Commit
·
b03629f
1
Parent(s):
a66f510
End of training
Browse files- README.md +19 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -255,4 +255,23 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 255 |
### Framework versions
|
| 256 |
|
| 257 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 258 |
- PEFT 0.6.2
|
|
|
|
| 255 |
### Framework versions
|
| 256 |
|
| 257 |
|
| 258 |
+
- PEFT 0.6.2
|
| 259 |
+
## Training procedure
|
| 260 |
+
|
| 261 |
+
|
| 262 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 263 |
+
- quant_method: bitsandbytes
|
| 264 |
+
- load_in_8bit: False
|
| 265 |
+
- load_in_4bit: True
|
| 266 |
+
- llm_int8_threshold: 6.0
|
| 267 |
+
- llm_int8_skip_modules: None
|
| 268 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 269 |
+
- llm_int8_has_fp16_weight: False
|
| 270 |
+
- bnb_4bit_quant_type: nf4
|
| 271 |
+
- bnb_4bit_use_double_quant: True
|
| 272 |
+
- bnb_4bit_compute_dtype: bfloat16
|
| 273 |
+
|
| 274 |
+
### Framework versions
|
| 275 |
+
|
| 276 |
+
|
| 277 |
- 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 67155338
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc99a2931389d45413aae2eef43baa7c084ecf943623540aa8edc951da1569ae
|
| 3 |
size 67155338
|