zariness00/smolk12_dpo
Full fine-tune vs HuggingFaceTB/SmolLM2-360M-Instruct. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 64/65, MLP 96/96, Attention 128/128
Full fine-tune vs HuggingFaceTB/SmolLM2-360M-Instruct. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 64/65, MLP 96/96, Attention 128/128