c-bone/CrystaLLM-pi_density
Full fine-tune vs c-bone/CrystaLLM-pi_base. Fine-tune: 100% of the weights changed, across 8 of 8 layers.
Tensors changed: Attention 32/32, Norms 34/34, MLP 32/32, Embeddings 1/1
Full fine-tune vs c-bone/CrystaLLM-pi_base. Fine-tune: 100% of the weights changed, across 8 of 8 layers.
Tensors changed: Attention 32/32, Norms 34/34, MLP 32/32, Embeddings 1/1