prithivMLmods/Viper-Coder-v1.7-Vsm6
Full fine-tune vs prithivMLmods/Viper-Coder-v1.6-r999. Fine-tune: 98% of the weights changed, across 48 of 48 layers.
Tensors changed: Norms 97/97, MLP 141/144, Attention 329/336
4 derivatives checked
- mradermacher/Viper-Coder-v1.7-Vsm6-i1-GGUF: Quantization
- DevQuasar-9/prithivMLmods.Viper-Coder-v1.7-Vsm6-GGUF: Same weights, converted
- mradermacher/Viper-Coder-v1.7-Vsm6-GGUF: Quantization
- prithivMLmods/Raptor-X5-UIGEN: Full fine-tune
1 other repos hold byte-for-byte identical weight files.