TareksGraveyard/Savant-Qwen2.5-72B

Full fine-tune vs EVA-UNIT-01/EVA-Qwen2.5-72B-v0.2. Fine-tune: 100% of the weights changed, across 80 of 80 layers.

Tensors changed: Norms 146/161, MLP 240/240, Attention 560/560

2 derivatives checked

• mradermacher/Savant-Qwen2.5-72B-i1-GGUF: Quantization
• mradermacher/Savant-Qwen2.5-72B-GGUF: Quantization

On the Hub · Downloads on Model Pulse