ehristoforu/Mistral-nemo-test-2layno-v2

Partial fine-tune vs unsloth/Mistral-Nemo-Instruct-2407. Partial change: 21% of the weights changed (attention, mlp, norms), in 7 of 28 layers.

Tensors changed: Embeddings 0/1, Norms 14/57, MLP 21/84, Attention 28/112, lm_head 0/1

On the Hub · Downloads on Model Pulse