jeffmeloy/Qwen2.5-7B-olm-v1.1

Full fine-tune vs Qwen/Qwen2.5-7B. Fine-tune: 97% of the weights changed, across 27 of 28 layers.

Tensors changed: Embeddings 1/1, Norms 53/57, MLP 81/84, Attention 189/196, lm_head 1/1

2 derivatives checked