Isabelle Cudlitz
commited on
Commit
·
87978fd
1
Parent(s):
8016d37
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -576,4 +576,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 576 |
### Framework versions
|
| 577 |
|
| 578 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 579 |
- PEFT 0.6.2
|
|
|
|
| 576 |
### Framework versions
|
| 577 |
|
| 578 |
|
| 579 |
+
- PEFT 0.6.2
|
| 580 |
+
## Training procedure
|
| 581 |
+
|
| 582 |
+
|
| 583 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 584 |
+
- load_in_8bit: False
|
| 585 |
+
- load_in_4bit: True
|
| 586 |
+
- llm_int8_threshold: 6.0
|
| 587 |
+
- llm_int8_skip_modules: None
|
| 588 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 589 |
+
- llm_int8_has_fp16_weight: False
|
| 590 |
+
- bnb_4bit_quant_type: fp4
|
| 591 |
+
- bnb_4bit_use_double_quant: True
|
| 592 |
+
- bnb_4bit_compute_dtype: float32
|
| 593 |
+
|
| 594 |
+
### Framework versions
|
| 595 |
+
|
| 596 |
+
|
| 597 |
- 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 sha256:7ff099763d3002edbb5845790ace2d065bfb142b09f6855dfc5556713db3ae4b
|
| 3 |
size 21020682
|