tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 12:35:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

ggml : remove q1_3 and q2_2

Browse Source 
* llama : remove the separate scale tensors of BitNet b1.58

They won't be needed, since the remaining ternary quant types have
built-in scales.
This commit is contained in:
Francis Couture-Harpin
2024-08-02 19:52:19 -04:00
parent 45719a2472
commit 04eec58112
12 changed files with 45 additions and 693 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/quantize/quantize.cpp
View File

@ -28,8 +28,6 @@ static const std::vector<struct quant_option> QUANT_OPTIONS = {
{ "IQ1_M", LLAMA_FTYPE_MOSTLY_IQ1_M, " 1.75 bpw quantization", },
{ "TQ1_0", LLAMA_FTYPE_MOSTLY_TQ1_0, " 1.69 bpw ternarization", },
{ "TQ2_0", LLAMA_FTYPE_MOSTLY_TQ2_0, " 2.06 bpw ternarization", },
{ "Q1_3", LLAMA_FTYPE_MOSTLY_Q1_3, " 1.63 bpw for BitNet b1.58", },
{ "Q2_2", LLAMA_FTYPE_MOSTLY_Q2_2, " 2.00 bpw for BitNet b1.58", },
{ "Q2_K", LLAMA_FTYPE_MOSTLY_Q2_K, " 2.96G, +3.5199 ppl @ Llama-3-8B", },
{ "Q2_K_S", LLAMA_FTYPE_MOSTLY_Q2_K_S, " 2.96G, +3.1836 ppl @ Llama-3-8B", },
{ "IQ3_XXS", LLAMA_FTYPE_MOSTLY_IQ3_XXS, " 3.06 bpw quantization", },