babybirdprd/Experiment26-7B-passthrough-6slice

Partial fine-tune vs yam-peleg/Experiment26-7B. Partial change: 75% of the weights changed (attention, mlp, norms), in 25 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 50/65, MLP 75/96, Attention 100/128, lm_head 0/1

On the Hub · Downloads on Model Pulse