Isabelle Cudlitz
commited on
Commit
·
14cc589
1
Parent(s):
c27f40d
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2754,4 +2754,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2754 |
### Framework versions
|
| 2755 |
|
| 2756 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2757 |
- PEFT 0.6.2
|
|
|
|
| 2754 |
### Framework versions
|
| 2755 |
|
| 2756 |
|
| 2757 |
+
- PEFT 0.6.2
|
| 2758 |
+
## Training procedure
|
| 2759 |
+
|
| 2760 |
+
|
| 2761 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2762 |
+
- load_in_8bit: True
|
| 2763 |
+
- load_in_4bit: False
|
| 2764 |
+
- llm_int8_threshold: 6.0
|
| 2765 |
+
- llm_int8_skip_modules: None
|
| 2766 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2767 |
+
- llm_int8_has_fp16_weight: False
|
| 2768 |
+
- bnb_4bit_quant_type: fp4
|
| 2769 |
+
- bnb_4bit_use_double_quant: False
|
| 2770 |
+
- bnb_4bit_compute_dtype: float32
|
| 2771 |
+
|
| 2772 |
+
### Framework versions
|
| 2773 |
+
|
| 2774 |
+
|
| 2775 |
- 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:c0de171de08431c605f67f1df8624303fb6cc857b4bac2b52c85e753a89030f3
|
| 3 |
size 21020682
|