leandronunes/qwen3.8-4b-a1-dare-ties
Full fine-tune vs InternScience/Agents-A1-4B. Fine-tune: 93% of the weights changed, across 32 of 32 layers.
Tensors changed: Attention 186/320, MLP 96/196, Embeddings 1/4, Norms 72/203
Full fine-tune vs InternScience/Agents-A1-4B. Fine-tune: 93% of the weights changed, across 32 of 32 layers.
Tensors changed: Attention 186/320, MLP 96/196, Embeddings 1/4, Norms 72/203