Isabelle Cudlitz
commited on
Commit
·
9c68675
1
Parent(s):
8302d48
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2790,4 +2790,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2790 |
### Framework versions
|
| 2791 |
|
| 2792 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2793 |
- PEFT 0.6.2
|
|
|
|
| 2790 |
### Framework versions
|
| 2791 |
|
| 2792 |
|
| 2793 |
+
- PEFT 0.6.2
|
| 2794 |
+
## Training procedure
|
| 2795 |
+
|
| 2796 |
+
|
| 2797 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2798 |
+
- load_in_8bit: True
|
| 2799 |
+
- load_in_4bit: False
|
| 2800 |
+
- llm_int8_threshold: 6.0
|
| 2801 |
+
- llm_int8_skip_modules: None
|
| 2802 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2803 |
+
- llm_int8_has_fp16_weight: False
|
| 2804 |
+
- bnb_4bit_quant_type: fp4
|
| 2805 |
+
- bnb_4bit_use_double_quant: False
|
| 2806 |
+
- bnb_4bit_compute_dtype: float32
|
| 2807 |
+
|
| 2808 |
+
### Framework versions
|
| 2809 |
+
|
| 2810 |
+
|
| 2811 |
- 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:c4edf7cece4e43aa2831dcf676ff18ec1bf127ff05debea9cdc0dfd7d6efcc9f
|
| 3 |
size 21020682
|