OPEA
/

Safetensors
llama
3-bit
gptq
cicdatopea commited on
Commit
f27f93b
·
verified ·
1 Parent(s): 54cea49

revert the change

Browse files
Files changed (1) hide show
  1. README.md +1 -3
README.md CHANGED
@@ -122,9 +122,7 @@ lm-eval --model hf --model_args pretrained=OPEA/Llama-3.3-70B-Instruct-int3-sym-
122
  | boolq | 0.8865 | 0.8911 | |
123
  | arc_easy | 0.8624 | 0.8523 | 0.2643 |
124
  | arc_challenge | 0.6109 | 0.5956 | |
125
- | gsm8k(5shot) strict match? | 0.9083 | 0.9067 | |
126
-
127
- gsm8k data is not correct, probably flexible-extract
128
 
129
  ## Generate the model
130
 
 
122
  | boolq | 0.8865 | 0.8911 | |
123
  | arc_easy | 0.8624 | 0.8523 | 0.2643 |
124
  | arc_challenge | 0.6109 | 0.5956 | |
125
+ | gsm8k(5shot) strict match | 0.9083 | 0.9067 | |
 
 
126
 
127
  ## Generate the model
128