Isabelle Cudlitz
commited on
Commit
·
b2887b6
1
Parent(s):
ed06531
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -756,4 +756,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 756 |
### Framework versions
|
| 757 |
|
| 758 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 759 |
- PEFT 0.6.2
|
|
|
|
| 756 |
### Framework versions
|
| 757 |
|
| 758 |
|
| 759 |
+
- PEFT 0.6.2
|
| 760 |
+
## Training procedure
|
| 761 |
+
|
| 762 |
+
|
| 763 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 764 |
+
- load_in_8bit: False
|
| 765 |
+
- load_in_4bit: True
|
| 766 |
+
- llm_int8_threshold: 6.0
|
| 767 |
+
- llm_int8_skip_modules: None
|
| 768 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 769 |
+
- llm_int8_has_fp16_weight: False
|
| 770 |
+
- bnb_4bit_quant_type: fp4
|
| 771 |
+
- bnb_4bit_use_double_quant: True
|
| 772 |
+
- bnb_4bit_compute_dtype: float32
|
| 773 |
+
|
| 774 |
+
### Framework versions
|
| 775 |
+
|
| 776 |
+
|
| 777 |
- 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 sha256:04b4d122d7b81c1883c11ac419f7e18ad4dc826d5c2dcef4ba30fb8e7b838a7b
|
| 3 |
size 21020682
|