Upload model
Browse files- README.md +3 -3
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -11,9 +11,9 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 11 |
- llm_int8_skip_modules: None
|
| 12 |
- llm_int8_enable_fp32_cpu_offload: False
|
| 13 |
- llm_int8_has_fp16_weight: False
|
| 14 |
-
- bnb_4bit_quant_type:
|
| 15 |
-
- bnb_4bit_use_double_quant:
|
| 16 |
-
- bnb_4bit_compute_dtype:
|
| 17 |
### Framework versions
|
| 18 |
|
| 19 |
|
|
|
|
| 11 |
- llm_int8_skip_modules: None
|
| 12 |
- llm_int8_enable_fp32_cpu_offload: False
|
| 13 |
- llm_int8_has_fp16_weight: False
|
| 14 |
+
- bnb_4bit_quant_type: nf4
|
| 15 |
+
- bnb_4bit_use_double_quant: True
|
| 16 |
+
- bnb_4bit_compute_dtype: bfloat16
|
| 17 |
### Framework versions
|
| 18 |
|
| 19 |
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 18898161
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48ddb6cd72566d00fe3025706171425a32e652cf00c05750a8665244cc838d3c
|
| 3 |
size 18898161
|