MikeLee000/olmo3-190m-zh-full-continue
Full fine-tune vs MikeLee000/olmo3-190m-zh-full. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, MLP 36/36, Norms 0/49, Attention 48/48
1 derivatives checked
- MikeLee000/olmo3-190m-zh-full-sft: Full fine-tune