Kadabra/Qwen3.5-4B-Gascon-CPT
Full fine-tune vs unsloth/Qwen3.5-4B-Base. Fine-tune: 93% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/4, Norms 0/203, Attention 152/320, MLP 96/196
Full fine-tune vs unsloth/Qwen3.5-4B-Base. Fine-tune: 93% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/4, Norms 0/203, Attention 152/320, MLP 96/196