Instructions to use FasterDecoding/medusa-vicuna-7b-v1.3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FasterDecoding/medusa-vicuna-7b-v1.3 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("FasterDecoding/medusa-vicuna-7b-v1.3", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update config.json
#2
by vishwasthedeveloper - opened
- config.json +2 -1
config.json
CHANGED
|
@@ -2,5 +2,6 @@
|
|
| 2 |
"base_model_name_or_path": "lmsys/vicuna-7b-v1.3",
|
| 3 |
"medusa_num_heads": 2,
|
| 4 |
"medusa_num_layers": 1,
|
| 5 |
-
"transformers_version": "4.31.0"
|
|
|
|
| 6 |
}
|
|
|
|
| 2 |
"base_model_name_or_path": "lmsys/vicuna-7b-v1.3",
|
| 3 |
"medusa_num_heads": 2,
|
| 4 |
"medusa_num_layers": 1,
|
| 5 |
+
"transformers_version": "4.31.0",
|
| 6 |
+
"model_type": "llama"
|
| 7 |
}
|