Update README.md
Browse filesCurrently hugging-face automatically has assigned the instruct model as a "quantized" version of the base model, maybe as it sees some weights in fp8? it automatically did it.
The `base_model_relation: finetune` will force set it to a finetune from the base model
README.md
CHANGED
|
@@ -3,6 +3,7 @@ license: mit
|
|
| 3 |
library_name: transformers
|
| 4 |
base_model:
|
| 5 |
- deepseek-ai/DeepSeek-V3.2-Exp-Base
|
|
|
|
| 6 |
---
|
| 7 |
# DeepSeek-V3.2-Exp
|
| 8 |
|
|
|
|
| 3 |
library_name: transformers
|
| 4 |
base_model:
|
| 5 |
- deepseek-ai/DeepSeek-V3.2-Exp-Base
|
| 6 |
+
base_model_relation: finetune
|
| 7 |
---
|
| 8 |
# DeepSeek-V3.2-Exp
|
| 9 |
|