Isabelle Cudlitz
commited on
Commit
·
09d33e0
1
Parent(s):
8f31831
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -792,4 +792,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 792 |
### Framework versions
|
| 793 |
|
| 794 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 795 |
- PEFT 0.6.2
|
|
|
|
| 792 |
### Framework versions
|
| 793 |
|
| 794 |
|
| 795 |
+
- PEFT 0.6.2
|
| 796 |
+
## Training procedure
|
| 797 |
+
|
| 798 |
+
|
| 799 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 800 |
+
- load_in_8bit: False
|
| 801 |
+
- load_in_4bit: True
|
| 802 |
+
- llm_int8_threshold: 6.0
|
| 803 |
+
- llm_int8_skip_modules: None
|
| 804 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 805 |
+
- llm_int8_has_fp16_weight: False
|
| 806 |
+
- bnb_4bit_quant_type: fp4
|
| 807 |
+
- bnb_4bit_use_double_quant: True
|
| 808 |
+
- bnb_4bit_compute_dtype: float32
|
| 809 |
+
|
| 810 |
+
### Framework versions
|
| 811 |
+
|
| 812 |
+
|
| 813 |
- 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:e3980a136f13032e5b6193b4a056277841bb68c938029b1487119313d533de80
|
| 3 |
size 21020682
|