JackHsieh/4B-base.4B-Instruct-reason-thoughts.stride-1.suffix-conditional

Full fine-tune vs Qwen/Qwen3-4B-Base. Fine-tune: 100% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 1/1, Norms 108/145, MLP 108/108, Attention 144/144

On the Hub · Downloads on Model Pulse