barunsaha/misti
Full fine-tune vs unsloth/mistral-7b-bnb-4bit. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: MLP 298/490, Attention 396/655, lm_head 1/1, Embeddings 1/1, Norms 0/65
Full fine-tune vs unsloth/mistral-7b-bnb-4bit. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: MLP 298/490, Attention 396/655, lm_head 1/1, Embeddings 1/1, Norms 0/65