Undi95/MiquTotalMix-70B

Partial fine-tune vs 152334H/miqu-1-70b-sf. Partial change: 66% of the weights changed (attention, mlp, norms), in 56 of 80 layers.

Tensors changed: Embeddings 0/1, Norms 76/161, MLP 160/240, Attention 212/320, lm_head 0/1

2 derivatives checked

On the Hub · Downloads on Model Pulse