Isabelle Cudlitz
commited on
Commit
·
40aba56
1
Parent(s):
8b11026
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1134,4 +1134,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1134 |
### Framework versions
|
| 1135 |
|
| 1136 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1137 |
- PEFT 0.6.2
|
|
|
|
| 1134 |
### Framework versions
|
| 1135 |
|
| 1136 |
|
| 1137 |
+
- PEFT 0.6.2
|
| 1138 |
+
## Training procedure
|
| 1139 |
+
|
| 1140 |
+
|
| 1141 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1142 |
+
- load_in_8bit: False
|
| 1143 |
+
- load_in_4bit: True
|
| 1144 |
+
- llm_int8_threshold: 6.0
|
| 1145 |
+
- llm_int8_skip_modules: None
|
| 1146 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1147 |
+
- llm_int8_has_fp16_weight: False
|
| 1148 |
+
- bnb_4bit_quant_type: fp4
|
| 1149 |
+
- bnb_4bit_use_double_quant: True
|
| 1150 |
+
- bnb_4bit_compute_dtype: float32
|
| 1151 |
+
|
| 1152 |
+
### Framework versions
|
| 1153 |
+
|
| 1154 |
+
|
| 1155 |
- 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 sha256:78f07dee66e548c48ff8e0188af7d347fd4b852b989ca9ab9f85f2750968baf8
|
| 3 |
size 21020682
|