Hugging Face's logo

mlabonne
/
codellama-2-7b

PEFT
PyTorch
Safetensors
llama
code
Model card Files Community
5
mlabonne commited on
Commit
e069b7b
·
1 Parent(s): b418918

Training in progress, step 1

Browse files
Files changed (3) hide show
  1. README.md +14 -0
  2. adapter_model.bin +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -88,5 +88,19 @@ The following `bitsandbytes` quantization config was used during training:
88
  - bnb_4bit_compute_dtype: bfloat16
89
  ### Framework versions
90
 
 
91
 
92
  - PEFT 0.5.0.dev0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
88
  - bnb_4bit_compute_dtype: bfloat16
89
  ### Framework versions
90
 
91
+ - PEFT 0.5.0.dev0
92
 
93
  - PEFT 0.5.0.dev0
94
+ ## Training procedure
95
+
96
+
97
+ The following `bitsandbytes` quantization config was used during training:
98
+ - load_in_8bit: False
99
+ - load_in_4bit: True
100
+ - llm_int8_threshold: 6.0
101
+ - llm_int8_skip_modules: None
102
+ - llm_int8_enable_fp32_cpu_offload: False
103
+ - llm_int8_has_fp16_weight: False
104
+ - bnb_4bit_quant_type: nf4
105
+ - bnb_4bit_use_double_quant: True
106
+ - bnb_4bit_compute_dtype: bfloat16
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:841fc29473ed1c7f63bee59a1e2398e150ad497baebeefb8f4b3ed024d8d2ee8
3
  size 16822989
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c5d978ae570471e739194ae6816c741737fc494e9cf7c486163848aa3ef036d
3
  size 16822989
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:90310b602d7ff2f29528124a2c8db35c0a494185ed644a6df6c4dd431436ac61
3
  size 4027
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:702b6e523eaa62e4570fc7a98259c31a6f26a72ca2e6e661a521913e0ada81ea
3
  size 4027