Isabelle Cudlitz
commited on
Commit
·
854a2ef
1
Parent(s):
4448e68
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2574,4 +2574,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2574 |
### Framework versions
|
| 2575 |
|
| 2576 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2577 |
- PEFT 0.6.2
|
|
|
|
| 2574 |
### Framework versions
|
| 2575 |
|
| 2576 |
|
| 2577 |
+
- PEFT 0.6.2
|
| 2578 |
+
## Training procedure
|
| 2579 |
+
|
| 2580 |
+
|
| 2581 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2582 |
+
- load_in_8bit: True
|
| 2583 |
+
- load_in_4bit: False
|
| 2584 |
+
- llm_int8_threshold: 6.0
|
| 2585 |
+
- llm_int8_skip_modules: None
|
| 2586 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2587 |
+
- llm_int8_has_fp16_weight: False
|
| 2588 |
+
- bnb_4bit_quant_type: fp4
|
| 2589 |
+
- bnb_4bit_use_double_quant: False
|
| 2590 |
+
- bnb_4bit_compute_dtype: float32
|
| 2591 |
+
|
| 2592 |
+
### Framework versions
|
| 2593 |
+
|
| 2594 |
+
|
| 2595 |
- 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:c28a7d9f6ae7468cb4eff21efd83a80e24a9ca144a9ce356984996c91f48f68d
|
| 3 |
size 21020682
|