingface.co/havenhq/mamba-chat","raw":"https://huggingface.co/havenhq/mamba-chat"},{"type":"new_line","raw":"\n"},{"type":"text","value":"2️⃣ Llama 2 70B 📝🦙: ","raw":"2️⃣ Llama 2 70B 📝🦙: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/llama2-70b","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/llama2-70b"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Model 🤖: ","raw":"Model 🤖: "},{"type":"resource","resource":{"type":"model","id":"meta-llama/Llama-2-70b"},"url":"https://huggingface.co/meta-llama/Llama-2-70b","raw":"https://huggingface.co/meta-llama/Llama-2-70b"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Text-To-Text Translation:","raw":"🔹 Text-To-Text Translation:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ SeamlessM4T V2 🌐🔄: ","raw":"1️⃣ SeamlessM4T V2 🌐🔄: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/seamless-m4t2-t2tt","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/seamless-m4t2-t2tt"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Model 🤖: ","raw":"Model 🤖: "},{"type":"resource","resource":{"type":"model","id":"facebook/seamless-m4t-v2-large"},"url":"https://huggingface.co/facebook/seamless-m4t-v2-large","raw":"https://huggingface.co/facebook/seamless-m4t-v2-large"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Image Generation:","raw":"🔹 Image Generation:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ Stable Diffusion XL 🎨🔍: ","raw":"1️⃣ Stable Diffusion XL 🎨🔍: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/sdxl","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/sdxl"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Image Conversation:","raw":"🔹 Image Conversation:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ NeVA-22B 🗨️📸: ","raw":"1️⃣ NeVA-22B 🗨️📸: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/neva-22b","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/neva-22b"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Image Classification and Object Detection:","raw":"🔹 Image Classification and Object Detection:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ CLIP 🖼️🔍: ","raw":"1️⃣ CLIP 🖼️🔍: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/clip","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/clip"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Voice Conversion:","raw":"🔹 Voice Conversion:"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ Maxine Voice Font 🗣️🎶: ","raw":"1️⃣ Maxine Voice Font 🗣️🎶: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/voice-font","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/voice-font"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔹 Multimodal LLM (MLLM):","raw":"🔹 Multimodal LLM (MLLM):"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1️⃣ Kosmos-2 🌐👁️: ","raw":"1️⃣ Kosmos-2 🌐👁️: "},{"type":"link","href":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/kosmos-2","raw":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/kosmos-2"}],"rawContent":"🌟✨ Exciting Announcement: NVIDIA AI Foundation Models ✨🌟\n\n🚀 Interact effortlessly with the latest SOTA AI model APIs, all optimized on the powerful NVIDIA accelerated computing stack-right from your browser! 💻⚡\n\n🔗 Web Page: https://catalog.ngc.nvidia.com/ai-foundation-models\n\n🌟🎯 Favorites:\n\n🔹 Code Generation:\n1️⃣ Code Llama 70B 📝🔥: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/codellama-70b\nModel 🤖: https://huggingface.co/codellama/CodeLlama-70b-hf\n\n🔹 Text and Code Generation:\n1️⃣ Gemma 7B 💬💻: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/gemma-7b\nModel 🤖: https://huggingface.co/google/gemma-7b\n2️⃣ Yi-34B 📚💡: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/yi-34b\nModel 🤖: https://huggingface.co/01-ai/Yi-34B\n\n🔹 Text Generation:\n1️⃣ Mamba-Chat 💬🐍: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/mamba-chat\nModel 🤖: https://huggingface.co/havenhq/mamba-chat\n2️⃣ Llama 2 70B 📝🦙: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/llama2-70b\nModel 🤖: https://huggingface.co/meta-llama/Llama-2-70b\n\n🔹 Text-To-Text Translation:\n1️⃣ SeamlessM4T V2 🌐🔄: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/seamless-m4t2-t2tt\nModel 🤖: https://huggingface.co/facebook/seamless-m4t-v2-large\n\n🔹 Image Generation:\n1️⃣ Stable Diffusion XL 🎨🔍: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/sdxl\n\n🔹 Image Conversation:\n1️⃣ NeVA-22B 🗨️📸: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/neva-22b\n\n🔹 Image Classification and Object Detection:\n1️⃣ CLIP 🖼️🔍: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/clip\n\n🔹 Voice Conversion:\n1️⃣ Maxine Voice Font 🗣️🎶: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/voice-font\n\n🔹 Multimodal LLM (MLLM):\n1️⃣ Kosmos-2 🌐👁️: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/ai-foundation/models/kosmos-2","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/nRCxbVng_PPBqKd-Z3KVc.jpeg","fullname":"Dmitry Ryumin","name":"DmitryRyumin","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"❤️","users":["MiSTe-R","vladbogo","ssbagpcm","osanseviero","samusenps","Josemalyson","eramax","s3nh","Alexandro14","Csplk","ArunkumarVR","xiaotianhan","Samoed","shubhamg2208","Antreas","dillfrescott","yahma","Ali-C137","Nymbo"],"count":19},{"reaction":"👍","users":["kim512","Antreas","dillfrescott"],"count":3}],"publishedAt":"2024-02-25T12:01:22.000Z","updatedAt":"2024-02-25T19:57:56.207Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/nRCxbVng_PPBqKd-Z3KVc.jpeg","fullname":"Dmitry Ryumin","name":"DmitryRyumin","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/DmitryRyumin/891608574268336","totalUniqueImpressions":148,"numComments":2},"comment":{"_id":"65db94fbc8df3e1363ea9885","raw":"would be cool to have some integration with the HF hub","html":"would be cool to have some integration with the HF hub
\n"}},{"time":"2024-02-21T19:05:37.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"587142414571229","content":[{"type":"text","value":"I am thrilled to announce Gemma, new 2B and 7B models from Google, based on the same research and technology used to train the Gemini models! These models achieve state-of-the-art performance for their size, and are launched across Transformers, Google Cloud, and many other surfaces worldwide starting today.","raw":"I am thrilled to announce Gemma, new 2B and 7B models from Google, based on the same research and technology used to train the Gemini models! These models achieve state-of-the-art performance for their size, and are launched across Transformers, Google Cloud, and many other surfaces worldwide starting today."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Get started using and adapting Gemma in the model Collection: ","raw":"Get started using and adapting Gemma in the model Collection: "},{"type":"resource","resource":{"type":"collection","id":"google/gemma-release-65d5efbccdbb8c4202ec078b"},"url":"https://huggingface.co/collections/google/gemma-release-65d5efbccdbb8c4202ec078b","raw":"https://huggingface.co/collections/google/gemma-release-65d5efbccdbb8c4202ec078b"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"These launches are the product of an outstanding collaboration between the Google DeepMind and Hugging Face teams over the last few months -- very proud of the work both teams have done, from integration with Vertex AI to optimization across the stack. Read more about the partnership in the main launch by ","raw":"These launches are the product of an outstanding collaboration between the Google DeepMind and Hugging Face teams over the last few months -- very proud of the work both teams have done, from integration with Vertex AI to optimization across the stack. Read more about the partnership in the main launch by "},{"type":"mention","user":"philschmid","raw":"@philschmid"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"osanseviero","raw":"@osanseviero"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"pcuenq","raw":"@pcuenq"},{"type":"text","value":" on the launch blog: ","raw":" on the launch blog: "},{"type":"link","href":"https://huggingface.co/blog/gemma","raw":"https://huggingface.co/blog/gemma"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"More information below if you are curious about training details, eval results, and safety characteristics!","raw":"More information below if you are curious about training details, eval results, and safety characteristics!"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Gemma Tech Report: ","raw":"Gemma Tech Report: "},{"type":"link","href":"https://goo.gle/GemmaReport","raw":"https://goo.gle/GemmaReport"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Launch announcement: ","raw":"Launch announcement: "},{"type":"link","href":"https://blog.google/technology/developers/gemma-open-models/","raw":"https://blog.google/technology/developers/gemma-open-models/"}],"rawContent":"I am thrilled to announce Gemma, new 2B and 7B models from Google, based on the same research and technology used to train the Gemini models! These models achieve state-of-the-art performance for their size, and are launched across Transformers, Google Cloud, and many other surfaces worldwide starting today.\n\nGet started using and adapting Gemma in the model Collection: https://huggingface.co/collections/google/gemma-release-65d5efbccdbb8c4202ec078b\n\nThese launches are the product of an outstanding collaboration between the Google DeepMind and Hugging Face teams over the last few months -- very proud of the work both teams have done, from integration with Vertex AI to optimization across the stack. Read more about the partnership in the main launch by @philschmid @osanseviero @pcuenq on the launch blog: https://huggingface.co/blog/gemma\n\nMore information below if you are curious about training details, eval results, and safety characteristics!\n\nGemma Tech Report: https://goo.gle/GemmaReport\nLaunch announcement: https://blog.google/technology/developers/gemma-open-models/","author":{"avatarUrl":"/avatars/5802fcc5a5f1e089df41232892f8f502.svg","fullname":"Tris Warkentin","name":"trisfromgoogle","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1617264212503-603d25b75f9d390ab190b777.jpeg","fullname":"Pedro Cuenca","name":"pcuenq","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1624629516652-5ff5d596f244529b3ec0fb89.png","fullname":"Philipp Schmid","name":"philschmid","type":"user","isPro":false,"isHf":true,"isMod":false}],"reactions":[{"reaction":"🤗","users":["algoartist","megrisdal","andysalerno","osanseviero","victor","BrigitteTousi","alvarobartt","abidlabs","clem","jeffboudier","joaogante","DmitryRyumin","manvinder","kramp","VictorSanh","Alexandro14","samforeman","vladbogo","taufiqdp","pcuenq","Kukedlc","Madisen","chansung","MoritzLaurer","m-ric","andrewrreed","Anthonyg5005","abhishek","mathiasn1","nicolasdec","Cstark","ethix"],"count":32},{"reaction":"❤️","users":["alvarobartt","abidlabs","clem","thibaultM","osanseviero","AP123","jeffboudier","joaogante","VictorSanh","Alexandro14","pcuenq","Kukedlc","Madisen","Bils","chansung","SanRosenberg","andrewrreed","Bolamide","ZenithBrew","allknowingroger","abhishek","sugatoray"],"count":22},{"reaction":"🤯","users":["clem","osanseviero","VictorSanh","Kukedlc","Madisen","chansung","abhishek"],"count":7},{"reaction":"🤝","users":["osanseviero","Kukedlc","Madisen","abhishek","Cstark"],"count":5}],"publishedAt":"2024-02-21T18:15:08.000Z","updatedAt":"2024-07-13T05:57:40.755Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/637bb1382d2d9c4f2486b458/eBV9PwY2Kxd6NsQjVqtzZ.jpeg","fullname":"AP","name":"AP123","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/21d7eb677e73fb7305d0d8776bc527c1.svg","fullname":"HY","name":"ethix","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1617264212503-603d25b75f9d390ab190b777.jpeg","fullname":"Pedro Cuenca","name":"pcuenq","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/807bef46ed74bbcc6bd9cac9b9acf82b.svg","fullname":"Manvinder Singh","name":"manvinder","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/trisfromgoogle/587142414571229","totalUniqueImpressions":101,"numComments":6},"comment":{"_id":"65d649815ef764405d3b39fd","raw":"This is awesome!","html":"This is awesome!
\n"}},{"time":"2024-02-12T18:14:36.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"811383627678984","content":[{"type":"text","value":"Hear, hear, AMD MI300Xs have started to emerge much sooner than expected. ","raw":"Hear, hear, AMD MI300Xs have started to emerge much sooner than expected. "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Here is a 2-part benchmarks report on performing BLOOM-176B inference using ","raw":"Here is a 2-part benchmarks report on performing BLOOM-176B inference using "},{"type":"mention","user":"MSFTDeepSpeed","raw":"@MSFTDeepSpeed"},{"type":"text","value":" optimized for AMD MI300X.","raw":" optimized for AMD MI300X."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"1. ","raw":"1. "},{"type":"link","href":"https://www.evp.cloud/post/diving-deeper-insights-from-our-llm-inference-testing","raw":"https://www.evp.cloud/post/diving-deeper-insights-from-our-llm-inference-testing"},{"type":"new_line","raw":"\n"},{"type":"text","value":"2. ","raw":"2. "},{"type":"link","href":"https://www.evp.cloud/post/diving-deeper-insights-from-our-llm-inference-testing-part-2","raw":"https://www.evp.cloud/post/diving-deeper-insights-from-our-llm-inference-testing-part-2"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"This was published in response to our BLOOM-176B super-fast inference blog post ","raw":"This was published in response to our BLOOM-176B super-fast inference blog post "},{"type":"link","href":"https://huggingface.co/blog/bloom-inference-pytorch-scripts","raw":"https://huggingface.co/blog/bloom-inference-pytorch-scripts"},{"type":"text","value":" ","raw":" "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Note that these have 192GB of HBM!","raw":"Note that these have 192GB of HBM!"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"The NVIDIA monopoly is strong, but it'll have to start sharing the pie and hopefully drive the costs down at least somewhat.","raw":"The NVIDIA monopoly is strong, but it'll have to start sharing the pie and hopefully drive the costs down at least somewhat."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Thanks to ","raw":"Thanks to "},{"type":"link","href":"https://www.linkedin.com/in/eliovp","raw":"https://www.linkedin.com/in/eliovp"},{"type":"text","value":" for sharing this writeup with me.","raw":" for sharing this writeup with me."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"p.s. at the PyTorch conference in the fall, the AMD representative said we will see MI300X available to us mortals in Q4-2024/Q1-2025.","raw":"p.s. at the PyTorch conference in the fall, the AMD representative said we will see MI300X available to us mortals in Q4-2024/Q1-2025."}],"rawContent":"Hear, hear, AMD MI300Xs have started to emerge much sooner than expected. \n\nHere is a 2-part benchmarks report on performing BLOOM-176B inference using @MSFTDeepSpeed optimized for AMD MI300X.\n\n1. https://www.evp.cloud/post/diving-deeper-insights-from-our-llm-inference-testing\n2. https://www.evp.cloud/post/diving-deeper-insights-from-our-llm-inference-testing-part-2\n\nThis was published in response to our BLOOM-176B super-fast inference blog post https://huggingface.co/blog/bloom-inference-pytorch-scripts \n\nNote that these have 192GB of HBM!\n\nThe NVIDIA monopoly is strong, but it'll have to start sharing the pie and hopefully drive the costs down at least somewhat.\n\nThanks to https://www.linkedin.com/in/eliovp for sharing this writeup with me.\n\np.s. at the PyTorch conference in the fall, the AMD representative said we will see MI300X available to us mortals in Q4-2024/Q1-2025.","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1594311341799-5f07383b19cb630495b812cd.jpeg","fullname":"Stas Bekman","name":"stas","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"❤️","users":["osanseviero","julien-c","davanstrien","tomaarsen","VictorSanh","clem","pcuenq","muhtasham","merve","samusenps","Eliovp","andrewrreed","rwightman","clefourrier"],"count":14}],"publishedAt":"2024-02-12T00:12:10.000Z","updatedAt":"2024-02-13T08:09:17.321Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/NQtzmrDdbG0H8qkZvRyGk.jpeg","fullname":"Julien Chaumond","name":"julien-c","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6317233cc92fd6fee317e030/cJHSvvimr1kqgQfHOjO5n.png","fullname":"Tom Aarsen","name":"tomaarsen","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1594311341799-5f07383b19cb630495b812cd.jpeg","fullname":"Stas Bekman","name":"stas","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/bc9d6b4976a2006fd53ce27b55d776d6.svg","fullname":"Eliovp","name":"Eliovp","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/stas/811383627678984","totalUniqueImpressions":95,"numComments":5},"comment":{"_id":"65ca600cab05dc46af58918f","raw":"very cool!","html":"very cool!
\n"}},{"time":"2024-02-08T17:10:03.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"291700537437132","content":[{"type":"text","value":"🌠 Let's try to figure out this one as a community.","raw":"🌠 Let's try to figure out this one as a community."},{"type":"new_line","raw":"\n"},{"type":"text","value":"What reactions should we add to Posts and discussions?","raw":"What reactions should we add to Posts and discussions?"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"If the reaction you want is already in the replies give it a thumbs up (👍) if it's not just add it as a reply.","raw":"If the reaction you want is already in the replies give it a thumbs up (👍) if it's not just add it as a reply."}],"rawContent":"🌠 Let's try to figure out this one as a community.\nWhat reactions should we add to Posts and discussions?\n\nIf the reaction you want is already in the replies give it a thumbs up (👍) if it's not just add it as a reply.","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1616001397867-5f17f0a0925b9863e28ad517.png","fullname":"Victor Mustar","name":"victor","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5f17f0a0925b9863e28ad517/0ZbXuU3elPJbPfzynr0e8.png"}],"mentions":[],"reactions":[{"reaction":"👍","users":["julien-c","Sylvestre","santiviquez","kramp","taufiqdp","osanseviero","ArthurZ","samusenps","clem","mvaloatto","Chunte"],"count":11},{"reaction":"❤️","users":["Ali-C137","johnny961"],"count":2},{"reaction":"🤗","users":["bunnycore","NePe"],"count":2},{"reaction":"🤝","users":["Bkarine"],"count":1}],"publishedAt":"2024-02-08T12:16:12.000Z","updatedAt":"2024-05-26T19:57:54.356Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1616001397867-5f17f0a0925b9863e28ad517.png","fullname":"Victor Mustar","name":"victor","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/703dd06469aaac724c94f622262b14e8.svg","fullname":"Tiezhen WANG","name":"xianbao","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1657144463525-629a173153a72d997d3f57d0.jpeg","fullname":"Santiago Viquez","name":"santiviquez","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674683851722-62441cb7456803e95009a08f.jpeg","fullname":"Arthur Zucker","name":"ArthurZ","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/tKcEdCfPIJ6gi9iqw9fPl.jpeg","fullname":"Jose Roca","name":"waroca","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6340651b388c3fa40f9a5bc0/av1C4_S7bHGxAzOu8lOmG.jpeg","fullname":"Adam Molnar","name":"lunarflu","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/c0f13b0d660c45394f55e97e3e7e627b.svg","fullname":"Abhishek Verma","name":"thor-x-me","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1648113222875-6141a88b3a0ec78603c9e784.png","fullname":"Merve Noyan","name":"merve","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6538119803519fddb4a17e10/ffJMkdx-rM7VvLTCM6ri_.jpeg","fullname":"samusenps","name":"samusenps","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6530994e70a88b63f007324d/3oRlaZg34975k1wCxII6W.png","fullname":"wbag","name":"Walmart-the-bag","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/97oWkFBYwQXo-YPrOzRgF.png","fullname":"Kristiyan Lukanov","name":"HDRobots","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1675778487155-63d4c8ce13ae45b780792f32.jpeg","fullname":"Ohenenoo","name":"PeepDaSlan9","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64aac16fd4a402e8dce11ebe/W640vNvHPWwwG_u4TsRo0.png","fullname":"Jorge Vallego","name":"neovalle","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63893d4c184615e463aa24b8/S1flsX_26OF6ZJBVcPlaf.jpeg","fullname":"Matt Valoatto","name":"mvaloatto","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/IPtQulJIe7DlzL3GT5LOk.png","fullname":"Source of Truth Data Labs","name":"sourceoftruthdata","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63a369d98c0c89dcae3b8329/6OUJ7Hc9T1jXynYH3FGaf.png","fullname":"Adina Yakefu","name":"AdinaY","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/d744307aff3a30aee73384e574d4a33e.svg","fullname":"Bartosz Glinski","name":"BGlinek","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/X9OwFnTRMr-lGYFjNH645.png","fullname":"Josh Habdas","name":"vhsdev","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6459fa0f5b3111fbe83286e1/CtzrB_KpBE1NgI9mBCLC9.jpeg","fullname":"Louis Brulé Naudet","name":"louisbrulenaudet","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/626237d9bbcbd1c34f1bb231/EJrOjvAL-68qMCYdnvOrq.png","fullname":"Ali Elfilali","name":"Ali-C137","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/NQtzmrDdbG0H8qkZvRyGk.jpeg","fullname":"Julien Chaumond","name":"julien-c","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6527e89a8808d80ccff88b7a/BRKGVgk_dJO34ZOi3Slb_.jpeg","fullname":"Lain","name":"not-lain","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6612aedf09f16e7347dfa7e1/bPYjBXCedY_1fSIPjoBTY.jpeg","fullname":"Nishith Jain","name":"KingNish","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/victor/291700537437132","totalUniqueImpressions":56,"numComments":44},"comment":{"_id":"65c50aeb9096960aa81b3eb2","raw":"This comment has been hidden","html":"This comment has been hidden"}},{"time":"2024-02-05T17:52:30.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"470307658539102","content":[{"type":"text","value":"These past months, I've been busy baking a special sort of Croissant 🥐 with an awesome team ! ","raw":"These past months, I've been busy baking a special sort of Croissant 🥐 with an awesome team ! "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🥐 CroissantLLM is a truly bilingual language model trained on 3 trillion tokens of French and English data. In its size category (<2B), it is the best model in French, but it also rivals the best monolingual English models ! ","raw":"🥐 CroissantLLM is a truly bilingual language model trained on 3 trillion tokens of French and English data. In its size category (<2B), it is the best model in French, but it also rivals the best monolingual English models ! "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"💾 To train it, we collected, filtered and cleaned huge quantities of permissively licensed French data, across various domains (legal, administrative, cultural, scientific), and different text modalities (speech transcriptions, movie subtitles, encyclopedias, forums, webpages)... ","raw":"💾 To train it, we collected, filtered and cleaned huge quantities of permissively licensed French data, across various domains (legal, administrative, cultural, scientific), and different text modalities (speech transcriptions, movie subtitles, encyclopedias, forums, webpages)... "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"⚖️ Assessing LLM performance is not easy, especially outside of English, and to this end we crafted a novel evaluation benchmark, FrenchBench, aiming to assess reasoning, factual knowledge, and linguistic capabilities of models in French !","raw":"⚖️ Assessing LLM performance is not easy, especially outside of English, and to this end we crafted a novel evaluation benchmark, FrenchBench, aiming to assess reasoning, factual knowledge, and linguistic capabilities of models in French !"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🔎 The best current LLMs are hidden behind a shroud of mystery, trained with undisclosed training data mixes or strategies. We go the opposite way, releasing all of the project's artefacts (model checkpoints, data, training details, evaluation benchmarks...) We obtain 81 % of the Stanford FMTI transparency criterias, far ahead of even most open initiatives !","raw":"🔎 The best current LLMs are hidden behind a shroud of mystery, trained with undisclosed training data mixes or strategies. We go the opposite way, releasing all of the project's artefacts (model checkpoints, data, training details, evaluation benchmarks...) We obtain 81 % of the Stanford FMTI transparency criterias, far ahead of even most open initiatives !"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🧪Beyond a powerful industrial resource, our transparent initiative is a stepping stone for many scientific questions ! How does teaching a model two languages instead of one splits its monolingual ability ? Does training on so much French help the model integrate French-centric knowledge and cultural biases ? How does the model memorize the training data ?","raw":"🧪Beyond a powerful industrial resource, our transparent initiative is a stepping stone for many scientific questions ! How does teaching a model two languages instead of one splits its monolingual ability ? Does training on so much French help the model integrate French-centric knowledge and cultural biases ? How does the model memorize the training data ?"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Many more things to say, for those interested, I recommend checking out:","raw":"Many more things to say, for those interested, I recommend checking out:"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🗞️ The blogpost: ","raw":"🗞️ The blogpost: "},{"type":"link","href":"https://huggingface.co/blog/manu/croissant-llm-blog","raw":"https://huggingface.co/blog/manu/croissant-llm-blog"},{"type":"new_line","raw":"\n"},{"type":"text","value":"📖 The 45 page report with lots of gems: ","raw":"📖 The 45 page report with lots of gems: "},{"type":"link","href":"https://arxiv.org/abs/2402.00786","raw":"https://arxiv.org/abs/2402.00786"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🤖 Models, Data, Demo: ","raw":"🤖 Models, Data, Demo: "},{"type":"link","href":"https://huggingface.co/croissantllm","raw":"https://huggingface.co/croissantllm"},{"type":"new_line","raw":"\n"}],"rawContent":"These past months, I've been busy baking a special sort of Croissant 🥐 with an awesome team ! \n\n🥐 CroissantLLM is a truly bilingual language model trained on 3 trillion tokens of French and English data. In its size category (<2B), it is the best model in French, but it also rivals the best monolingual English models ! \n\n💾 To train it, we collected, filtered and cleaned huge quantities of permissively licensed French data, across various domains (legal, administrative, cultural, scientific), and different text modalities (speech transcriptions, movie subtitles, encyclopedias, forums, webpages)... \n\n⚖️ Assessing LLM performance is not easy, especially outside of English, and to this end we crafted a novel evaluation benchmark, FrenchBench, aiming to assess reasoning, factual knowledge, and linguistic capabilities of models in French !\n\n🔎 The best current LLMs are hidden behind a shroud of mystery, trained with undisclosed training data mixes or strategies. We go the opposite way, releasing all of the project's artefacts (model checkpoints, data, training details, evaluation benchmarks...) We obtain 81 % of the Stanford FMTI transparency criterias, far ahead of even most open initiatives !\n\n🧪Beyond a powerful industrial resource, our transparent initiative is a stepping stone for many scientific questions ! How does teaching a model two languages instead of one splits its monolingual ability ? Does training on so much French help the model integrate French-centric knowledge and cultural biases ? How does the model memorize the training data ?\n\nMany more things to say, for those interested, I recommend checking out:\n\n🗞️ The blogpost: https://huggingface.co/blog/manu/croissant-llm-blog\n📖 The 45 page report with lots of gems: https://arxiv.org/abs/2402.00786\n🤖 Models, Data, Demo: https://huggingface.co/croissantllm\n","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1654090481550-60f2e021adf471cbdf8bb660.jpeg","fullname":"Manuel Faysse","name":"manu","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"❤️","users":["eliolio","gsarti","euclaise","eliebak","kramp","yozozaya","clem","blanchon","victor","mvaloatto","osanseviero","samusenps","santiviquez","ksiabani","batmac","nouamanetazi","davanstrien","fffiloni","Ali-C137","sbrandeis"],"count":20},{"reaction":"🤯","users":["fffiloni"],"count":1}],"publishedAt":"2024-02-05T15:12:40.000Z","updatedAt":"2024-02-06T07:27:15.701Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670231290373-5e7749883d77a72421292d07.jpeg","fullname":"Gabriele Sarti","name":"gsarti","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64955109ac70da05b7aacb9a/bZKEz24ZfaWDSI33yHUmR.png","fullname":"Kostas Siabanis","name":"ksiabani","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/manu/470307658539102","totalUniqueImpressions":1015,"numComments":3},"comment":{"_id":"65c1205e65086cabf47250fa","raw":"🇫🇷🇫🇷🇫🇷","html":"🇫🇷🇫🇷🇫🇷
\n"}},{"time":"2024-01-31T19:19:39.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"183807800492559","content":[{"type":"text","value":"🚀 The Open Source AI community needs more open datasets for improving Open LLMs. Excited to share our new open dataset for boosting chat models:","raw":"🚀 The Open Source AI community needs more open datasets for improving Open LLMs. Excited to share our new open dataset for boosting chat models:"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🎉 Welcome Distilabel Capybara DPO, a multi-turn, high-quality preference dataset. ","raw":"🎉 Welcome Distilabel Capybara DPO, a multi-turn, high-quality preference dataset. "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"dataset","id":"argilla/distilabel-capybara-dpo-7k-binarized"},"url":"https://huggingface.co/datasets/argilla/distilabel-capybara-dpo-7k-binarized","raw":"https://huggingface.co/datasets/argilla/distilabel-capybara-dpo-7k-binarized"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Why?","raw":"Why?"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Best closed chat models are built on top of multi-turn dialogue preference data. The OSS community lacks these datasets. This dataset is the first in the series to close this gap.","raw":"Best closed chat models are built on top of multi-turn dialogue preference data. The OSS community lacks these datasets. This dataset is the first in the series to close this gap."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Is this dataset useful?","raw":"Is this dataset useful?"},{"type":"new_line","raw":"\n"},{"type":"text","value":"To test this dataset, we've built our virtual launching partner:","raw":"To test this dataset, we've built our virtual launching partner:"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"🎉 Welcome CapybaraHermes, a preference tuned OpenHermes with increased second turn capabilities on MTBench","raw":"🎉 Welcome CapybaraHermes, a preference tuned OpenHermes with increased second turn capabilities on MTBench"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"model","id":"argilla/CapybaraHermes-2.5-Mistral-7B"},"url":"https://huggingface.co/argilla/CapybaraHermes-2.5-Mistral-7B","raw":"https://huggingface.co/argilla/CapybaraHermes-2.5-Mistral-7B"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"As usual, models are the least important to us. We like to focus on the data. Our mission is to build and share high-quality datasets, sharing our methods in the open so the community can improve upon them.","raw":"As usual, models are the least important to us. We like to focus on the data. Our mission is to build and share high-quality datasets, sharing our methods in the open so the community can improve upon them."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"That's why, we took some time to describe the full methodology on the dataset card, check it out and give us feedback! Data and methods are never perfect!","raw":"That's why, we took some time to describe the full methodology on the dataset card, check it out and give us feedback! Data and methods are never perfect!"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Finally, this is just a preview version and would love to collaborate with you to add more benchmarking results, what hyperparams work for DPO'ing models, what mix of datasets, etc.","raw":"Finally, this is just a preview version and would love to collaborate with you to add more benchmarking results, what hyperparams work for DPO'ing models, what mix of datasets, etc."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Expect some more datasets in the coming weeks. Let's build the best data for AI, together.","raw":"Expect some more datasets in the coming weeks. Let's build the best data for AI, together."}],"rawContent":"🚀 The Open Source AI community needs more open datasets for improving Open LLMs. Excited to share our new open dataset for boosting chat models:\n\n🎉 Welcome Distilabel Capybara DPO, a multi-turn, high-quality preference dataset. \n\nhttps://huggingface.co/datasets/argilla/distilabel-capybara-dpo-7k-binarized\n\nWhy?\nBest closed chat models are built on top of multi-turn dialogue preference data. The OSS community lacks these datasets. This dataset is the first in the series to close this gap.\n\nIs this dataset useful?\nTo test this dataset, we've built our virtual launching partner:\n\n🎉 Welcome CapybaraHermes, a preference tuned OpenHermes with increased second turn capabilities on MTBench\n\nhttps://huggingface.co/argilla/CapybaraHermes-2.5-Mistral-7B\n\nAs usual, models are the least important to us. We like to focus on the data. Our mission is to build and share high-quality datasets, sharing our methods in the open so the community can improve upon them.\n\nThat's why, we took some time to describe the full methodology on the dataset card, check it out and give us feedback! Data and methods are never perfect!\n\nFinally, this is just a preview version and would love to collaborate with you to add more benchmarking results, what hyperparams work for DPO'ing models, what mix of datasets, etc.\n\nExpect some more datasets in the coming weeks. Let's build the best data for AI, together.","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60420dccc15e823a685f2b03/Dn7QTyy9SZ7jKN6xpufVD.png","fullname":"Daniel Vila","name":"dvilasuero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/60420dccc15e823a685f2b03/ddfzg2ruWBF4O2X6agDIP.png"}],"mentions":[],"reactions":[{"reaction":"❤️","users":["davanstrien","lhoestq","osanseviero","victor","severo","julien-c","samusenps","clem","Smorty100","Arjun4707","gabrielmbmb","kramp","mayurninama","davidberenstein1957","hfthorvaldur","leonardlin","alvarobartt","Ali-C137","sbrandeis"],"count":19},{"reaction":"🤗","users":["davanstrien","osanseviero","severo","julien-c","clem","Smorty100","Arjun4707","Ali-C137"],"count":8}],"publishedAt":"2024-01-31T17:25:59.000Z","updatedAt":"2024-01-31T19:19:40.329Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/dvilasuero/183807800492559","totalUniqueImpressions":20,"numComments":1},"comment":{"_id":"65ba9d4b00e143d6727d0a03","raw":"Very cool!","html":"Very cool!
\n"}},{"time":"2024-01-31T19:18:32.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"898508490022091","content":[{"type":"text","value":"🔥 New LLM leaderboard on the hub: an Enterprise Scenarios Leaderboard!","raw":"🔥 New LLM leaderboard on the hub: an Enterprise Scenarios Leaderboard!"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"This work evaluates LLMs on several real world use cases (Finance documents, Legal confidentiality, Customer support, ...), which makes it grounded, and interesting for companies! 🏢","raw":"This work evaluates LLMs on several real world use cases (Finance documents, Legal confidentiality, Customer support, ...), which makes it grounded, and interesting for companies! 🏢"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Bonus: the test set is private, so it's hard to game 🔥","raw":"Bonus: the test set is private, so it's hard to game 🔥"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"space","id":"PatronusAI/enterprise_scenarios_leaderboard"},"url":"https://huggingface.co/spaces/PatronusAI/enterprise_scenarios_leaderboard","raw":"https://huggingface.co/spaces/PatronusAI/enterprise_scenarios_leaderboard"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Side note: I discovered through this benchmark that you could evaluate \"Engagingness\" of an LLM, which could also be interesting for our LLM fine-tuning community out there.","raw":"Side note: I discovered through this benchmark that you could evaluate \"Engagingness\" of an LLM, which could also be interesting for our LLM fine-tuning community out there."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Read more about their different tasks and metrics in the intro blog: ","raw":"Read more about their different tasks and metrics in the intro blog: "},{"type":"link","href":"https://huggingface.co/blog/leaderboards-on-the-hub-patronus","raw":"https://huggingface.co/blog/leaderboards-on-the-hub-patronus"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Congrats to ","raw":"Congrats to "},{"type":"mention","user":"sunitha98","raw":"@sunitha98"},{"type":"text","value":" who led the leaderboard implementation, and to ","raw":" who led the leaderboard implementation, and to "},{"type":"mention","user":"rebeccaqian","raw":"@rebeccaqian"},{"type":"text","value":" and ","raw":" and "},{"type":"mention","user":"anandnk24","raw":"@anandnk24"},{"type":"text","value":", all at Patronus AI !","raw":", all at Patronus AI !"}],"rawContent":"🔥 New LLM leaderboard on the hub: an Enterprise Scenarios Leaderboard!\n\nThis work evaluates LLMs on several real world use cases (Finance documents, Legal confidentiality, Customer support, ...), which makes it grounded, and interesting for companies! 🏢\nBonus: the test set is private, so it's hard to game 🔥\nhttps://huggingface.co/spaces/PatronusAI/enterprise_scenarios_leaderboard\n\nSide note: I discovered through this benchmark that you could evaluate \"Engagingness\" of an LLM, which could also be interesting for our LLM fine-tuning community out there.\n\nRead more about their different tasks and metrics in the intro blog: https://huggingface.co/blog/leaderboards-on-the-hub-patronus\n\nCongrats to @sunitha98 who led the leaderboard implementation, and to @rebeccaqian and @anandnk24, all at Patronus AI !","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1644340617257-noauth.png","fullname":"Clémentine Fourrier","name":"clefourrier","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64800401dd29d0de8121fa3d/jQyJeDW1-NsX5sd1tsRjF.jpeg","fullname":"Anand Kannappan","name":"anandnk24","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670593076071-noauth.jpeg","fullname":"Rebecca Qian","name":"rebeccaqian","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/631f35b6e488fcef857cdbda/XZsR6bfelyUX32-T8UTtu.jpeg","fullname":"Selvan Sunitha Ravi","name":"sunitha98","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"❤️","users":["osanseviero","RebeccaQian1","clem","marcsun13","samusenps","Pelmeshek","Ali-C137","anandnk24","gblazex"],"count":9}],"publishedAt":"2024-01-31T17:49:26.000Z","updatedAt":"2024-02-01T07:33:06.944Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1644340617257-noauth.png","fullname":"Clémentine Fourrier","name":"clefourrier","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/clefourrier/898508490022091","totalUniqueImpressions":45,"numComments":2},"comment":{"_id":"65ba9d0804c8a5d626ae2487","raw":"very useful! This is the link to the leaderboard btw: https://huggingface.co/spaces/PatronusAI/enterprise_scenarios_leaderboard\n","html":"very useful! This is the link to the leaderboard btw: https://huggingface.co/spaces/PatronusAI/enterprise_scenarios_leaderboard
\n"}},{"time":"2024-01-31T19:18:12.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"306425415509800","content":[{"type":"text","value":"📣 NEW on HF","raw":"📣 NEW on HF"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"the Dataset Viewer is now available on *private datasets* too","raw":"the Dataset Viewer is now available on *private datasets* too"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"You need to be a PRO or a Enterprise Hub user. 🔥","raw":"You need to be a PRO or a Enterprise Hub user. 🔥"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Great work from our Datasets team 🥰: ","raw":"Great work from our Datasets team 🥰: "},{"type":"mention","user":"lhoestq","raw":"@lhoestq"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"severo","raw":"@severo"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"polinaeterna","raw":"@polinaeterna"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"asoria","raw":"@asoria"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"albertvillanova","raw":"@albertvillanova"},{"type":"text","value":" and the whole team 🥰 ","raw":" and the whole team 🥰 "}],"rawContent":"📣 NEW on HF\n\nthe Dataset Viewer is now available on *private datasets* too\n\nYou need to be a PRO or a Enterprise Hub user. 🔥\n\nGreat work from our Datasets team 🥰: @lhoestq @severo @polinaeterna @asoria @albertvillanova and the whole team 🥰 ","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/NQtzmrDdbG0H8qkZvRyGk.jpeg","fullname":"Julien Chaumond","name":"julien-c","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5dd96eb166059660ed1ee413/k7X14tuI4OD1h4-QS0N01.png"}],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1606406298765-noauth.jpeg","fullname":"Albert Villanova","name":"albertvillanova","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674055965173-noauth.jpeg","fullname":"Andrea Soria","name":"asoria","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1594214747713-5e9ecfc04957053f60648a3e.png","fullname":"Quentin Lhoest","name":"lhoestq","type":"user","isPro":true,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1645527878855-61dd9f18f187b39868bd157e.jpeg","fullname":"Polina Kazakova","name":"polinaeterna","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60a76b174e24361791fe822d/inEvYwrd4z0xvRQN3ikdE.jpeg","fullname":"Sylvain Lesage","name":"severo","type":"user","isPro":true,"isHf":true,"isMod":false}],"reactions":[{"reaction":"❤️","users":["lhoestq","severo","clem","blanchon","osanseviero","pierrci","albertvillanova","VanshGehlot","vladbogo","AdinaY","BrigitteTousi","michaljunczyk","not-lain","Nymbo"],"count":14},{"reaction":"👍","users":["fffiloni","clem","osanseviero","Pelmeshek","kramp","mvaloatto","awacke1","AdinaY","BrigitteTousi","Tonic","not-lain","Nymbo"],"count":12},{"reaction":"🤯","users":["pierrci","BrigitteTousi","not-lain"],"count":3},{"reaction":"🤗","users":["albertvillanova","BrigitteTousi","not-lain"],"count":3}],"publishedAt":"2024-01-31T18:39:12.000Z","updatedAt":"2024-01-31T19:18:12.723Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/julien-c/306425415509800","totalUniqueImpressions":18,"numComments":1},"comment":{"_id":"65ba9cf479d8f2796ef9f1d5","raw":"very cool!","html":"very cool!
\n"}},{"time":"2024-01-30T17:02:31.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"251812180396517","content":[{"type":"text","value":"So impressed with the speed and accuracy of ","raw":"So impressed with the speed and accuracy of "},{"type":"resource","resource":{"type":"space","id":"vikhyatk/moondream1"},"url":"https://huggingface.co/spaces/vikhyatk/moondream1","raw":"https://huggingface.co/spaces/vikhyatk/moondream1"},{"type":"text","value":" by ","raw":" by "},{"type":"new_line","raw":"\n"},{"type":"mention","user":"vikhyatk","raw":"@vikhyatk"},{"type":"text","value":" (especially the last answer 😝😝😝). ","raw":" (especially the last answer 😝😝😝). "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Open multi-modal models have gone a long way!","raw":"Open multi-modal models have gone a long way!"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Model: ","raw":"Model: "},{"type":"resource","resource":{"type":"model","id":"vikhyatk/moondream1"},"url":"https://huggingface.co/vikhyatk/moondream1","raw":"https://huggingface.co/vikhyatk/moondream1"}],"rawContent":"So impressed with the speed and accuracy of https://huggingface.co/spaces/vikhyatk/moondream1 by \r\n@vikhyatk (especially the last answer 😝😝😝). \r\n\r\nOpen multi-modal models have gone a long way!\r\n\r\nModel: https://huggingface.co/vikhyatk/moondream1","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/EZl2FLjsXK_npzClZAvWr.png"},{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/HbDx5UyOyy56lNLdJFQJT.png"},{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/rzBdopT5rKUy4iYgJLhWY.png"},{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/LMRWKzlpbPE2CjIuF9n-7.png"}],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63117568fa95534e218da163/8h9zN8aKRxPLBnXW7sqY9.jpeg","fullname":"vik","name":"vikhyatk","type":"user","isPro":true,"isHf":false,"isMod":false}],"reactions":[{"reaction":"❤️","users":["vikhyatk","victor","osanseviero","clem","GloriaMK","fffiloni","akhaliq","karankatke","masonchambersiii","abidlabs","samusenps","floattingabby","gate369","jeffboudier","ArthurZ","combox","snarktank","michaelkleyn","ivanfioravanti"],"count":19},{"reaction":"🤯","users":["pierrci","akhaliq"],"count":2}],"publishedAt":"2024-01-30T17:02:31.000Z","updatedAt":"2024-01-30T17:38:54.297Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/clem/251812180396517","totalUniqueImpressions":341,"numComments":1}},{"time":"2024-01-29T16:36:17.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"954032654837367","content":[{"type":"text","value":"With the Google announcement last week, I think we're now officially the only AI startup out there who has commercial collaborations with all the major cloud providers (AWS, GCP, Azure) and hardware providers (Nvidia, AMD, Intel, Qualcomm,...), making our vision of being the independent and agnostic platform for all AI builders truer than ever! ","raw":"With the Google announcement last week, I think we're now officially the only AI startup out there who has commercial collaborations with all the major cloud providers (AWS, GCP, Azure) and hardware providers (Nvidia, AMD, Intel, Qualcomm,...), making our vision of being the independent and agnostic platform for all AI builders truer than ever! "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Let's go!","raw":"Let's go!"}],"rawContent":"With the Google announcement last week, I think we're now officially the only AI startup out there who has commercial collaborations with all the major cloud providers (AWS, GCP, Azure) and hardware providers (Nvidia, AMD, Intel, Qualcomm,...), making our vision of being the independent and agnostic platform for all AI builders truer than ever! \r\n\r\nLet's go!","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"❤️","users":["on1onmangoes","osanseviero","santiviquez","mvaloatto","abbasm2","donbale","gsarti","vladbogo","MadElf1337","SivilTaram","taufiqdp","rreed-pha","victor","neovalle","qnixsynapse","andrewrreed","vivekrp","merve","joaogante","Bondstreet","LucasThil","maywell","duraad","hysts","thibaultM"],"count":25}],"publishedAt":"2024-01-29T16:36:17.000Z","updatedAt":"2024-01-29T16:36:17.971Z","commentators":[],"url":"/posts/clem/954032654837367","totalUniqueImpressions":22,"numComments":0}},{"time":"2024-01-26T18:27:47.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"575326441220299","content":[{"type":"text","value":"In 2024, we're expanding from open weights to open EVERYTHING (datasets, training scripts,...). ","raw":"In 2024, we're expanding from open weights to open EVERYTHING (datasets, training scripts,...). "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Excited to see this dataset release in French by ","raw":"Excited to see this dataset release in French by "},{"type":"mention","user":"Pclanglais","raw":"@Pclanglais"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"carbonbasedLLM","raw":"@carbonbasedLLM"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"anastasiastasenko","raw":"@anastasiastasenko"},{"type":"text","value":":","raw":":"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"dataset","id":"PleIAs/French-PD-Newspapers"},"url":"https://huggingface.co/datasets/PleIAs/French-PD-Newspapers","raw":"https://huggingface.co/datasets/PleIAs/French-PD-Newspapers"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"\"To give you an idea of the size, the full French Wikipedia is about 2 billon words. This is 40 times larger.\"","raw":"\"To give you an idea of the size, the full French Wikipedia is about 2 billon words. This is 40 times larger.\""}],"rawContent":"In 2024, we're expanding from open weights to open EVERYTHING (datasets, training scripts,...). \r\n\r\nExcited to see this dataset release in French by @Pclanglais @carbonbasedLLM @anastasiastasenko:\r\nhttps://huggingface.co/datasets/PleIAs/French-PD-Newspapers\r\n\r\n\"To give you an idea of the size, the full French Wikipedia is about 2 billon words. This is 40 times larger.\"","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/eNUY-tTX5_EB3TKN3fJoE.jpeg","fullname":"Anastasia Stasenko","name":"anastasiastasenko","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65b3ddd382d38451342aca57/wYBAou-q38bKVROYd1YhN.jpeg","fullname":"Carlos Rosas","name":"carbonbasedLLM","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64ce091a9e9ca8123d7a42b0/OEPggp82RwigxNLL35LgT.jpeg","fullname":"Pierre-Carl Langlais","name":"Pclanglais","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"❤️","users":["Maani","Pclanglais","afrideva","tomaarsen","osanseviero","samusenps","glenn-parham","mlabonne","ucalyptus","abidlabs","Sylvestre","JaiSurya","anastasiastasenko","mvaloatto","sparkycollier","linoyts","MexIvanov","AdinaY","brishtiteveja"],"count":19}],"publishedAt":"2024-01-26T18:27:47.000Z","updatedAt":"2024-01-26T18:27:54.982Z","commentators":[],"url":"/posts/clem/575326441220299","totalUniqueImpressions":20,"numComments":0}},{"time":"2024-01-26T11:26:44.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"163107901738446","content":[{"type":"text","value":"My TED talk is finally live!! I proposed the recipe for the \"Foundation Agent\": a single model that learns how to act in different worlds. LLM scales across lots and lots of texts. Foundation Agent scales across lots of lots of realities. If it is able to master 10,000 diverse simulated realities, it may well generalize to our physical world, which is simply the 10,001st reality.","raw":"My TED talk is finally live!! I proposed the recipe for the \"Foundation Agent\": a single model that learns how to act in different worlds. LLM scales across lots and lots of texts. Foundation Agent scales across lots of lots of realities. If it is able to master 10,000 diverse simulated realities, it may well generalize to our physical world, which is simply the 10,001st reality."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Why do we want a single Foundation Agent instead of many smaller models? I'd like to quote the idea from my friend Prof. Yuke Zhu's CoRL keynote talk. If we trace each AI field's evolution, we'd find this pattern:","raw":"Why do we want a single Foundation Agent instead of many smaller models? I'd like to quote the idea from my friend Prof. Yuke Zhu's CoRL keynote talk. If we trace each AI field's evolution, we'd find this pattern:"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Specialist -> Generalist -> Specializing Generalist","raw":"Specialist -> Generalist -> Specializing Generalist"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"And the \"specialized generalist\" is often way more powerful than the original specialist. Just like distilled versions of LlaMA are way better than custom-built NLP systems 5 years ago.","raw":"And the \"specialized generalist\" is often way more powerful than the original specialist. Just like distilled versions of LlaMA are way better than custom-built NLP systems 5 years ago."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"TED talks do not have teleprompters!! All I have is a \"confidence monitor\" at my foot, showing only the current slide and timer. That means I need to memorize the whole speech. It sounds intimidating at first, but turns out to be the best way to connect with the audience and deliver the ideas right to heart. ","raw":"TED talks do not have teleprompters!! All I have is a \"confidence monitor\" at my foot, showing only the current slide and timer. That means I need to memorize the whole speech. It sounds intimidating at first, but turns out to be the best way to connect with the audience and deliver the ideas right to heart. "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Happy to share my slides with all of you! ","raw":"Happy to share my slides with all of you! "},{"type":"link","href":"https://drive.google.com/file/d/1NSY6MxMu3OPQ4U6hx0OxPq7EQB5XTcAG/view?usp=sharing","raw":"https://drive.google.com/file/d/1NSY6MxMu3OPQ4U6hx0OxPq7EQB5XTcAG/view?usp=sharing"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"The video is only 10 min. I promise it's well worth your time! ","raw":"The video is only 10 min. I promise it's well worth your time! "},{"type":"new_line","raw":"\n"},{"type":"link","href":"https://www.ted.com/talks/jim_fan_the_next_grand_challenge_for_ai","raw":"https://www.ted.com/talks/jim_fan_the_next_grand_challenge_for_ai"}],"rawContent":"My TED talk is finally live!! I proposed the recipe for the \"Foundation Agent\": a single model that learns how to act in different worlds. LLM scales across lots and lots of texts. Foundation Agent scales across lots of lots of realities. If it is able to master 10,000 diverse simulated realities, it may well generalize to our physical world, which is simply the 10,001st reality.\r\n\r\nWhy do we want a single Foundation Agent instead of many smaller models? I'd like to quote the idea from my friend Prof. Yuke Zhu's CoRL keynote talk. If we trace each AI field's evolution, we'd find this pattern:\r\n\r\nSpecialist -> Generalist -> Specializing Generalist\r\n\r\nAnd the \"specialized generalist\" is often way more powerful than the original specialist. Just like distilled versions of LlaMA are way better than custom-built NLP systems 5 years ago.\r\n\r\nTED talks do not have teleprompters!! All I have is a \"confidence monitor\" at my foot, showing only the current slide and timer. That means I need to memorize the whole speech. It sounds intimidating at first, but turns out to be the best way to connect with the audience and deliver the ideas right to heart. \r\n\r\nHappy to share my slides with all of you! https://drive.google.com/file/d/1NSY6MxMu3OPQ4U6hx0OxPq7EQB5XTcAG/view?usp=sharing\r\n\r\nThe video is only 10 min. I promise it's well worth your time! \r\nhttps://www.ted.com/talks/jim_fan_the_next_grand_challenge_for_ai","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/628352f309aa80237c6c3e4b/W2dkeN-jJ137HfbYKo088.jpeg","fullname":"Jim Fan","name":"jimfan","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/628352f309aa80237c6c3e4b/Mz22ypyZX3dwsmsOseVfx.png"}],"mentions":[],"reactions":[{"reaction":"❤️","users":["truong-xuan-linh","s3nh","gsarti","samusenps","osanseviero","clem","victor","julien-c","akhaliq","juancopi81","afrideva","frankdata"],"count":12},{"reaction":"🤯","users":["samusenps","osanseviero","clem","julien-c","akhaliq","sbrandeis"],"count":6},{"reaction":"👍","users":["jm8","mvaloatto","akhaliq"],"count":3}],"publishedAt":"2024-01-26T07:53:26.000Z","updatedAt":"2024-01-26T11:55:28.779Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/w6hr-DEQot4VVkoyRIBiy.png","fullname":"Omar Sanseviero","name":"osanseviero","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/NQtzmrDdbG0H8qkZvRyGk.jpeg","fullname":"Julien Chaumond","name":"julien-c","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/jimfan/163107901738446","totalUniqueImpressions":27,"numComments":3},"comment":{"_id":"65b396f47934f3ef33b7bcae","raw":"Very cool!","html":"Very cool!
\n"}},{"time":"2024-01-25T19:52:38.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"521953578632447","content":[{"type":"text","value":"Google + Hugging Face + Open-Source AI = 🔥🔥🔥","raw":"Google + Hugging Face + Open-Source AI = 🔥🔥🔥"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"link","href":"https://huggingface.co/blog/gcp-partnership","raw":"https://huggingface.co/blog/gcp-partnership"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"link","href":"https://finance.yahoo.com/video/google-hugging-face-alliance-spur-173016882.html","raw":"https://finance.yahoo.com/video/google-hugging-face-alliance-spur-173016882.html"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"link","href":"https://www.theverge.com/2024/1/25/24050445/google-cloud-hugging-face-ai-developer-access","raw":"https://www.theverge.com/2024/1/25/24050445/google-cloud-hugging-face-ai-developer-access"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"link","href":"https://www.bloomberg.com/news/articles/2024-01-25/google-to-team-up-with-startup-hugging-face-to-host-ai-software","raw":"https://www.bloomberg.com/news/articles/2024-01-25/google-to-team-up-with-startup-hugging-face-to-host-ai-software"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"link","href":"https://www.reuters.com/technology/google-cloud-partners-with-hugging-face-attract-ai-developers-2024-01-25/","raw":"https://www.reuters.com/technology/google-cloud-partners-with-hugging-face-attract-ai-developers-2024-01-25/"}],"rawContent":"Google + Hugging Face + Open-Source AI = 🔥🔥🔥\r\n\r\nhttps://huggingface.co/blog/gcp-partnership\r\n\r\nhttps://finance.yahoo.com/video/google-hugging-face-alliance-spur-173016882.html\r\n\r\nhttps://www.theverge.com/2024/1/25/24050445/google-cloud-hugging-face-ai-developer-access\r\n\r\nhttps://www.bloomberg.com/news/articles/2024-01-25/google-to-team-up-with-startup-hugging-face-to-host-ai-software\r\n\r\nhttps://www.reuters.com/technology/google-cloud-partners-with-hugging-face-attract-ai-developers-2024-01-25/","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/bLm9EPbhttz4YfonBNba7.png"}],"mentions":[],"reactions":[{"reaction":"👍","users":["EmilyWitko","osanseviero","vladbogo","neovalle","FranzNo","mrfakename","radames","akhaliq","not-lain","afrideva","Sylvestre"],"count":11},{"reaction":"❤️","users":["osanseviero","Paul-HF","vicgalle","akhaliq","koisose","stas","not-lain","taufiqdp","sbrandeis"],"count":9},{"reaction":"🤗","users":["mrfakename","akhaliq","not-lain","kramp","mvaloatto"],"count":5},{"reaction":"🤝","users":["notsahil"],"count":1}],"publishedAt":"2024-01-25T19:52:38.000Z","updatedAt":"2024-01-25T21:17:05.575Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62e54f0eae9d3f10acb95cb9/VAyk05hqB3OZWXEZW-B0q.png","fullname":"mrfakename","name":"mrfakename","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6538119803519fddb4a17e10/ffJMkdx-rM7VvLTCM6ri_.jpeg","fullname":"samusenps","name":"samusenps","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/521953578632447","totalUniqueImpressions":33,"numComments":3}},{"time":"2024-01-22T19:38:40.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"970025506569107","content":[{"type":"text","value":"Re-posting ","raw":"Re-posting "},{"type":"mention","user":"karpathy","raw":"@karpathy"},{"type":"text","value":"'s blogpost here because it's down on ","raw":"'s blogpost here because it's down on "},{"type":"link","href":"https://karpathy.github.io/2024/01/21/selfdriving-agi","raw":"https://karpathy.github.io/2024/01/21/selfdriving-agi"},{"type":"text","value":". What do you all think?","raw":". What do you all think?"}],"rawContent":"Re-posting @karpathy's blogpost here because it's down on https://karpathy.github.io/2024/01/21/selfdriving-agi. What do you all think?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/6n3rfOGCL-J7ETz_0Iphm.png"},{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/itmgDQNABi9_x2o0s5qji.png"},{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/F8yJ2CY9oqQDwjypku-eH.png"},{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/sX1rNJXHbGBWqV0VoNUIO.png"},{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/O2F0HRuUXnL_LQSDLFdlV.png"}],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1660434061546-62f83661fe21cc4875221c0f.jpeg","fullname":"Andrej K","name":"karpathy","type":"user","isPro":false,"isHf":false,"isMod":false}],"reactions":[{"reaction":"👍","users":["elpreneurAbdo","RodolpheD","Paul-HF","SaylorTwift","stefan-jo","arnaudliotta","lunarflu","julien-c","merve","radames","pierrci"],"count":11},{"reaction":"🤯","users":["osanseviero","chansung","samu","lunarflu","julien-c","merve"],"count":6},{"reaction":"❤️","users":["taufiqdp","lunarflu","julien-c","merve"],"count":4}],"publishedAt":"2024-01-22T19:38:40.000Z","updatedAt":"2024-01-23T14:30:48.322Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/61caeda441f9432649f03ab6/0UdRCrzIqhedZblgfpMBk.png","fullname":"s3nh","name":"s3nh","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6538119803519fddb4a17e10/ffJMkdx-rM7VvLTCM6ri_.jpeg","fullname":"samusenps","name":"samusenps","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65ae8908b64e1c2389aafd94/bPoQffBL8fASGtvgDpf1F.png","fullname":"Luke Shaffer","name":"LukeES","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/a307da2b473607b1dbf668e7e1168b07.svg","fullname":"Srulik Ben David","name":"Srulikbd","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/970025506569107","totalUniqueImpressions":32,"numComments":4}},{"time":"2024-01-21T12:00:04.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"986151192658340","content":[{"type":"text","value":"There's a lot of interest in machine learning models that generate 3D objects, so Gradio now supports previewing STL files natively in the ","raw":"There's a lot of interest in machine learning models that generate 3D objects, so Gradio now supports previewing STL files natively in the "},{"type":"inline_code","code":"Model3D","raw":"`Model3D`"},{"type":"text","value":" component. Huge thanks to ","raw":" component. Huge thanks to "},{"type":"inline_code","code":"Monius","raw":"`Monius`"},{"type":"text","value":" for the contribution 🔥🔥","raw":" for the contribution 🔥🔥"},{"type":"new_line","raw":"\n"}],"rawContent":"There's a lot of interest in machine learning models that generate 3D objects, so Gradio now supports previewing STL files natively in the `Model3D` component. Huge thanks to `Monius` for the contribution 🔥🔥\r\n","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1621947938344-noauth.png","fullname":"Abubakar Abid","name":"abidlabs","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/608b8bb39d7c9519b4adae19/td-cW9V5W_e4oMke2ElqX.png"}],"mentions":[],"reactions":[{"reaction":"❤️","users":["samusenps","Maykeye","osanseviero","NerdN","Dlbk","clem","rreed","linoyts","johko","Chunte","HuggyMonkey","lunarflu","ciCic","AgastyaPatel","marianbasti","goendalf666","sugatoray","conceptron"],"count":18},{"reaction":"🤗","users":["samusenps","NerdN","clem","abidlabs","taufiqdp","NeuralNovel","Chunte","lunarflu"],"count":8}],"publishedAt":"2024-01-21T01:05:22.000Z","updatedAt":"2024-01-21T12:00:04.408Z","commentators":[{"avatarUrl":"/avatars/73b36acc7734d31784b9311cd5e143f2.svg","fullname":"sayed ismail","name":"sayedismail","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false}],"url":"/posts/abidlabs/986151192658340","totalUniqueImpressions":23,"numComments":2},"comment":{"_id":"65ad074414e6582c30bb6b8e","raw":"very cool!","html":"very cool!
\n"}},{"time":"2024-01-21T11:59:07.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"898169785250170","content":[{"type":"text","value":"I've prepared a Google Colab notebook which allows you to play with interpolating between different people using IP-Adapter SDXL Face-ID Plus. ","raw":"I've prepared a Google Colab notebook which allows you to play with interpolating between different people using IP-Adapter SDXL Face-ID Plus. "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"code_fence","code":"\n#Prepare a list t of num_of_results values between 0 and 1\nt_space = torch.linspace(0, 1, num_of_results)\nfor t in tqdm(t_space):\n mix_factor = t.item()\n # interpolate between the two face images \n image = (image1 * (1 - mix_factor) + image2 * mix_factor).astype(np.uint8)\n # interpolate between the two face embedding \n faceid_embeds = torch.lerp(faceid_embeds1, faceid_embeds2, t)\n #generate interpolated result\n images = ip_model.generate(prompt=prompt, negative_prompt=negative_prompt, face_image=image, faceid_embeds=faceid_embeds, shortcut=v2, num_samples=2, scale=scale, s_scale=s_scale, guidance_scale=guidance_scale, width=width, height=height, num_inference_steps=steps, seed=seed)\n ","raw":"```\n\n#Prepare a list t of num_of_results values between 0 and 1\nt_space = torch.linspace(0, 1, num_of_results)\nfor t in tqdm(t_space):\n mix_factor = t.item()\n # interpolate between the two face images \n image = (image1 * (1 - mix_factor) + image2 * mix_factor).astype(np.uint8)\n # interpolate between the two face embedding \n faceid_embeds = torch.lerp(faceid_embeds1, faceid_embeds2, t)\n #generate interpolated result\n images = ip_model.generate(prompt=prompt, negative_prompt=negative_prompt, face_image=image, faceid_embeds=faceid_embeds, shortcut=v2, num_samples=2, scale=scale, s_scale=s_scale, guidance_scale=guidance_scale, width=width, height=height, num_inference_steps=steps, seed=seed)\n \n```"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Link to notebook:","raw":"Link to notebook:"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"dataset","id":"Norod78/face_id_v2_test_code"},"url":"https://colab.research.google.com/#fileId=https%3A//huggingface.co/datasets/Norod78/face_id_v2_test_code/blob/main/norod78_faceid_sdxl_plus_v2_test.ipynb","raw":"https://colab.research.google.com/#fileId=https%3A//huggingface.co/datasets/Norod78/face_id_v2_test_code/blob/main/norod78_faceid_sdxl_plus_v2_test.ipynb"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Link to Face-ID Repo:","raw":"Link to Face-ID Repo:"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"model","id":"h94/IP-Adapter-FaceID"},"url":"https://huggingface.co/h94/IP-Adapter-FaceID","raw":"https://huggingface.co/h94/IP-Adapter-FaceID"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"Link to all sorts of generated examples (Use the file tab):","raw":"Link to all sorts of generated examples (Use the file tab):"},{"type":"new_line","raw":"\n"},{"type":"resource","resource":{"type":"dataset","id":"Norod78/face_id_v2_test_code"},"url":"https://huggingface.co/datasets/Norod78/face_id_v2_test_code/tree/main/sdxl_plus_v2_outputs","raw":"https://huggingface.co/datasets/Norod78/face_id_v2_test_code/tree/main/sdxl_plus_v2_outputs"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"}],"rawContent":"I've prepared a Google Colab notebook which allows you to play with interpolating between different people using IP-Adapter SDXL Face-ID Plus. \r\n\r\n```\r\n\r\n#Prepare a list t of num_of_results values between 0 and 1\r\nt_space = torch.linspace(0, 1, num_of_results)\r\nfor t in tqdm(t_space):\r\n mix_factor = t.item()\r\n # interpolate between the two face images \r\n image = (image1 * (1 - mix_factor) + image2 * mix_factor).astype(np.uint8)\r\n # interpolate between the two face embedding \r\n faceid_embeds = torch.lerp(faceid_embeds1, faceid_embeds2, t)\r\n #generate interpolated result\r\n images = ip_model.generate(prompt=prompt, negative_prompt=negative_prompt, face_image=image, faceid_embeds=faceid_embeds, shortcut=v2, num_samples=2, scale=scale, s_scale=s_scale, guidance_scale=guidance_scale, width=width, height=height, num_inference_steps=steps, seed=seed)\r\n \r\n```\r\n\r\nLink to notebook:\r\nhttps://colab.research.google.com/#fileId=https%3A//huggingface.co/datasets/Norod78/face_id_v2_test_code/blob/main/norod78_faceid_sdxl_plus_v2_test.ipynb\r\n\r\nLink to Face-ID Repo:\r\nhttps://huggingface.co/h94/IP-Adapter-FaceID\r\n\r\nLink to all sorts of generated examples (Use the file tab):\r\nhttps://huggingface.co/datasets/Norod78/face_id_v2_test_code/tree/main/sdxl_plus_v2_outputs\r\n\r\n","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1671292294864-5e00e3bdcbfd036a99df0da3.jpeg","fullname":"Doron Adler","name":"Norod78","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},"attachments":[{"type":"video","url":"https://cdn-uploads.huggingface.co/production/uploads/5e00e3bdcbfd036a99df0da3/6XVam08oGvebcHqDEkx4P.qt"},{"type":"video","url":"https://cdn-uploads.huggingface.co/production/uploads/5e00e3bdcbfd036a99df0da3/_V0wHMxqlqBY67NnQSqn5.mp4"}],"mentions":[],"reactions":[{"reaction":"🤯","users":["osanseviero","linoyts","clem","merve","HuggyMonkey","tonyassi"],"count":6},{"reaction":"❤️","users":["s3nh","clem","kevinstonge","samusenps","tonyassi"],"count":5},{"reaction":"👍","users":["abdd199719"],"count":1}],"publishedAt":"2024-01-21T09:24:07.000Z","updatedAt":"2024-03-19T08:54:34.138Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/61caeda441f9432649f03ab6/0UdRCrzIqhedZblgfpMBk.png","fullname":"s3nh","name":"s3nh","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1671292294864-5e00e3bdcbfd036a99df0da3.jpeg","fullname":"Doron Adler","name":"Norod78","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/638f308fc4444c6ca870b60a/Q11NK-8-JbiilJ-vk2LAR.png","fullname":"Linoy Tsaban","name":"linoyts","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/3ae664485d8df12c9b0e7b91e8d0457b.svg","fullname":"maxells","name":"maksim717","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/Norod78/898169785250170","totalUniqueImpressions":1153,"numComments":9},"comment":{"_id":"65ad070bb68db4f26eba869f","raw":"very cool!","html":"very cool!
\n"}},{"time":"2024-01-18T16:13:06.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"488383924115005","content":[{"type":"text","value":"Most upvoted papers of 2023 on HF. What do you think are going to be the most prominent research topics in AI for 2024 (also, don't forget to add your papers to the hub this year!). ","raw":"Most upvoted papers of 2023 on HF. What do you think are going to be the most prominent research topics in AI for 2024 (also, don't forget to add your papers to the hub this year!). "},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"From: ","raw":"From: "},{"type":"resource","resource":{"type":"space","id":"hysts/daily-papers"},"url":"https://huggingface.co/spaces/hysts/daily-papers","raw":"https://huggingface.co/spaces/hysts/daily-papers"}],"rawContent":"Most upvoted papers of 2023 on HF. What do you think are going to be the most prominent research topics in AI for 2024 (also, don't forget to add your papers to the hub this year!). \r\n\r\nFrom: https://huggingface.co/spaces/hysts/daily-papers","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5e67bdd61009063689407479/Ado86z8i--4EmuQgNAXfN.png"}],"mentions":[],"reactions":[{"reaction":"❤️","users":["pierrci","osanseviero","vwxyzjn","mvaloatto","linoyts","gsarti","victor","samusenps","julien-c","Dlbk","yukiarimo","mlabonne","MaziyarPanahi","Jofthomas","dvilasuero","snarktank","hysts","mfarajtabar","avinash02"],"count":19}],"publishedAt":"2024-01-18T16:13:06.000Z","updatedAt":"2024-01-22T13:49:29.159Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/o28fcV8uFf7mXtB3Asws0.jpeg","fullname":"Ryan","name":"snarktank","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/488383924115005","totalUniqueImpressions":81,"numComments":1}},{"time":"2024-01-16T12:43:15.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"533874509800797","content":[{"type":"text","value":"Is synthetic data the future of AI? 🔥🔥🔥","raw":"Is synthetic data the future of AI? 🔥🔥🔥"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"mention","user":"HugoLaurencon","raw":"@HugoLaurencon"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"Leyo","raw":"@Leyo"},{"type":"text","value":" & ","raw":" & "},{"type":"mention","user":"VictorSanh","raw":"@VictorSanh"},{"type":"text","value":" are introducing ","raw":" are introducing "},{"type":"resource","resource":{"type":"dataset","id":"HuggingFaceM4/WebSight"},"url":"https://huggingface.co/datasets/HuggingFaceM4/WebSight","raw":"https://huggingface.co/datasets/HuggingFaceM4/WebSight"},{"type":"text","value":" , a multimodal dataset featuring 823,000 pairs of synthetically generated HTML/CSS codes along with screenshots of the corresponding rendered websites to train GPT4-V-like models 🌐💻","raw":" , a multimodal dataset featuring 823,000 pairs of synthetically generated HTML/CSS codes along with screenshots of the corresponding rendered websites to train GPT4-V-like models 🌐💻"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"While crafting their upcoming foundation vision language model, they faced the challenge of converting website screenshots into usable HTML/CSS codes. Most VLMs suck at this and there was no public dataset available for this specific task, so they decided to create their own.","raw":"While crafting their upcoming foundation vision language model, they faced the challenge of converting website screenshots into usable HTML/CSS codes. Most VLMs suck at this and there was no public dataset available for this specific task, so they decided to create their own."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"They prompted existing LLMs to generate 823k HTML/CSS codes of very simple websites. Through supervised fine-tuning of a vision language model on WebSight, they were able to generate the code to reproduce a website component, given a screenshot.","raw":"They prompted existing LLMs to generate 823k HTML/CSS codes of very simple websites. Through supervised fine-tuning of a vision language model on WebSight, they were able to generate the code to reproduce a website component, given a screenshot."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"You can explore the dataset here: ","raw":"You can explore the dataset here: "},{"type":"resource","resource":{"type":"dataset","id":"HuggingFaceM4/WebSight"},"url":"https://huggingface.co/datasets/HuggingFaceM4/WebSight","raw":"https://huggingface.co/datasets/HuggingFaceM4/WebSight"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"What do you think?","raw":"What do you think?"}],"rawContent":"Is synthetic data the future of AI? 🔥🔥🔥\r\n\r\n@HugoLaurencon @Leyo & @VictorSanh are introducing https://huggingface.co/datasets/HuggingFaceM4/WebSight , a multimodal dataset featuring 823,000 pairs of synthetically generated HTML/CSS codes along with screenshots of the corresponding rendered websites to train GPT4-V-like models 🌐💻\r\n\r\nWhile crafting their upcoming foundation vision language model, they faced the challenge of converting website screenshots into usable HTML/CSS codes. Most VLMs suck at this and there was no public dataset available for this specific task, so they decided to create their own.\r\n\r\nThey prompted existing LLMs to generate 823k HTML/CSS codes of very simple websites. Through supervised fine-tuning of a vision language model on WebSight, they were able to generate the code to reproduce a website component, given a screenshot.\r\n\r\nYou can explore the dataset here: https://huggingface.co/datasets/HuggingFaceM4/WebSight\r\n\r\nWhat do you think?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1635201569275-noauth.jpeg","fullname":"Hugo Laurençon","name":"HugoLaurencon","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1652185658647-6244866a456803e9500d0f6a.jpeg","fullname":"Leo Tronchon","name":"Leyo","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1619623771844-5ecea265968f6028e0559fa5.jpeg","fullname":"Victor Sanh","name":"VictorSanh","type":"user","isPro":true,"isHf":false,"isMod":false}],"reactions":[{"reaction":"🤗","users":["Leyo","vishaal27","lhoestq","AdinaY","merve","Hotsnow","s3nh","florentgbelidji","HugoLaurencon","cabralski","ronzyponz","mishig","HugoDzz","victor","chansung","kramp","ameerazam08","wjbmattingly","mcpotato","taufiqdp","osanseviero","phanes","Ali-C137","duxans","aliabd","taesiri","pierrci","lvwerra","VictorSanh","Triyansha","mrfakename","dvilasuero","Srulikbd","julien-c","samusenps","Chinar","PeepDaSlan9","nic221","DEKUBE","LouMa","Mithilss","rohanbalkondekar","SaylorTwift","peterschmidt85","yihaopeng","mkemka","radames","faust01","lunarflu","vztay","iblub","zeynel","rreed"],"count":53},{"reaction":"👍","users":["s3nh","lhoestq","AdinaY","BramVanroy","merve","cabralski","davanstrien","Pwlot","mishig","HansKlepitko","HugoDzz","bstraehle","Sylvestre","neovalle","ameerazam08","wjbmattingly","osanseviero","aliabd","sungkim11","clementperrot","cyberandy","dvilasuero","julien-c","samusenps","mrdbourke","PeepDaSlan9","Frantz103","LouMa","Kernel","MoritzLaurer","yihaopeng","lunarflu","Kathakali","zeynel","z11h","rreed","freddyaboulton"],"count":37},{"reaction":"❤️","users":["happymanda","cabralski","mishig","ayut","HugoDzz","clefourrier","ameerazam08","osanseviero","aliabd","dvilasuero","julien-c","samusenps","PeepDaSlan9","DEKUBE","rohanbalkondekar","MaziyarPanahi","yihaopeng","lunarflu","rreed","jeffboudier","sbrandeis"],"count":21},{"reaction":"🤯","users":["taufiqdp","cabralski","mishig","Wauplin","holymonkey47","aliabd","doxiy","dreamworks2050","Srulikbd","julien-c","lunarflu","binga"],"count":12},{"reaction":"🤝","users":["Sharathhebbar24","lunarflu"],"count":2}],"publishedAt":"2024-01-16T12:40:13.000Z","updatedAt":"2024-01-17T10:51:53.167Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1676700305651-63f0698e5c2ceb16fc745be8.png","fullname":"amanda besemer","name":"happymanda","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1679186088116-63fa68b0d4349b157a15cffe.jpeg","fullname":"Val","name":"vdeva","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1627505688463-60107b385ac3e86b3ea4fc34.jpeg","fullname":"Daniel van Strien","name":"davanstrien","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/61caeda441f9432649f03ab6/0UdRCrzIqhedZblgfpMBk.png","fullname":"s3nh","name":"s3nh","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64aac16fd4a402e8dce11ebe/W640vNvHPWwwG_u4TsRo0.png","fullname":"Jorge Vallego","name":"neovalle","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/635964636a61954080850e1d/0bfExuDTrHTtm8c-40cDM.png","fullname":"William Lamkin","name":"phanes","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1619623771844-5ecea265968f6028e0559fa5.jpeg","fullname":"Victor Sanh","name":"VictorSanh","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64aea8ff67511bd3d965697b/Jxn52EmDF5RApJh8antxn.jpeg","fullname":"Feynman Innovations","name":"ajibawa-2023","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/NQtzmrDdbG0H8qkZvRyGk.jpeg","fullname":"Julien Chaumond","name":"julien-c","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/c96bb76b227f884114a040e9062bb005.svg","fullname":"Aaron Berdanier","name":"berdaniera","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/533874509800797","totalUniqueImpressions":23,"numComments":12},"comment":{"_id":"65a679e3bd0c4a98e9cf0f8a","raw":"🔥🔥🔥","html":"🔥🔥🔥
\n"}},{"time":"2024-01-16T12:40:13.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post","socialPost":{"slug":"533874509800797","content":[{"type":"text","value":"Is synthetic data the future of AI? 🔥🔥🔥","raw":"Is synthetic data the future of AI? 🔥🔥🔥"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"mention","user":"HugoLaurencon","raw":"@HugoLaurencon"},{"type":"text","value":" ","raw":" "},{"type":"mention","user":"Leyo","raw":"@Leyo"},{"type":"text","value":" & ","raw":" & "},{"type":"mention","user":"VictorSanh","raw":"@VictorSanh"},{"type":"text","value":" are introducing ","raw":" are introducing "},{"type":"resource","resource":{"type":"dataset","id":"HuggingFaceM4/WebSight"},"url":"https://huggingface.co/datasets/HuggingFaceM4/WebSight","raw":"https://huggingface.co/datasets/HuggingFaceM4/WebSight"},{"type":"text","value":" , a multimodal dataset featuring 823,000 pairs of synthetically generated HTML/CSS codes along with screenshots of the corresponding rendered websites to train GPT4-V-like models 🌐💻","raw":" , a multimodal dataset featuring 823,000 pairs of synthetically generated HTML/CSS codes along with screenshots of the corresponding rendered websites to train GPT4-V-like models 🌐💻"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"While crafting their upcoming foundation vision language model, they faced the challenge of converting website screenshots into usable HTML/CSS codes. Most VLMs suck at this and there was no public dataset available for this specific task, so they decided to create their own.","raw":"While crafting their upcoming foundation vision language model, they faced the challenge of converting website screenshots into usable HTML/CSS codes. Most VLMs suck at this and there was no public dataset available for this specific task, so they decided to create their own."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"They prompted existing LLMs to generate 823k HTML/CSS codes of very simple websites. Through supervised fine-tuning of a vision language model on WebSight, they were able to generate the code to reproduce a website component, given a screenshot.","raw":"They prompted existing LLMs to generate 823k HTML/CSS codes of very simple websites. Through supervised fine-tuning of a vision language model on WebSight, they were able to generate the code to reproduce a website component, given a screenshot."},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"You can explore the dataset here: ","raw":"You can explore the dataset here: "},{"type":"resource","resource":{"type":"dataset","id":"HuggingFaceM4/WebSight"},"url":"https://huggingface.co/datasets/HuggingFaceM4/WebSight","raw":"https://huggingface.co/datasets/HuggingFaceM4/WebSight"},{"type":"new_line","raw":"\n"},{"type":"new_line","raw":"\n"},{"type":"text","value":"What do you think?","raw":"What do you think?"}],"rawContent":"Is synthetic data the future of AI? 🔥🔥🔥\r\n\r\n@HugoLaurencon @Leyo & @VictorSanh are introducing https://huggingface.co/datasets/HuggingFaceM4/WebSight , a multimodal dataset featuring 823,000 pairs of synthetically generated HTML/CSS codes along with screenshots of the corresponding rendered websites to train GPT4-V-like models 🌐💻\r\n\r\nWhile crafting their upcoming foundation vision language model, they faced the challenge of converting website screenshots into usable HTML/CSS codes. Most VLMs suck at this and there was no public dataset available for this specific task, so they decided to create their own.\r\n\r\nThey prompted existing LLMs to generate 823k HTML/CSS codes of very simple websites. Through supervised fine-tuning of a vision language model on WebSight, they were able to generate the code to reproduce a website component, given a screenshot.\r\n\r\nYou can explore the dataset here: https://huggingface.co/datasets/HuggingFaceM4/WebSight\r\n\r\nWhat do you think?","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1635201569275-noauth.jpeg","fullname":"Hugo Laurençon","name":"HugoLaurencon","type":"user","isPro":false,"isHf":true,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1652185658647-6244866a456803e9500d0f6a.jpeg","fullname":"Leo Tronchon","name":"Leyo","type":"user","isPro":false,"isHf":false,"isMod":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1619623771844-5ecea265968f6028e0559fa5.jpeg","fullname":"Victor Sanh","name":"VictorSanh","type":"user","isPro":true,"isHf":false,"isMod":false}],"reactions":[{"reaction":"🤗","users":["Leyo","vishaal27","lhoestq","AdinaY","merve","Hotsnow","s3nh","florentgbelidji","HugoLaurencon","cabralski","ronzyponz","mishig","HugoDzz","victor","chansung","kramp","ameerazam08","wjbmattingly","mcpotato","taufiqdp","osanseviero","phanes","Ali-C137","duxans","aliabd","taesiri","pierrci","lvwerra","VictorSanh","Triyansha","mrfakename","dvilasuero","Srulikbd","julien-c","samusenps","Chinar","PeepDaSlan9","nic221","DEKUBE","LouMa","Mithilss","rohanbalkondekar","SaylorTwift","peterschmidt85","yihaopeng","mkemka","radames","faust01","lunarflu","vztay","iblub","zeynel","rreed"],"count":53},{"reaction":"👍","users":["s3nh","lhoestq","AdinaY","BramVanroy","merve","cabralski","davanstrien","Pwlot","mishig","HansKlepitko","HugoDzz","bstraehle","Sylvestre","neovalle","ameerazam08","wjbmattingly","osanseviero","aliabd","sungkim11","clementperrot","cyberandy","dvilasuero","julien-c","samusenps","mrdbourke","PeepDaSlan9","Frantz103","LouMa","Kernel","MoritzLaurer","yihaopeng","lunarflu","Kathakali","zeynel","z11h","rreed","freddyaboulton"],"count":37},{"reaction":"❤️","users":["happymanda","cabralski","mishig","ayut","HugoDzz","clefourrier","ameerazam08","osanseviero","aliabd","dvilasuero","julien-c","samusenps","PeepDaSlan9","DEKUBE","rohanbalkondekar","MaziyarPanahi","yihaopeng","lunarflu","rreed","jeffboudier","sbrandeis"],"count":21},{"reaction":"🤯","users":["taufiqdp","cabralski","mishig","Wauplin","holymonkey47","aliabd","doxiy","dreamworks2050","Srulikbd","julien-c","lunarflu","binga"],"count":12},{"reaction":"🤝","users":["Sharathhebbar24","lunarflu"],"count":2}],"publishedAt":"2024-01-16T12:40:13.000Z","updatedAt":"2024-01-17T10:51:53.167Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1676700305651-63f0698e5c2ceb16fc745be8.png","fullname":"amanda besemer","name":"happymanda","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1679186088116-63fa68b0d4349b157a15cffe.jpeg","fullname":"Val","name":"vdeva","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1627505688463-60107b385ac3e86b3ea4fc34.jpeg","fullname":"Daniel van Strien","name":"davanstrien","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/61caeda441f9432649f03ab6/0UdRCrzIqhedZblgfpMBk.png","fullname":"s3nh","name":"s3nh","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64aac16fd4a402e8dce11ebe/W640vNvHPWwwG_u4TsRo0.png","fullname":"Jorge Vallego","name":"neovalle","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/635964636a61954080850e1d/0bfExuDTrHTtm8c-40cDM.png","fullname":"William Lamkin","name":"phanes","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1619623771844-5ecea265968f6028e0559fa5.jpeg","fullname":"Victor Sanh","name":"VictorSanh","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64aea8ff67511bd3d965697b/Jxn52EmDF5RApJh8antxn.jpeg","fullname":"Feynman Innovations","name":"ajibawa-2023","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/NQtzmrDdbG0H8qkZvRyGk.jpeg","fullname":"Julien Chaumond","name":"julien-c","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/c96bb76b227f884114a040e9062bb005.svg","fullname":"Aaron Berdanier","name":"berdaniera","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/533874509800797","totalUniqueImpressions":23,"numComments":12}},{"time":"2024-01-16T12:34:03.000Z","user":"clem","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","type":"social-post-comment","socialPost":{"slug":"372730116768493","content":[{"type":"text","value":"Happy to announce, brand new, open-source Hugging Face Competitions platform 🚀 Now, create a machine learning competition for your friends, colleagues or the world for FREE* and host it on Hugging Face: the AI community building the future. Creating a competition requires only two steps: ","raw":"Happy to announce, brand new, open-source Hugging Face Competitions platform 🚀 Now, create a machine learning competition for your friends, colleagues or the world for FREE* and host it on Hugging Face: the AI community building the future. Creating a competition requires only two steps: "},{"type":"inline_code","code":"pip install competitions","raw":"`pip install competitions`"},{"type":"text","value":", then run ","raw":", then run "},{"type":"inline_code","code":"competitions create","raw":"`competitions create`"},{"type":"text","value":" and create competition by answering a few questions 💥 Checkout the github repo: ","raw":" and create competition by answering a few questions 💥 Checkout the github repo: "},{"type":"link","href":"https://github.com/huggingface/competitions","raw":"https://github.com/huggingface/competitions"},{"type":"text","value":" and docs: ","raw":" and docs: "},{"type":"link","href":"https://hf.co/docs/competitions","raw":"https://hf.co/docs/competitions"}],"rawContent":"Happy to announce, brand new, open-source Hugging Face Competitions platform 🚀 Now, create a machine learning competition for your friends, colleagues or the world for FREE* and host it on Hugging Face: the AI community building the future. Creating a competition requires only two steps: `pip install competitions`, then run `competitions create` and create competition by answering a few questions 💥 Checkout the github repo: https://github.com/huggingface/competitions and docs: https://hf.co/docs/competitions","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1604427578240-noauth.png","fullname":"Abhishek Thakur","name":"abhishek","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[{"type":"image","url":"https://cdn-uploads.huggingface.co/production/uploads/5fa19f4ba13e063b8b2b5e11/fWArXA3y_x5sxHz7_zYTE.png"}],"mentions":[],"reactions":[{"reaction":"❤️","users":["clem","merve","robmarkcole","linoyts","kurianbenoy","victor","Ali-C137","abhishek","AdinaY","osanseviero","Tonic","gsarti","nezubn","VictorSanh","julien-c","mrdbourke","MaziyarPanahi","samusenps","Smooke","d0rj","dillfrescott","Ammok","ecevinoth","PinakHolder"],"count":24},{"reaction":"👍","users":["happymanda","ThetaPhiPsi","chansung","victor","AdinaY","osanseviero","doxiy","julien-c","kramp","Smooke","Solshine","dillfrescott","Ammok"],"count":13},{"reaction":"🤗","users":["mc0ps","Ali-C137","pacoid","osanseviero","julien-c","TuringsSolutions","Smooke","mrfakename","Solshine","dillfrescott","IndiraMandalapu"],"count":11},{"reaction":"🤯","users":["clem","dreamworks2050","julien-c","dillfrescott"],"count":4}],"publishedAt":"2024-01-16T12:14:56.000Z","updatedAt":"2024-01-23T04:04:15.932Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1676700305651-63f0698e5c2ceb16fc745be8.png","fullname":"amanda besemer","name":"happymanda","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1604427578240-noauth.png","fullname":"Abhishek Thakur","name":"abhishek","type":"user","isPro":false,"isHf":true,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62a3bb1cd0d8c2c2169f0b88/eT2TS0IlQbZtz-F_zHLz9.jpeg","fullname":"Joseph Pollack","name":"Tonic","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/eXiGSHeBcKbFAQIhjiQ8r.jpeg","fullname":"Caleb DeLeeuw","name":"Solshine","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/abhishek/372730116768493","totalUniqueImpressions":993,"numComments":6},"comment":{"_id":"65a677bb5c58475cf9bc9c16","raw":"very cool!","html":"very cool!
\n"}}],"activityType":"post","blogPosts":[],"canReadDatabase":false,"canManageEntities":false,"canImpersonate":false,"canManageBilling":false,"canPost":false,"communityScore":129,"collections":[{"slug":"clem/favorite-demos-650cec2a72afb1e60e77242b","title":"Favorite demos","description":"","lastUpdated":"2023-09-22T01:22:27.862Z","owner":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"items":[{"_id":"650cec4229ff71d4d6c14e5a","position":1,"type":"space","author":"stabilityai","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/643feeb67bc3fbde1385cc25/7vmYr2XwVcPtkLzac_jxQ.png","fullname":"Stability AI","name":"stabilityai","type":"org","isHf":false,"isMod":false,"isEnterprise":true},"colorFrom":"yellow","colorTo":"gray","createdAt":"2022-08-04T09:41:12.000Z","emoji":"🔥","id":"stabilityai/stable-diffusion","lastModified":"2023-09-15T16:33:07.000Z","likes":10587,"pinned":true,"private":false,"repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":"cpu-upgrade","requested":"cpu-upgrade"},"storage":null,"gcTimeout":null,"replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"stabilityai-stable-diffusion.hf.space","isCustom":false,"stage":"READY"}],"sha":"7da827c2024da3963cbaca23d3fcc3d0370c56db"},"title":"Stable Diffusion 2-1","isLikedByUser":false},{"_id":"650cec4c9169ea733171f47c","position":2,"type":"space","author":"dalle-mini","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1628725656410-5f061f3c5d08220171a0ae35.png","fullname":"DALLE mini","name":"dalle-mini","type":"org","isHf":false,"isMod":false,"isEnterprise":false},"colorFrom":"yellow","colorTo":"green","createdAt":"2022-04-07T00:55:48.000Z","emoji":"🥑","id":"dalle-mini/dalle-mini","lastModified":"2023-02-08T06:32:28.000Z","likes":5484,"pinned":true,"private":false,"repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":null,"requested":null},"storage":null,"replicas":{"requested":1,"current":1}},"title":"DALL·E mini","isLikedByUser":false},{"_id":"650cec53fb7a5108874aa87c","position":3,"type":"space","author":"jbilcke-hf","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/2RK8J_YSNAK2ob8XZH7w2.jpeg","fullname":"Julian Bilcke","name":"jbilcke-hf","type":"user","isPro":false,"isHf":true,"isMod":false},"colorFrom":"red","colorTo":"yellow","createdAt":"2023-08-25T14:46:47.000Z","emoji":"👩🎨","id":"jbilcke-hf/ai-comic-factory","lastModified":"2024-08-07T15:54:46.000Z","likes":7716,"pinned":true,"private":false,"repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":"cpu-upgrade","requested":"cpu-upgrade"},"storage":null,"gcTimeout":172800,"replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"aicomicfactory.app","isCustom":true,"stage":"READY"},{"domain":"jbilcke-hf-ai-comic-factory.hf.space","isCustom":false,"stage":"READY"}],"sha":"1763a89c85b47fc38bc12c7df516259ca47b166e"},"shortDescription":"Create your own AI comic with a single prompt","title":"AI Comic Factory","isLikedByUser":false}],"position":1,"theme":"orange","private":false,"shareUrl":"https://huggingface.co/collections/clem/favorite-demos-650cec2a72afb1e60e77242b","upvotes":2,"isUpvotedByUser":false},{"slug":"clem/most-influential-papers-in-ai-6556a182c7c44ce359f47817","title":"Most influential papers in AI","description":"","lastUpdated":"2023-11-16T23:16:41.552Z","owner":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"items":[{"_id":"6556a1893541e7603457c835","position":0,"type":"paper","id":"1706.03762","title":"Attention Is All You Need","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/1706.03762.png","upvotes":41,"publishedAt":"2017-06-12T17:57:34.000Z","isUpvotedByUser":false},{"_id":"6556a1c238754a6b38f2ad56","position":1,"type":"paper","id":"1810.04805","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language\n Understanding","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/1810.04805.png","upvotes":14,"publishedAt":"2018-10-11T00:50:01.000Z","isUpvotedByUser":false},{"_id":"6556a1fd6539c9a5a5306bc0","position":2,"type":"paper","id":"1801.06146","title":"Universal Language Model Fine-tuning for Text Classification","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/1801.06146.png","upvotes":6,"publishedAt":"2018-01-18T17:54:52.000Z","isUpvotedByUser":false},{"_id":"6556a2828616d8d740bb101c","position":3,"type":"paper","id":"2005.14165","title":"Language Models are Few-Shot Learners","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2005.14165.png","upvotes":11,"publishedAt":"2020-05-28T17:29:03.000Z","isUpvotedByUser":false}],"position":1,"theme":"pink","private":false,"shareUrl":"https://huggingface.co/collections/clem/most-influential-papers-in-ai-6556a182c7c44ce359f47817","upvotes":27,"isUpvotedByUser":false},{"slug":"clem/read-papers-6556a31147c93d37a87af193","title":"read papers","description":"This is a collection of some papers I've read in the past few months","lastUpdated":"2023-11-21T18:28:01.540Z","owner":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"items":[{"_id":"6556a37885d43542fa1834d3","position":0,"type":"paper","id":"2311.05640","title":"FinGPT: Large Generative Models for a Small Language","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2311.05640.png","upvotes":26,"publishedAt":"2023-11-03T08:05:04.000Z","isUpvotedByUser":false},{"_id":"6556a37d07bf46c11a8be558","position":1,"type":"paper","id":"2311.05556","title":"LCM-LoRA: A Universal Stable-Diffusion Acceleration Module","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2311.05556.png","upvotes":79,"publishedAt":"2023-11-09T18:04:15.000Z","isUpvotedByUser":false},{"_id":"6556a382c7c44ce359f4d421","position":2,"type":"paper","id":"2106.10207","title":"Distributed Deep Learning in Open Collaborations","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2106.10207.png","upvotes":1,"publishedAt":"2021-06-18T16:23:13.000Z","isUpvotedByUser":false},{"_id":"6556a38526100eef4a35a7cf","position":3,"type":"paper","id":"2109.02846","title":"Datasets: A Community Library for Natural Language Processing","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2109.02846.png","upvotes":8,"publishedAt":"2021-09-07T03:59:22.000Z","isUpvotedByUser":false}],"position":2,"theme":"blue","private":false,"shareUrl":"https://huggingface.co/collections/clem/read-papers-6556a31147c93d37a87af193","upvotes":46,"isUpvotedByUser":false}],"datasets":[{"author":"clem","downloads":0,"gated":false,"id":"clem/prompts","lastModified":"2023-09-22T01:19:50.000Z","private":false,"repoType":"dataset","likes":2,"isLikedByUser":false},{"author":"clem","downloads":3,"gated":false,"id":"clem/autonlp-data-french_word_detection","lastModified":"2021-09-14T09:45:38.000Z","private":false,"repoType":"dataset","likes":1,"isLikedByUser":false}],"hasMoreActivities":false,"models":[{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":0,"gated":false,"id":"clem/ai-french-prime-minister","inference":"library-not-detected","lastModified":"2024-07-02T12:17:37.000Z","likes":18,"private":false,"repoType":"model","isLikedByUser":false},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":0,"gated":false,"id":"clem/gemini","inference":"library-not-detected","lastModified":"2023-12-06T15:20:59.000Z","likes":91,"private":false,"repoType":"model","isLikedByUser":false},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":0,"gated":false,"id":"clem/test","inference":"library-not-detected","lastModified":"2023-07-27T22:06:58.000Z","likes":0,"private":false,"repoType":"model","isLikedByUser":false},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":11,"gated":false,"id":"clem/maxdekdt","inference":"not-popular-enough","lastModified":"2023-01-05T20:11:04.000Z","likes":3,"pipeline_tag":"text-to-image","private":false,"repoType":"model","isLikedByUser":false},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":1,"gated":false,"id":"clem/friedeberg","inference":"not-popular-enough","lastModified":"2022-12-24T18:21:14.000Z","likes":0,"pipeline_tag":"text-to-image","private":false,"repoType":"model","isLikedByUser":false},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":8,"gated":false,"id":"clem/autonlp-test3-2101787","inference":"not-popular-enough","lastModified":"2021-06-29T04:32:06.000Z","likes":0,"pipeline_tag":"text-classification","private":false,"repoType":"model","isLikedByUser":false,"widgetOutputUrls":[]},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":7,"gated":false,"id":"clem/autonlp-test3-2101782","inference":"not-popular-enough","lastModified":"2021-06-29T04:19:34.000Z","likes":0,"pipeline_tag":"text-classification","private":false,"repoType":"model","isLikedByUser":false,"widgetOutputUrls":[]},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":6,"gated":false,"id":"clem/autonlp-test3-2101779","inference":"not-popular-enough","lastModified":"2021-06-29T04:15:35.000Z","likes":0,"pipeline_tag":"text-classification","private":false,"repoType":"model","isLikedByUser":false,"widgetOutputUrls":[]},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"downloads":0,"gated":false,"id":"clem/bert_portuguese","inference":"library-not-detected","lastModified":"2020-12-18T19:35:08.000Z","likes":1,"private":false,"repoType":"model","isLikedByUser":false}],"numberLikes":966,"papers":[{"id":"2109.02846","title":"Datasets: A Community Library for Natural Language Processing","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2109.02846.png","upvotes":8,"publishedAt":"2021-09-07T03:59:22.000Z","isUpvotedByUser":false},{"id":"1910.03771","title":"HuggingFace's Transformers: State-of-the-art Natural Language Processing","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/1910.03771.png","upvotes":16,"publishedAt":"2019-10-09T03:23:22.000Z","isUpvotedByUser":false},{"id":"1901.08149","title":"TransferTransfo: A Transfer Learning Approach for Neural Network Based\n Conversational Agents","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/1901.08149.png","upvotes":3,"publishedAt":"2019-01-23T22:08:01.000Z","isUpvotedByUser":false},{"id":"1805.05758","title":"Continuous Learning in a Hierarchical Multiscale Neural Network","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/1805.05758.png","upvotes":1,"publishedAt":"2018-05-15T13:37:33.000Z","isUpvotedByUser":false},{"id":"1803.10631","title":"Meta-Learning a Dynamical Language Model","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/1803.10631.png","upvotes":0,"publishedAt":"2018-03-28T14:08:12.000Z","isUpvotedByUser":false}],"posts":[{"slug":"530713412165384","content":[{"type":"text","value":"This is the week of small AI language models!","raw":"This is the week of small AI language models!"}],"rawContent":"This is the week of small AI language models!","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"👍","users":["ZeroWw","rumbleFTW","ZachZimm","osanseviero","GPT007","chenduo","jpacifico","thesven","AtAndDev","Fredithefish"],"count":10},{"reaction":"🤝","users":["Bharath182004","YaTharThShaRma999","rumbleFTW","osanseviero","GPT007","jessicagab","AtAndDev","louisbrulenaudet","Fredithefish"],"count":9},{"reaction":"🔥","users":["jeffboudier","AtAndDev"],"count":2}],"publishedAt":"2024-07-18T15:44:06.000Z","updatedAt":"2024-07-20T15:36:06.401Z","commentators":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65d883893a52cd9bcd8ab7cf/tRsCJlHNZo1D02kBTmfy9.jpeg","fullname":"leroy Samuel Dyer","name":"LeroyDyer","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/641caf6c043963b1c0a27256/CD7ktICDsldVJlpiND5kl.png","fullname":"PseudoTerminal X","name":"ptx0","type":"user","isPro":true,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/630f3e4002ce39336c411048/FXJON7b-aRUiH0_V2uRsi.jpeg","fullname":"alkinun","name":"AtAndDev","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false},{"avatarUrl":"/avatars/54483699273ac58a4a6fe1fa4aab65fe.svg","fullname":"Robert Sinclair","name":"ZeroWw","type":"user","isPro":false,"isHf":false,"isMod":false,"isFollowing":false}],"url":"/posts/clem/530713412165384","totalUniqueImpressions":2463,"numComments":4},{"slug":"215969412331682","content":[{"type":"text","value":"5,000 new repos (models, datasets, spaces) are created EVERY DAY on HF now. The community is amazing!","raw":"5,000 new repos (models, datasets, spaces) are created EVERY DAY on HF now. The community is amazing!"}],"rawContent":"5,000 new repos (models, datasets, spaces) are created EVERY DAY on HF now. The community is amazing!","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false,"isFollowing":false},"attachments":[],"mentions":[],"reactions":[{"reaction":"❤️","users":["umuthopeyildirim","osanseviero","victor","tail-call","paulml","jeremy-london","brunatrevelin","Clausss","ZeroWw","xi0v","yuriachermann","merterbak","nroggendorff","sted97","Jaromir","not-lain","netynet","Heresynetwork","SixOpen","tarob0ba","3thn","digiplay","prithivMLmods","GPT007","Nymbo"],"count":25},{"reaction":"🚀","users":["victor","ngxson","Clausss","ZeroWw","xi0v","nold","yuriachermann","nroggendorff","InferenceIllusionist","sted97","danielus","not-lain","netynet","Heresynetwork","moock","3thn","kramp","GPT007","nampdn-ai"],"count":19},{"reaction":"🤗","users":["digiplay","louisbrulenaudet","GPT007","emirhanbilgic"],"count":4},{"reaction":"👍","users":["digiplay","GPT007"],"count":2},{"reaction":"🤝","users":["digiplay","GPT007"],"count":2},{"reaction":"😎","users":["GPT007"],"count":1}],"publishedAt":"2024-07-01T12:40:49.000Z","updatedAt":"2024-07-01T12:40:49.355Z","commentators":[],"url":"/posts/clem/215969412331682","totalUniqueImpressions":5690,"numComments":0}],"totalPosts":18,"spaces":[{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"colorFrom":"yellow","colorTo":"pink","createdAt":"2023-01-08T15:08:16.000Z","emoji":"🔥","id":"clem/comparing-captioning-models","lastModified":"2023-01-08T15:08:17.000Z","likes":4,"pinned":false,"private":false,"repoType":"space","runtime":{"stage":"RUNTIME_ERROR","hardware":{"current":null,"requested":"cpu-basic"},"storage":null,"gcTimeout":86400,"errorMessage":"Launch timed out, workload was not healthy after 30 min","replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"clem-comparing-captioning-models.hf.space","isCustom":false,"stage":"READY"}]},"title":"Comparing Captioning Models","isLikedByUser":false,"originSpace":{"name":"nielsr/comparing-captioning-models","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1608042047613-5f1158120c833276f61f1a84.jpeg","fullname":"Niels Rogge","name":"nielsr","type":"user","isPro":false,"isHf":true,"isMod":false}}},{"author":"clem","authorData":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","fullname":"Clem 🤗","name":"clem","type":"user","isPro":true,"isHf":true,"isMod":false},"colorFrom":"blue","colorTo":"gray","createdAt":"2022-12-22T18:37:57.000Z","emoji":"📈","id":"clem/Image_Face_Upscale_Restoration-GFPGAN","lastModified":"2022-12-22T18:37:57.000Z","likes":130,"pinned":false,"private":false,"repoType":"space","runtime":{"stage":"RUNTIME_ERROR","hardware":{"current":null,"requested":"cpu-basic"},"storage":null,"gcTimeout":86400,"errorMessage":" 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0curl: (6) Could not resolve host: huggingface.co\nWarning: Problem : timeout. Will retry in 1 seconds. 5 retries left.\n\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0curl: (6) Could not resolve host: huggingface.co\nWarning: Problem : timeout. Will retry in 1 seconds. 4 retries left.\n\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0curl: (6) Could not resolve host: huggingface.co\nWarning: Problem : timeout. Will retry in 1 seconds. 3 retries left.\n\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0curl: (6) Could not resolve host: huggingface.co\nWarning: Problem : timeout. Will retry in 1 seconds. 2 retries left.\n\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0curl: (6) Could not resolve host: huggingface.co\nWarning: Problem : timeout. Will retry in 1 seconds. 1 retries left.\n\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0\r 0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0curl: (6) Could not resolve host: huggingface.co\n","replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"clem-image-face-upscale-restoration-gfpgan.hf.space","isCustom":false,"stage":"READY"}]},"title":"Image Face Upscale Restoration-GFPGAN","isLikedByUser":false,"originSpace":{"name":"nightfury/Image_Face_Upscale_Restoration-GFPGAN","author":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1664442251110-63329fde2217725f3adfff27.png","fullname":"spartan","name":"nightfury","type":"user","isPro":false,"isHf":false,"isMod":false}}}],"u":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583857146757-5e67bdd61009063689407479.jpeg","isPro":true,"fullname":"Clem 🤗","user":"clem","orgs":[{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1583856921041-5dd96eb166059660ed1ee413.png","fullname":"Hugging Face","name":"huggingface","userRole":"write","type":"org","isHf":true},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1595976246100-5ef393fd518622264685b0cf.png","fullname":"Objective Function","name":"ObjectiveFunction","userRole":"admin","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1599731665502-5e67bdd61009063689407479.jpeg","fullname":"Pied Piper","name":"piedpiper","userRole":"admin","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674034531134-61e6a54836fa261c76dc3760.png","fullname":"Society & Ethics","name":"society-ethics","userRole":"write","type":"org","isHf":false},{"avatarUrl":"https://www.gravatar.com/avatar/0cbc6611f5540bd0809a388dc95a615b?d=retro&size=100","fullname":"Organization","name":"Test","userRole":"admin","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f17f0a0925b9863e28ad517/NXI_YNqaf9ZvhVczj0kpz.png","fullname":"Text Generation Inference","name":"text-generation-inference","userRole":"write","type":"org","isHf":false},{"avatarUrl":"https://www.gravatar.com/avatar/4ce78c5773a0225a1bba3a167ee4c2ea?d=retro&size=100","fullname":"testifly","name":"testifly","userRole":"admin","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1647529507227-5f10cb46636b661bdc42318b.png","fullname":"HugGAN Community","name":"huggan","userRole":"write","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1649681117572-6141a88b3a0ec78603c9e784.png","fullname":"Hugging Face Fellows","name":"hugging-fellows","userRole":"write","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/608b8bb39d7c9519b4adae19/uSNsnGAAMM4-z2jYf8sqZ.png","fullname":"Gradio-Blocks-Party","name":"Gradio-Blocks","userRole":"contributor","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1653062669432-60741a2e69a66931a0273f0c.png","fullname":"HuggingFaceM4","name":"HuggingFaceM4","userRole":"admin","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1652969609470-61ac8f8a00d01045fca0ad2f.png","fullname":"Workshop June 13 Classroom","name":"Classroom-workshop","userRole":"admin","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60f1abe7544c2adfd699860c/2lt9ePTp9MaZ2RNaDw6Ql.jpeg","fullname":"Open-Source AI Meetup","name":"SFEvent","userRole":"admin","type":"org","isHf":false},{"avatarUrl":"https://www.gravatar.com/avatar/bd6f32283561374ab3e454f334600863?d=retro&size=100","fullname":"Hugging Face Smol Cluster","name":"HFSmolCluster","userRole":"write","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6032802e1f993496bc14d9e3/3s4vnB6_xPaPPWcZkF6jT.jpeg","fullname":"huggingPartyParis","name":"HuggingPartyParis","userRole":"read","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/61b85ce86eb1f2c5e6233736/YaCgid8n_VBFISGh1Eity.png","fullname":"Unofficial Mistral Community","name":"mistral-community","userRole":"contributor","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63691c3eda9b693c2730b2a2/WoOIHdJahrAnLo1wpyVjc.png","fullname":"Journalists on Hugging Face","name":"JournalistsonHF","userRole":"write","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6304c06eeb6d777a838eab63/qkILUu9lckQxgxfxOn2xm.png","fullname":"Major TOM","name":"Major-TOM","userRole":"read","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/623c830997ddced06d78699b/3qTjC7d3YFCJTwpxd2noq.png","fullname":"MLX Community","name":"mlx-community","userRole":"contributor","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/644206b566a62c605c98dbe5/Da4Zg541W3iEg5CajZ5P_.png","fullname":"Miami AI Hub","name":"miamiaihub","userRole":"write","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f17f0a0925b9863e28ad517/nxmdd6m86cxu55UZBlQeg.jpeg","fullname":"Social Post Explorers","name":"social-post-explorers","userRole":"admin","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60a551a34ecc5d054c8ad93e/zj4Vyk5keZrNRfy1wWR4D.png","fullname":"Paris AI Running Club","name":"paris-ai-running-club","userRole":"read","type":"org","isHf":false},{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6459fa0f5b3111fbe83286e1/TCv1NK707PxgaE5XsXWOx.png","fullname":"Hugging Face for Legal","name":"HFforLegal","userRole":"contributor","type":"org","isHf":false}],"signup":{"github":"clmnt","homepage":"http://huggingface.co","twitter":"clementdelangue","details":"multi-modal, time-series, biology and chemistry"},"isHf":true,"isMod":false,"type":"user"},"upvotes":309,"repoFilterModels":{"sortKey":"modified"},"repoFilterDatasets":{"sortKey":"modified"},"repoFilterSpaces":{"sortKey":"modified"},"numFollowers":1469,"numFollowing":930,"isFollowing":false,"isFollower":false,"sampleFollowers":[{"user":"ChrisLer","fullname":"Chris Ler","_id":"648d885bef7d519f2639bf9a","type":"user","isPro":false,"avatarUrl":"/avatars/a6360ddc674d6818e9f551aec4720e16.svg"},{"user":"florencejiang","fullname":"Florence Jiang","_id":"60ec80058362d78848a8be00","type":"user","isPro":false,"avatarUrl":"/avatars/b1ce73d25eeb791ea2c44561aced52ed.svg"},{"user":"nalasadi","fullname":"Noor Alasadi","_id":"64ae392a4beffa272d021c2a","type":"user","isPro":false,"avatarUrl":"/avatars/ac10fa5dac78d3b87c5108fb9f81cd57.svg"},{"user":"juancopi81","fullname":"Juan Carlos Piñeros","_id":"61e17815f28a6640312005db","type":"user","isPro":false,"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1644587669520-61e17815f28a6640312005db.png"}],"isWatching":false,"hardwareItems":[{"sku":["Apple Silicon","-","Apple M2 Max"],"mem":32,"num":1}]}">Clem 🤗 PRO
clem
AI & ML interests
multi-modal, time-series, biology and chemistry
Organizations
clem's activity
very cool!
Beautiful team work!
posted
an
update
about 1 month ago
Post
2463
This is the week of small AI language models!
- 4 replies
posted
an
update
about 2 months ago
Post
5690
5,000 new repos (models, datasets, spaces) are created EVERY DAY on HF now. The community is amazing!
replied to
louisbrulenaudet's
post
about 2 months ago
very cool! feel free to create the HF for Legal org and share about it and we can amplify!
posted
an
update
2 months ago
Post
3619
Who said you couldn't build a big business based on open-source AI? Congrats Mistral team: https://huggingface.co/mistralai
Beautiful work!
omg would be sick!
posted
an
update
3 months ago
Post
1525
I would pick
@ylecun
over
@elonmuskceo
every single day of the week.
Despite getting much less $$, recognition & visibility than entrepreneurs, the scientists who publish their groundbreaking research openly are the cornerstone of technological progress & massively contribute to making the world a better place!
Despite getting much less $$, recognition & visibility than entrepreneurs, the scientists who publish their groundbreaking research openly are the cornerstone of technological progress & massively contribute to making the world a better place!
- 1 reply
replied to
singhsidhukuldeep's
post
3 months ago
very cool! cc @clefourrier
replied to
their
post
3 months ago
replied to
their
post
3 months ago
any info on when it's going to be released though?
replied to
their
post
3 months ago
do you have a link?
posted
an
update
3 months ago
Post
1560
What are you excited about from Google I/O?
- 9 replies
congrats!
replied to
HeshamHaroon's
post
3 months ago
very cool, thanks for sharing!
replied to
singhsidhukuldeep's
post
3 months ago
Interesting update! They can open-source GPT4 now haha
replied to
danielhanchen's
post
4 months ago
congratulations! well deserved!
gotta catch them all!
you should create an org on HF for it
posted
an
update
4 months ago
Post
2516
Great in-depth Llama-3 tests from
@wolfram
, of the models from Meta of course but also
@MaziyarPanahi
@emozilla
@turboderp
: https://huggingface.co/blog/wolfram/llm-comparison-test-llama-3
Spotted by @jack-kumar
Spotted by @jack-kumar
- 2 replies
replied to
their
post
4 months ago
posted
an
update
4 months ago
Post
2895
Already almost 1,000 llama3 model variations have been shared publicly on HF (many more in private use at companies): https://huggingface.co/models?p=5&sort=trending&search=llama3.
Everyone should fine-tune their own models for their use-cases, languages, industry, infra constraints,...
10,000 llama3 variants by the end of next week?
Everyone should fine-tune their own models for their use-cases, languages, industry, infra constraints,...
10,000 llama3 variants by the end of next week?
- 4 replies
replied to
visheratin's
post
4 months ago
Thank you! You should tweet it mentioning @elonmuskceo !
posted
an
update
4 months ago
Post
2671
We noticed that all the open-source models and datasets from https://huggingface.co/WizardLM in their personal Hugging Face account & in the Microsoft Hugging Face organization (https://huggingface.co/microsoft) have been made private by the author, which will lead some demos to fail (these models were collectively downloaded over a hundred thousand times a month).
This is the explanation that @WizardLM communicated a few hours ago: https://huggingface.co/posts/WizardLM/329547800484476#661e0d17bca1a6038b60503e
We apologize for the inconvenience & are trying to get in touch with the author & Microsoft in order to try to find a good resolution for community members. Let us know if you have any questions!
This is the explanation that @WizardLM communicated a few hours ago: https://huggingface.co/posts/WizardLM/329547800484476#661e0d17bca1a6038b60503e
We apologize for the inconvenience & are trying to get in touch with the author & Microsoft in order to try to find a good resolution for community members. Let us know if you have any questions!
- 1 reply
posted
an
update
4 months ago
Post
2427
Fun dataset added last week by
@esind
from https://huggingface.co/Anthropic to compare persuasiveness between AI and human outputs:
Anthropic/persuasion
Anthropic/persuasion
- 2 replies
posted
an
update
5 months ago
Post
2518
Introducing
gretelai/synthetic_text_to_sql by https://huggingface.co/gretelai
It stands as the largest and most diverse synthetic Text-to-SQL dataset available to-date.
The dataset includes:
- 105,851 records partitioned into 100,000 train and 5,851 test records
~23M total tokens, including ~12M SQL tokens
- Coverage across 100 distinct domains/verticals
- Comprehensive array of SQL tasks: data definition, retrieval, manipulation, analytics & reporting
- Wide range of SQL complexity levels, including subqueries, single joins, multiple joins, aggregations, window functions, set operations
- Database context, including table and view create statements
- Natural language explanations of what the SQL query is doing
- Contextual tags to optimize model training
Blogpost: https://gretel.ai/blog/synthetic-text-to-sql-dataset
Dataset: gretelai/synthetic_text_to_sql
It stands as the largest and most diverse synthetic Text-to-SQL dataset available to-date.
The dataset includes:
- 105,851 records partitioned into 100,000 train and 5,851 test records
~23M total tokens, including ~12M SQL tokens
- Coverage across 100 distinct domains/verticals
- Comprehensive array of SQL tasks: data definition, retrieval, manipulation, analytics & reporting
- Wide range of SQL complexity levels, including subqueries, single joins, multiple joins, aggregations, window functions, set operations
- Database context, including table and view create statements
- Natural language explanations of what the SQL query is doing
- Contextual tags to optimize model training
Blogpost: https://gretel.ai/blog/synthetic-text-to-sql-dataset
Dataset: gretelai/synthetic_text_to_sql
- 1 reply
Thanks for sharing!
Welcome @josefprusa !
posted
an
update
6 months ago
Post
Terribly excited about open-source + on-device AI these days! Great to see
@qualcomm
release 80+ models optimized and curated for their devices and chips on HF: https://huggingface.co/qualcomm
- 1 reply
replied to
dvilasuero's
post
6 months ago
Unpopular opinion: this is the most impactful release of the day (because open)!
replied to
DmitryRyumin's
post
6 months ago
would be cool to have some integration with the HF hub
replied to
trisfromgoogle's
post
6 months ago
This is awesome!
very cool!
This comment has been hidden
🇫🇷🇫🇷🇫🇷
replied to
dvilasuero's
post
7 months ago
Very cool!
replied to
clefourrier's
post
7 months ago
very useful! This is the link to the leaderboard btw: https://huggingface.co/spaces/PatronusAI/enterprise_scenarios_leaderboard
very cool!
posted
an
update
7 months ago
Post
So impressed with the speed and accuracy of
vikhyatk/moondream1 by
@vikhyatk (especially the last answer 😝😝😝).
Open multi-modal models have gone a long way!
Model: vikhyatk/moondream1
@vikhyatk (especially the last answer 😝😝😝).
Open multi-modal models have gone a long way!
Model: vikhyatk/moondream1
- 1 reply
posted
an
update
7 months ago
Post
With the Google announcement last week, I think we're now officially the only AI startup out there who has commercial collaborations with all the major cloud providers (AWS, GCP, Azure) and hardware providers (Nvidia, AMD, Intel, Qualcomm,...), making our vision of being the independent and agnostic platform for all AI builders truer than ever!
Let's go!
Let's go!
posted
an
update
7 months ago
Post
In 2024, we're expanding from open weights to open EVERYTHING (datasets, training scripts,...).
Excited to see this dataset release in French by @Pclanglais @carbonbasedLLM @anastasiastasenko :
PleIAs/French-PD-Newspapers
"To give you an idea of the size, the full French Wikipedia is about 2 billon words. This is 40 times larger."
Excited to see this dataset release in French by @Pclanglais @carbonbasedLLM @anastasiastasenko :
PleIAs/French-PD-Newspapers
"To give you an idea of the size, the full French Wikipedia is about 2 billon words. This is 40 times larger."
Very cool!
posted
an
update
7 months ago
Post
Google + Hugging Face + Open-Source AI = 🔥🔥🔥
https://huggingface.co/blog/gcp-partnership
https://finance.yahoo.com/video/google-hugging-face-alliance-spur-173016882.html
https://www.theverge.com/2024/1/25/24050445/google-cloud-hugging-face-ai-developer-access
https://www.bloomberg.com/news/articles/2024-01-25/google-to-team-up-with-startup-hugging-face-to-host-ai-software
https://www.reuters.com/technology/google-cloud-partners-with-hugging-face-attract-ai-developers-2024-01-25/
https://huggingface.co/blog/gcp-partnership
https://finance.yahoo.com/video/google-hugging-face-alliance-spur-173016882.html
https://www.theverge.com/2024/1/25/24050445/google-cloud-hugging-face-ai-developer-access
https://www.bloomberg.com/news/articles/2024-01-25/google-to-team-up-with-startup-hugging-face-to-host-ai-software
https://www.reuters.com/technology/google-cloud-partners-with-hugging-face-attract-ai-developers-2024-01-25/
- 3 replies
posted
an
update
7 months ago
Post
Re-posting
@karpathy
's blogpost here because it's down on https://karpathy.github.io/2024/01/21/selfdriving-agi. What do you all think?
- 4 replies
very cool!
very cool!
posted
an
update
7 months ago
Post
Most upvoted papers of 2023 on HF. What do you think are going to be the most prominent research topics in AI for 2024 (also, don't forget to add your papers to the hub this year!).
From: hysts/daily-papers
From: hysts/daily-papers
- 1 reply
replied to
their
post
7 months ago
🔥🔥🔥
posted
an
update
7 months ago
Post
Is synthetic data the future of AI? 🔥🔥🔥
@HugoLaurencon @Leyo & @VictorSanh are introducing HuggingFaceM4/WebSight , a multimodal dataset featuring 823,000 pairs of synthetically generated HTML/CSS codes along with screenshots of the corresponding rendered websites to train GPT4-V-like models 🌐💻
While crafting their upcoming foundation vision language model, they faced the challenge of converting website screenshots into usable HTML/CSS codes. Most VLMs suck at this and there was no public dataset available for this specific task, so they decided to create their own.
They prompted existing LLMs to generate 823k HTML/CSS codes of very simple websites. Through supervised fine-tuning of a vision language model on WebSight, they were able to generate the code to reproduce a website component, given a screenshot.
You can explore the dataset here: HuggingFaceM4/WebSight
What do you think?
@HugoLaurencon @Leyo & @VictorSanh are introducing HuggingFaceM4/WebSight , a multimodal dataset featuring 823,000 pairs of synthetically generated HTML/CSS codes along with screenshots of the corresponding rendered websites to train GPT4-V-like models 🌐💻
While crafting their upcoming foundation vision language model, they faced the challenge of converting website screenshots into usable HTML/CSS codes. Most VLMs suck at this and there was no public dataset available for this specific task, so they decided to create their own.
They prompted existing LLMs to generate 823k HTML/CSS codes of very simple websites. Through supervised fine-tuning of a vision language model on WebSight, they were able to generate the code to reproduce a website component, given a screenshot.
You can explore the dataset here: HuggingFaceM4/WebSight
What do you think?
- 12 replies
very cool!