OPEA
/

Safetensors
llama
2-bit
intel/auto-round
cicdatopea commited on
Commit
49da73c
·
verified ·
1 Parent(s): 6616a74

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -1
README.md CHANGED
@@ -129,9 +129,10 @@ auto-round \
129
  --device 0 \
130
  --group_size 32 \
131
  --nsamples 1024 \
132
- --bits 4 \
133
  --iter 2000 \
134
  --disable_eval \
 
135
  --enable_norm_bias_tuning \
136
  --low_gpu_mem_usage \
137
  --format 'auto_gptq,auto_round' \
 
129
  --device 0 \
130
  --group_size 32 \
131
  --nsamples 1024 \
132
+ --bits 2 \
133
  --iter 2000 \
134
  --disable_eval \
135
+ --model_dtype "fp16" \
136
  --enable_norm_bias_tuning \
137
  --low_gpu_mem_usage \
138
  --format 'auto_gptq,auto_round' \