Skywork/MindLink-72B-0801
Full fine-tune vs Qwen/Qwen2.5-72B-Instruct. Fine-tune: 100% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 1/1, Norms 152/161, MLP 240/240, Attention 560/560, lm_head 1/1
3 derivatives checked
- bartowski/Skywork_MindLink-72B-0801-GGUF: Quantization
- mradermacher/MindLink-72B-0801-i1-GGUF: Quantization
- lmstudio-community/MindLink-72B-0801-GGUF: Quantization
2 other repos hold byte-for-byte identical weight files.