Upload model
Browse files- README.md +19 -0
- adapter_model.safetensors +1 -1
README.md
CHANGED
|
@@ -78,4 +78,23 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 78 |
### Framework versions
|
| 79 |
|
| 80 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
- PEFT 0.6.2
|
|
|
|
| 78 |
### Framework versions
|
| 79 |
|
| 80 |
|
| 81 |
+
- PEFT 0.6.2
|
| 82 |
+
## Training procedure
|
| 83 |
+
|
| 84 |
+
|
| 85 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 86 |
+
- quant_method: bitsandbytes
|
| 87 |
+
- load_in_8bit: False
|
| 88 |
+
- load_in_4bit: True
|
| 89 |
+
- llm_int8_threshold: 6.0
|
| 90 |
+
- llm_int8_skip_modules: None
|
| 91 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 92 |
+
- llm_int8_has_fp16_weight: False
|
| 93 |
+
- bnb_4bit_quant_type: nf4
|
| 94 |
+
- bnb_4bit_use_double_quant: False
|
| 95 |
+
- bnb_4bit_compute_dtype: bfloat16
|
| 96 |
+
|
| 97 |
+
### Framework versions
|
| 98 |
+
|
| 99 |
+
|
| 100 |
- PEFT 0.6.2
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 75518144
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7aac51f19cf38ffb25e0a5397d86a69721f534c1d20417a8d7e2c364f435b444
|
| 3 |
size 75518144
|