SakanaAI/RePo-OLMo2-1B-stage2-L5
Full fine-tune vs allenai/OLMo-2-0425-1B. Fine-tune: 100% of the weights changed, across 16 of 16 layers.
Tensors changed: Embeddings 1/1, MLP 48/48, Norms 65/65, Attention 64/64, lm_head 1/1
Full fine-tune vs allenai/OLMo-2-0425-1B. Fine-tune: 100% of the weights changed, across 16 of 16 layers.
Tensors changed: Embeddings 1/1, MLP 48/48, Norms 65/65, Attention 64/64, lm_head 1/1