ehristoforu/mistral-distil-test-2

Partial fine-tune vs unsloth/Mistral-Nemo-Instruct-2407. Partial change: 40% of the weights changed (attention, mlp, norms), in 14 of 30 layers.

Tensors changed: Embeddings 0/1, Norms 28/61, MLP 42/90, Attention 56/120, lm_head 0/1

On the Hub · Downloads on Model Pulse