Instructions to use WillHeld/via-llama with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use WillHeld/via-llama with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("WillHeld/via-llama", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
tokenizer fix 831511c
Helw150 commited on
tokenizer fix 9ce6bce
Helw150 commited on
tokenizer fix 6232d79
Helw150 commited on
llama 3 5819f88
Helw150 commited on
Soft Detokenize 257b73a
Helw150 commited on
Update config.json e9df8b7 verified
Soft Detokenize 64cb2d1
Helw150 commited on
Soft Detokenize 15aa1e2
Helw150 commited on
448 Query Tokens 389ceef
Helw150 commited on
Full NAR Window 5aca5eb
Helw150 commited on
Update config.json 6b67330 verified
Update config.json e1bdd50 verified
Update config.json 1b2d511 verified
Kinda Querying Transformer b8234e4
Helw150 commited on
Decoder.Model 29e9579
Helw150 commited on
LM Head b6b8e1d
Helw150 commited on
Out with the old 1b1139b
Helw150 commited on
Fix Safetensor Format df054ea
Helw150 commited on
Better Safetensor Format 08ce4b7
Helw150 commited on
Index e5c4b80
Helw150 commited on
Upload tokenizer 5481e76 verified
Pretrained 1e77bed
Helw150 commited on