MMOPD/Qwen3-4B-OT3-law
Full fine-tune vs MMOPD/Qwen3-4B-OT3-2ep. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 117/145, MLP 108/108, Attention 144/144, lm_head 1/1
Full fine-tune vs MMOPD/Qwen3-4B-OT3-2ep. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 117/145, MLP 108/108, Attention 144/144, lm_head 1/1