Skip to content

Commit

Permalink
better js object
Browse files Browse the repository at this point in the history
  • Loading branch information
@mishig25
mishig25 committed Apr 10, 2024
1 parent fc27b4c commit 240f0df
Showing 1 changed file with 95 additions and 37 deletions.
132 changes: 95 additions & 37 deletions packages/gguf/src/quant_descriptions.ts
View file
Original file line number Diff line number Diff line change
@@ -1,40 +1,98 @@
import { GGMLQuantizationType } from "./types";

export const QUANT_DESCRIPTIONS: Record<GGMLQuantizationType, string> = {
[GGMLQuantizationType.F32]: "32-bit standard IEEE 754 single-precision floating-point number.", // src: https://en.wikipedia.org/wiki/Single-precision_floating-point_format
[GGMLQuantizationType.F16]: "16-bit standard IEEE 754 half-precision floating-point number.", // src: https://en.wikipedia.org/wiki/Half-precision_floating-point_format
[GGMLQuantizationType.Q4_0]:
"4-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale. Legacy quantization method (not used widely as of today)", // src: https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557654249
[GGMLQuantizationType.Q4_1]:
"4-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale + block_minimum. Legacy quantization method (not used widely as of today)", // src: https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557682290
[GGMLQuantizationType.Q5_0]:
"5-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale. Legacy quantization method (not used widely as of today)", // src: https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557654249
[GGMLQuantizationType.Q5_1]:
"5-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale + block_minimum. Legacy quantization method (not used widely as of today)", // src: https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557682290
[GGMLQuantizationType.Q8_0]:
"8-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale. Legacy quantization method (not used widely as of today)", // src: https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557654249
[GGMLQuantizationType.Q8_1]:
"8-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale + block_minimum. Legacy quantization method (not used widely as of today)", // src: https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557682290
[GGMLQuantizationType.Q2_K]: `2-bit quantization (q). Super-blocks with 16 blocks, each block has 16 weight. Weight formula: w = q * block_scale(4-bit) + block_min(4-bit), resulting in 2.5625 bits-per-weight.`, // src: https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305
[GGMLQuantizationType.Q3_K]: `3-bit quantization (q). Super-blocks with 16 blocks, each block has 16 weights. Weight formula: w = q * block_scale(6-bit), resulting. 3.4375 bits-per-weight`, // src: https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305
[GGMLQuantizationType.Q4_K]: `4-bit quantization (q). Super-blocks with 8 blocks, each block has 32 weights. Weight formula: w = q * block_scale(6-bit) + block_min(6-bit), resulting in 4.5 bits-per-weight.`, // src: https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305
[GGMLQuantizationType.Q5_K]: `5-bit quantization (q). Super-blocks with 8 blocks, each block has 32 weights. Weight formula: w = q * block_scale(6-bit) + block_min(6-bit), resulting in 5.5 bits-per-weight.`, // src: https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305
[GGMLQuantizationType.Q6_K]: `6-bit quantization (q). Super-blocks with 16 blocks, each block has 16 weights. Weight formula: w = q * block_scale(8-bit), resulting in 6.5625 bits-per-weight.`, // src: https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305
[GGMLQuantizationType.Q8_K]: `8-bit quantization (q). Each block has 256 weights. Only used for quantizing intermediate results. All 2-6 bit dot products are implemented for this quantization type. Weight formula: w = q * block_scale.`, // src: https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305
[GGMLQuantizationType.IQ2_XXS]:
"2-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 2.06 bits-per-weight.", // src: https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70
[GGMLQuantizationType.IQ2_XS]:
"2-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 2.31 bits-per-weight.", // src: https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70
[GGMLQuantizationType.IQ3_XXS]:
"3-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 3.06 bits-per-weight.", // src: https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70
[GGMLQuantizationType.IQ1_S]:
"1-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 1.56 bits-per-weight.", // src: https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70
[GGMLQuantizationType.IQ4_NL]:
"4-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix",
[GGMLQuantizationType.IQ3_S]:
"3-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 3.44 bits-per-weight.", // src: https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70
[GGMLQuantizationType.IQ2_S]:
"2-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 2.5 bits-per-weight.", // src: https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70
[GGMLQuantizationType.IQ4_XS]:
"4-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 4.25 bits-per-weight.", // src: https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70
export const QUANT_DESCRIPTIONS: Record<GGMLQuantizationType, { txt: string; src_url?: string }> = {
[GGMLQuantizationType.F32]: {
txt: "32-bit standard IEEE 754 single-precision floating-point number.",
src_url: "https://en.wikipedia.org/wiki/Single-precision_floating-point_format",
},
[GGMLQuantizationType.F16]: {
txt: "16-bit standard IEEE 754 half-precision floating-point number.",
src_url: "https://en.wikipedia.org/wiki/Half-precision_floating-point_format",
},
[GGMLQuantizationType.Q4_0]: {
txt: "4-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale. Legacy quantization method (not used widely as of today)",
src_url: "https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557654249",
},
[GGMLQuantizationType.Q4_1]: {
txt: "4-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale + block_minimum. Legacy quantization method (not used widely as of today)",
src_url: "https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557682290",
},
[GGMLQuantizationType.Q5_0]: {
txt: "5-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale. Legacy quantization method (not used widely as of today)",
src_url: "https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557654249",
},
[GGMLQuantizationType.Q5_1]: {
txt: "5-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale + block_minimum. Legacy quantization method (not used widely as of today)",
src_url: "https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557682290",
},
[GGMLQuantizationType.Q8_0]: {
txt: "8-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale. Legacy quantization method (not used widely as of today)",
src_url: "https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557654249",
},
[GGMLQuantizationType.Q8_1]: {
txt: "8-bit round-to-nearest quantization (q). Each block has 32 weights. Weight formula: w = q * block_scale + block_minimum. Legacy quantization method (not used widely as of today)",
src_url: "https://github.com/huggingface/huggingface.js/pull/615#discussion_r1557682290",
},
[GGMLQuantizationType.Q2_K]: {
txt: `2-bit quantization (q). Super-blocks with 16 blocks, each block has 16 weight. Weight formula: w = q * block_scale(4-bit) + block_min(4-bit), resulting in 2.5625 bits-per-weight.`,
src_url: "https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305",
},
[GGMLQuantizationType.Q3_K]: {
txt: `3-bit quantization (q). Super-blocks with 16 blocks, each block has 16 weights. Weight formula: w = q * block_scale(6-bit), resulting. 3.4375 bits-per-weight`,
src_url: "https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305",
},
[GGMLQuantizationType.Q4_K]: {
txt: `4-bit quantization (q). Super-blocks with 8 blocks, each block has 32 weights. Weight formula: w = q * block_scale(6-bit) + block_min(6-bit), resulting in 4.5 bits-per-weight.`,
src_url: "https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305",
},
[GGMLQuantizationType.Q5_K]: {
txt: `5-bit quantization (q). Super-blocks with 8 blocks, each block has 32 weights. Weight formula: w = q * block_scale(6-bit) + block_min(6-bit), resulting in 5.5 bits-per-weight.`,
src_url: "https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305",
},
[GGMLQuantizationType.Q6_K]: {
txt: `6-bit quantization (q). Super-blocks with 16 blocks, each block has 16 weights. Weight formula: w = q * block_scale(8-bit), resulting in 6.5625 bits-per-weight.`,
src_url: "https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305",
},
[GGMLQuantizationType.Q8_K]: {
txt: `8-bit quantization (q). Each block has 256 weights. Only used for quantizing intermediate results. All 2-6 bit dot products are implemented for this quantization type. Weight formula: w = q * block_scale.`,
src_url: "https://github.com/ggerganov/llama.cpp/pull/1684#issue-1739619305",
},
[GGMLQuantizationType.IQ2_XXS]: {
txt: "2-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 2.06 bits-per-weight.",
src_url:
"https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70",
},
[GGMLQuantizationType.IQ2_XS]: {
txt: "2-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 2.31 bits-per-weight.",
src_url:
"https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70",
},
[GGMLQuantizationType.IQ3_XXS]: {
txt: "3-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 3.06 bits-per-weight.",
src_url:
"https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70",
},
[GGMLQuantizationType.IQ1_S]: {
txt: "1-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 1.56 bits-per-weight.",
src_url:
"https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70",
},
[GGMLQuantizationType.IQ4_NL]: {
txt: "4-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix",
},
[GGMLQuantizationType.IQ3_S]: {
txt: "3-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 3.44 bits-per-weight.",
src_url:
"https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70",
},
[GGMLQuantizationType.IQ2_S]: {
txt: "2-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 2.5 bits-per-weight.",
src_url:
"https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70",
},
[GGMLQuantizationType.IQ4_XS]: {
txt: "4-bit quantization (q). Super-blocks with 256 weights. Weight w is obtained using super_block_scale & importance matrix, resulting in 4.25 bits-per-weight.",
src_url:
"https://huggingface.co/CISCai/OpenCodeInterpreter-DS-6.7B-SOTA-GGUF/blob/main/README.md?code=true#L59-L70",
},
};

0 comments on commit 240f0df

Please sign in to comment.