Isabelle Cudlitz commited on
Commit ·
35d0c6b
1
Parent(s): b3332b2
Upload model
Browse files- README.md +19 -0
- adapter_model.safetensors +1 -1
README.md
CHANGED
|
@@ -2858,4 +2858,23 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2858 |
### Framework versions
|
| 2859 |
|
| 2860 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2861 |
- PEFT 0.6.2
|
|
|
|
| 2858 |
### Framework versions
|
| 2859 |
|
| 2860 |
|
| 2861 |
+
- PEFT 0.6.2
|
| 2862 |
+
## Training procedure
|
| 2863 |
+
|
| 2864 |
+
|
| 2865 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2866 |
+
- quant_method: bitsandbytes
|
| 2867 |
+
- load_in_8bit: True
|
| 2868 |
+
- load_in_4bit: False
|
| 2869 |
+
- llm_int8_threshold: 6.0
|
| 2870 |
+
- llm_int8_skip_modules: None
|
| 2871 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2872 |
+
- llm_int8_has_fp16_weight: False
|
| 2873 |
+
- bnb_4bit_quant_type: fp4
|
| 2874 |
+
- bnb_4bit_use_double_quant: False
|
| 2875 |
+
- bnb_4bit_compute_dtype: float32
|
| 2876 |
+
|
| 2877 |
+
### Framework versions
|
| 2878 |
+
|
| 2879 |
+
|
| 2880 |
- PEFT 0.6.2
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 20991608
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4caf2ce9f6c1f74d4efd4bb31ce6db781712e01b071bc2d8b3835ae5b54c4586
|
| 3 |
size 20991608
|