Isabelle Cudlitz
commited on
Commit
·
83f7a7a
1
Parent(s):
38f0bda
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -720,4 +720,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 720 |
### Framework versions
|
| 721 |
|
| 722 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 723 |
- PEFT 0.6.2
|
|
|
|
| 720 |
### Framework versions
|
| 721 |
|
| 722 |
|
| 723 |
+
- PEFT 0.6.2
|
| 724 |
+
## Training procedure
|
| 725 |
+
|
| 726 |
+
|
| 727 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 728 |
+
- load_in_8bit: False
|
| 729 |
+
- load_in_4bit: True
|
| 730 |
+
- llm_int8_threshold: 6.0
|
| 731 |
+
- llm_int8_skip_modules: None
|
| 732 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 733 |
+
- llm_int8_has_fp16_weight: False
|
| 734 |
+
- bnb_4bit_quant_type: fp4
|
| 735 |
+
- bnb_4bit_use_double_quant: True
|
| 736 |
+
- bnb_4bit_compute_dtype: float32
|
| 737 |
+
|
| 738 |
+
### Framework versions
|
| 739 |
+
|
| 740 |
+
|
| 741 |
- 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 sha256:876cb2832c92d1b9b17c88400fa35caca87d5fa00c9430a94d014f2f58ef3311
|
| 3 |
size 21020682
|