Isabelle Cudlitz
commited on
Commit
·
01efb9d
1
Parent(s):
8d2f676
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -3114,4 +3114,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 3114 |
### Framework versions
|
| 3115 |
|
| 3116 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3117 |
- PEFT 0.6.2
|
|
|
|
| 3114 |
### Framework versions
|
| 3115 |
|
| 3116 |
|
| 3117 |
+
- PEFT 0.6.2
|
| 3118 |
+
## Training procedure
|
| 3119 |
+
|
| 3120 |
+
|
| 3121 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 3122 |
+
- load_in_8bit: True
|
| 3123 |
+
- load_in_4bit: False
|
| 3124 |
+
- llm_int8_threshold: 6.0
|
| 3125 |
+
- llm_int8_skip_modules: None
|
| 3126 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 3127 |
+
- llm_int8_has_fp16_weight: False
|
| 3128 |
+
- bnb_4bit_quant_type: fp4
|
| 3129 |
+
- bnb_4bit_use_double_quant: False
|
| 3130 |
+
- bnb_4bit_compute_dtype: float32
|
| 3131 |
+
|
| 3132 |
+
### Framework versions
|
| 3133 |
+
|
| 3134 |
+
|
| 3135 |
- 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 sha256:13131186d0c0c791ca5bece7cbf8ad46d810e605ff0b7307840b7192ac3ec499
|
| 3 |
size 21020682
|