shaafsalman/Qwen2.5-32B-Med-Merge

Full fine-tune vs Qwen/Qwen2.5-32B-Instruct. Fine-tune: 98% of the weights changed, across 64 of 64 layers.

Tensors changed: Norms 128/129, MLP 189/192, Attention 441/448

2 derivatives checked

• mradermacher/MedQWEN-2.5-32B-i1-GGUF: Quantization
• mradermacher/MedQWEN-2.5-32B-GGUF: Quantization

On the Hub · Downloads on Model Pulse