Text Generation
Transformers
Safetensors
PyTorch
nemotron_h
nvidia
conversational
custom_code
Eval Results

Recommended parameters?

#30
by leonsarmiento - opened

Temperature: ?
TopK: ?
MinP: ?
TopP: ?

thanks!

temperature=1.0 and top_p=1.0 are recommended for reasoning tasks, while temperature=0.6 and top_p=0.95 are recommended for tool calling.

From the model card - these are the recommended settings.

Thanks!

Sign up or log in to comment