SGK86/dsi-mistral-7b-searchtome-subset
Full fine-tune vs mistralai/Mistral-7B-Instruct-v0.2. Fine-tune: 98% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 14/65, MLP 96/96, Attention 128/128, lm_head 1/1
Full fine-tune vs mistralai/Mistral-7B-Instruct-v0.2. Fine-tune: 98% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 14/65, MLP 96/96, Attention 128/128, lm_head 1/1