Isabelle Cudlitz
commited on
Commit
·
21c4a6b
1
Parent(s):
4d40456
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -3618,4 +3618,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 3618 |
### Framework versions
|
| 3619 |
|
| 3620 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3621 |
- PEFT 0.6.2
|
|
|
|
| 3618 |
### Framework versions
|
| 3619 |
|
| 3620 |
|
| 3621 |
+
- PEFT 0.6.2
|
| 3622 |
+
## Training procedure
|
| 3623 |
+
|
| 3624 |
+
|
| 3625 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 3626 |
+
- load_in_8bit: True
|
| 3627 |
+
- load_in_4bit: False
|
| 3628 |
+
- llm_int8_threshold: 6.0
|
| 3629 |
+
- llm_int8_skip_modules: None
|
| 3630 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 3631 |
+
- llm_int8_has_fp16_weight: False
|
| 3632 |
+
- bnb_4bit_quant_type: fp4
|
| 3633 |
+
- bnb_4bit_use_double_quant: False
|
| 3634 |
+
- bnb_4bit_compute_dtype: float32
|
| 3635 |
+
|
| 3636 |
+
### Framework versions
|
| 3637 |
+
|
| 3638 |
+
|
| 3639 |
- 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:a5c16f28740ef79ae8f13ee477d73943fd3b6c3d3396acf44c718368ff4c47b3
|
| 3 |
size 21020682
|