Hugging Face's logo

ISTA-DASLab
/
Meta-Llama-3-70B-AQLM-2Bit-1x16

Text Generation
Transformers
Safetensors
llama
facebook
meta
llama-3
conversational
text-generation-inference
Inference Endpoints
aqlm
Model card Files Community
SpiridonSunRotator commited on
Commit
ddb56ce
•
1 Parent(s): c5c2ca6

Updated metrics

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -16,5 +16,5 @@ For this quantization, we used 1 codebook of 16 bits.
16
  Results (measured with `lm_eval==4.0`):
17
  | Model | Quantization | MMLU (5-shot) | ArcC| ArcE| Hellaswag | Winogrande | PiQA | Model size, Gb |
18
  |------|------|-------|------|------|------|------|------|------|
19
- |meta-llama/Meta-Llama-3-70B | - | 0.7868 | 0.6596 | 0.8935 | 0.6921 | 0.8444 | 0.8090 | 141.2 |
20
- | | 1x16 | 0.7475| 0.6049 | 0.8468 | 0.6520 | 0.8248 | 0.7790 | 21.9 |
 
16
  Results (measured with `lm_eval==4.0`):
17
  | Model | Quantization | MMLU (5-shot) | ArcC| ArcE| Hellaswag | Winogrande | PiQA | Model size, Gb |
18
  |------|------|-------|------|------|------|------|------|------|
19
+ |meta-llama/Meta-Llama-3-70B | - | 0.7868 | 0.6041 | 0.8670 | 0.6634 | 0.8248 | 0.8090 | 141.2 |
20
+ | | 1x16 | 0.7486 | 0.5009 | 0.7896 | 0.6346 | 0.7954 | 0.7814 | 21.9 |