SriyaM/MHC-Coach

Full fine-tune vs meta-llama/Meta-Llama-3-70B. Fine-tune: 100% of the weights changed, across 80 of 80 layers.

Tensors changed: Embeddings 1/1, Norms 161/161, MLP 240/240, Attention 320/320, lm_head 1/1

1 derivatives checked