Minerva Llama3 Instruct

This is a https://huggingface.co/sapienzanlp/Minerva-3B-base-v1.0 fine-tune with a Llama3 prompt and the camoscio dataset in the Italian language.

It is fine-tuned with the llama3 instructor prompt so it is suitable for chat.

Note: the EOS is </s>

Safetensors

Model size

2.89B params

Tensor type

BF16

Collection including mudler/Minerva-3B-Llama3-Instruct-v0.1