utter-project/EuroMoE-2.6B-A0.6B-Instruct-Preview
Full fine-tune vs utter-project/EuroMoE-2.6B-A0.6B-2512. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Embeddings 1/1, MLP 2830/2830, other 24/24, Norms 42/49, Attention 96/96
2 derivatives checked
- nyuuzyou/EuroMoE-2.6B-A0.6B-Instruct-Preview-GGUF: Same weights, converted
- DevQuasar/utter-project.EuroMoE-2.6B-A0.6B-Instruct-Preview-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.