prithivMLmods/Raptor-X5-UIGEN
Full fine-tune vs prithivMLmods/Viper-Coder-v1.7-Vsm6. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Norms 97/97, MLP 144/144, Attention 336/336
2 derivatives checked
- mradermacher/Raptor-X5-UIGEN-GGUF: Quantization
- DevQuasar-2/prithivMLmods.Raptor-X5-UIGEN-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.