Hugging Face's logo

Tirendaz
/
llama-2-7b-chat-guanaco

PEFT
Safetensors
Model card Files
xet
 Community
Tirendaz commited on
Commit
3088e21
·
1 Parent(s): 051df21

Upload model

Browse files
Files changed (2) hide show
  1. README.md +2 -2
  2. adapter_model.safetensors +1 -1
README.md CHANGED
@@ -209,9 +209,9 @@ The following `bitsandbytes` quantization config was used during training:
209
  - llm_int8_skip_modules: None
210
  - llm_int8_enable_fp32_cpu_offload: False
211
  - llm_int8_has_fp16_weight: False
212
- - bnb_4bit_quant_type: nf4
213
  - bnb_4bit_use_double_quant: False
214
- - bnb_4bit_compute_dtype: bfloat16
215
 
216
  ### Framework versions
217
 
 
209
  - llm_int8_skip_modules: None
210
  - llm_int8_enable_fp32_cpu_offload: False
211
  - llm_int8_has_fp16_weight: False
212
+ - bnb_4bit_quant_type: fp4
213
  - bnb_4bit_use_double_quant: False
214
+ - bnb_4bit_compute_dtype: float32
215
 
216
  ### Framework versions
217
 
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7de6dfafc655e985366c6d6920234645d8e3db9d0b734d588cc272856e31c599
3
  size 134235048
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdf5f96806491fc204d7043a16221f2416ab7858aa63c8e7db8f3a98a01a439f
3
  size 134235048