MaziyarPanahi/calme-2.8-qwen2-7b
Full fine-tune vs Qwen/Qwen2-7B. Fine-tune: 93% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 55/57, MLP 84/84, Attention 196/196, lm_head 1/1
4 derivatives checked
- MaziyarPanahi/calme-2.8-qwen2-7b-GGUF: Same weights, converted
- featherless-ai-quants/MaziyarPanahi-calme-2.8-qwen2-7b-GGUF: Quantization
- Nelathan/Qwen2-7B-FocusMix: Full fine-tune
- win10/high-speed-mixing-7B-V1: Full fine-tune