Isabelle Cudlitz
commited on
Commit
·
15ec4da
1
Parent(s):
eb7ab23
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1332,4 +1332,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1332 |
### Framework versions
|
| 1333 |
|
| 1334 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1335 |
- PEFT 0.6.2
|
|
|
|
| 1332 |
### Framework versions
|
| 1333 |
|
| 1334 |
|
| 1335 |
+
- PEFT 0.6.2
|
| 1336 |
+
## Training procedure
|
| 1337 |
+
|
| 1338 |
+
|
| 1339 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1340 |
+
- load_in_8bit: False
|
| 1341 |
+
- load_in_4bit: True
|
| 1342 |
+
- llm_int8_threshold: 6.0
|
| 1343 |
+
- llm_int8_skip_modules: None
|
| 1344 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1345 |
+
- llm_int8_has_fp16_weight: False
|
| 1346 |
+
- bnb_4bit_quant_type: fp4
|
| 1347 |
+
- bnb_4bit_use_double_quant: True
|
| 1348 |
+
- bnb_4bit_compute_dtype: float32
|
| 1349 |
+
|
| 1350 |
+
### Framework versions
|
| 1351 |
+
|
| 1352 |
+
|
| 1353 |
- 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:cc09084218a8f6ef9e08c770201b87ef3fe398b63ef70e8852cbdf763b194a43
|
| 3 |
size 21020682
|