arcee-ai/WitchLM-1.5B
Full fine-tune vs Qwen/Qwen2-1.5B. 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
2 derivatives checked
- mradermacher/WitchLM-1.5B-i1-GGUF: Quantization
- mradermacher/WitchLM-1.5B-GGUF: Same weights, converted