This is a refactor of arman-bd/guppylm-9M to be compliant with transformers's custom_model.
python inference.py guppylm-9M
GuppyLMForCausalLM loaded: 8.7M params
Guppy Chat (type 'quit' to exit)
You> is there a cat in the room?
Guppy> i don't like it. it puts its face on the glass by the bubbles.
You> I'm sorry. are you hungry?
Guppy> i don't eat it.
This is a good small LLM for doing execise with SGLang.
- Downloads last month
- 22