nebchi/solar-merge-slerp

Full fine-tune vs chihoonlee10/T3Q-ko-solar-dpo-v6.0. Fine-tune: 97% of the weights changed, across 47 of 47 layers.

Tensors changed: Embeddings 0/1, Norms 12/95, MLP 138/141, Attention 184/188, lm_head 1/1

On the Hub · Downloads on Model Pulse