regnant-io/kw5-149M-instruct-r2
Full fine-tune vs regnant-io/kw5-149M. Fine-tune: 100% of the weights changed, across 20 of 20 layers.
Tensors changed: Norms 81/81, Attention 100/100, other 40/40, MLP 80/80, lm_head 1/1, Embeddings 1/1
Full fine-tune vs regnant-io/kw5-149M. Fine-tune: 100% of the weights changed, across 20 of 20 layers.
Tensors changed: Norms 81/81, Attention 100/100, other 40/40, MLP 80/80, lm_head 1/1, Embeddings 1/1