cisimcik/tau-4b
Partial fine-tune vs cisimcik/cisimcik-base. Partial change: 85% of the weights changed (attention, mlp, norms), in 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 87/105, Attention 205/224, MLP 96/96
Partial fine-tune vs cisimcik/cisimcik-base. Partial change: 85% of the weights changed (attention, mlp, norms), in 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 87/105, Attention 205/224, MLP 96/96