Isabelle Cudlitz commited on
Commit ·
5fb690c
1
Parent(s): d113f7c
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1206,4 +1206,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1206 |
### Framework versions
|
| 1207 |
|
| 1208 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1209 |
- PEFT 0.6.2
|
|
|
|
| 1206 |
### Framework versions
|
| 1207 |
|
| 1208 |
|
| 1209 |
+
- PEFT 0.6.2
|
| 1210 |
+
## Training procedure
|
| 1211 |
+
|
| 1212 |
+
|
| 1213 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1214 |
+
- load_in_8bit: False
|
| 1215 |
+
- load_in_4bit: True
|
| 1216 |
+
- llm_int8_threshold: 6.0
|
| 1217 |
+
- llm_int8_skip_modules: None
|
| 1218 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1219 |
+
- llm_int8_has_fp16_weight: False
|
| 1220 |
+
- bnb_4bit_quant_type: fp4
|
| 1221 |
+
- bnb_4bit_use_double_quant: True
|
| 1222 |
+
- bnb_4bit_compute_dtype: float32
|
| 1223 |
+
|
| 1224 |
+
### Framework versions
|
| 1225 |
+
|
| 1226 |
+
|
| 1227 |
- 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:fe2ca4d287c5b79e5970a460d2d31b33f9d6031b17cee649adf465177dd2ab99
|
| 3 |
size 21020682
|