Na0s/Mixtral-8x7B-Instruct-v0.1-exhaustive-LoRA
Partial fine-tune vs mistralai/Mixtral-8x7B-Instruct-v0.1. Partial change: 3% of the weights changed (attention, other), in 32 of 32 layers.
Tensors changed: Embeddings 0/1, MLP 0/768, other 32/32, Norms 0/65, Attention 128/128, lm_head 0/1