Isabelle Cudlitz
commited on
Commit
·
f217c84
1
Parent(s):
5b005fa
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -468,4 +468,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 468 |
### Framework versions
|
| 469 |
|
| 470 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 471 |
- PEFT 0.6.2
|
|
|
|
| 468 |
### Framework versions
|
| 469 |
|
| 470 |
|
| 471 |
+
- PEFT 0.6.2
|
| 472 |
+
## Training procedure
|
| 473 |
+
|
| 474 |
+
|
| 475 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 476 |
+
- load_in_8bit: False
|
| 477 |
+
- load_in_4bit: True
|
| 478 |
+
- llm_int8_threshold: 6.0
|
| 479 |
+
- llm_int8_skip_modules: None
|
| 480 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 481 |
+
- llm_int8_has_fp16_weight: False
|
| 482 |
+
- bnb_4bit_quant_type: fp4
|
| 483 |
+
- bnb_4bit_use_double_quant: True
|
| 484 |
+
- bnb_4bit_compute_dtype: float32
|
| 485 |
+
|
| 486 |
+
### Framework versions
|
| 487 |
+
|
| 488 |
+
|
| 489 |
- 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:b9d58f41ca29f8e0359fa1cbcea9ca8f327b147b779de9b011578ed6696484b1
|
| 3 |
size 21020682
|