Isabelle Cudlitz
commited on
Commit
·
0b39a73
1
Parent(s):
690e9e2
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -3924,4 +3924,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 3924 |
### Framework versions
|
| 3925 |
|
| 3926 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3927 |
- PEFT 0.6.2
|
|
|
|
| 3924 |
### Framework versions
|
| 3925 |
|
| 3926 |
|
| 3927 |
+
- PEFT 0.6.2
|
| 3928 |
+
## Training procedure
|
| 3929 |
+
|
| 3930 |
+
|
| 3931 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 3932 |
+
- load_in_8bit: True
|
| 3933 |
+
- load_in_4bit: False
|
| 3934 |
+
- llm_int8_threshold: 6.0
|
| 3935 |
+
- llm_int8_skip_modules: None
|
| 3936 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 3937 |
+
- llm_int8_has_fp16_weight: False
|
| 3938 |
+
- bnb_4bit_quant_type: fp4
|
| 3939 |
+
- bnb_4bit_use_double_quant: False
|
| 3940 |
+
- bnb_4bit_compute_dtype: float32
|
| 3941 |
+
|
| 3942 |
+
### Framework versions
|
| 3943 |
+
|
| 3944 |
+
|
| 3945 |
- PEFT 0.6.2
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 21020682
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb1e0eabd31ba4ac435c35da841ba32ff2edb5ec4ad8cb857a625c9b21268e54
|
| 3 |
size 21020682
|