Isabelle Cudlitz
commited on
Commit
·
efd8eef
1
Parent(s):
b8d88f9
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2844,4 +2844,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2844 |
### Framework versions
|
| 2845 |
|
| 2846 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2847 |
- PEFT 0.6.2
|
|
|
|
| 2844 |
### Framework versions
|
| 2845 |
|
| 2846 |
|
| 2847 |
+
- PEFT 0.6.2
|
| 2848 |
+
## Training procedure
|
| 2849 |
+
|
| 2850 |
+
|
| 2851 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2852 |
+
- load_in_8bit: True
|
| 2853 |
+
- load_in_4bit: False
|
| 2854 |
+
- llm_int8_threshold: 6.0
|
| 2855 |
+
- llm_int8_skip_modules: None
|
| 2856 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2857 |
+
- llm_int8_has_fp16_weight: False
|
| 2858 |
+
- bnb_4bit_quant_type: fp4
|
| 2859 |
+
- bnb_4bit_use_double_quant: False
|
| 2860 |
+
- bnb_4bit_compute_dtype: float32
|
| 2861 |
+
|
| 2862 |
+
### Framework versions
|
| 2863 |
+
|
| 2864 |
+
|
| 2865 |
- 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:ce14f6c5b7ac22f3a3633560f3f1c0cc1f9ec2f2ce0f1e1f564170502963da30
|
| 3 |
size 21020682
|