Isabelle Cudlitz
commited on
Commit
·
55d98a5
1
Parent(s):
03d02e6
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -270,4 +270,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 270 |
### Framework versions
|
| 271 |
|
| 272 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 273 |
- PEFT 0.6.2
|
|
|
|
| 270 |
### Framework versions
|
| 271 |
|
| 272 |
|
| 273 |
+
- PEFT 0.6.2
|
| 274 |
+
## Training procedure
|
| 275 |
+
|
| 276 |
+
|
| 277 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 278 |
+
- load_in_8bit: False
|
| 279 |
+
- load_in_4bit: True
|
| 280 |
+
- llm_int8_threshold: 6.0
|
| 281 |
+
- llm_int8_skip_modules: None
|
| 282 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 283 |
+
- llm_int8_has_fp16_weight: False
|
| 284 |
+
- bnb_4bit_quant_type: fp4
|
| 285 |
+
- bnb_4bit_use_double_quant: True
|
| 286 |
+
- bnb_4bit_compute_dtype: float32
|
| 287 |
+
|
| 288 |
+
### Framework versions
|
| 289 |
+
|
| 290 |
+
|
| 291 |
- 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 sha256:3f76f59c13a7409c2bf2b7b4fac05cc23ce7c1051dcf6d51af662ca6753748b6
|
| 3 |
size 21020682
|