Isabelle Cudlitz
commited on
Commit
·
8b11026
1
Parent(s):
998e274
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1116,4 +1116,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1116 |
### Framework versions
|
| 1117 |
|
| 1118 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1119 |
- PEFT 0.6.2
|
|
|
|
| 1116 |
### Framework versions
|
| 1117 |
|
| 1118 |
|
| 1119 |
+
- PEFT 0.6.2
|
| 1120 |
+
## Training procedure
|
| 1121 |
+
|
| 1122 |
+
|
| 1123 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1124 |
+
- load_in_8bit: False
|
| 1125 |
+
- load_in_4bit: True
|
| 1126 |
+
- llm_int8_threshold: 6.0
|
| 1127 |
+
- llm_int8_skip_modules: None
|
| 1128 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1129 |
+
- llm_int8_has_fp16_weight: False
|
| 1130 |
+
- bnb_4bit_quant_type: fp4
|
| 1131 |
+
- bnb_4bit_use_double_quant: True
|
| 1132 |
+
- bnb_4bit_compute_dtype: float32
|
| 1133 |
+
|
| 1134 |
+
### Framework versions
|
| 1135 |
+
|
| 1136 |
+
|
| 1137 |
- 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:ef3667c8c96825d1f16691d938772b870179d9179c7c2521cd3f0af76189a605
|
| 3 |
size 21020682
|