mergekit-community/Hammer2-LLM-Qwen-1.5B
Full fine-tune vs MadeAgents/Hammer2.1-1.5b. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 196/196
1 derivatives checked
- mradermacher/Hammer2-LLM-Qwen-1.5B-GGUF: Same weights, converted