Nohobby/YetAnotherMerge-v0.5

Full fine-tune vs MarinaraSpaghetti/NemoRemix-12B. Fine-tune: 93% of the weights changed, across 40 of 40 layers.

Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 2/81, MLP 110/120, Attention 146/160

2 derivatives checked

On the Hub · Downloads on Model Pulse