Text Generation
Transformers
PyTorch
Italian
English
mistral
conversational
text-generation-inference
Inference Endpoints

Model max length

#5
by prodPc - opened

hello what is the maximum token length supported?

Hi it is actually 8192 tokens.
There is a mistake in the https://huggingface.co/galatolo/cerbero-7b/blob/main/config.json (sliding_window is set to 4096) but the model was trained with a context size of 8192 so it can be safely used within those borders

galatolo changed discussion status to closed

Sign up or log in to comment