12384-0ashntoi/Qwen3-4B-Think-GRPO-MLX
Full fine-tune vs Qwen/Qwen3-4B-Instruct-2507. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 0/145, MLP 108/108, Attention 144/144
Full fine-tune vs Qwen/Qwen3-4B-Instruct-2507. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 0/145, MLP 108/108, Attention 144/144