open-athena/Grug-67B-A2B-Datakit-SFT-262K-2026.09.21-dr-doom

Full fine-tune vs open-athena/Grug-67B-A2B-Datakit-SFT-262K-2026.09.21. Fine-tune: 92% of the weights changed, across 26 of 26 layers.

Tensors changed: lm_head 1/1, Embeddings 2/4, Norms 106/159, MLP 173/208, Attention 130/130

1 derivatives checked

• open-athena/Grug-67B-A2B-Dr-Doom-Science-SFT-Step5989-2026.10.02: Partial fine-tune

On the Hub · Downloads on Model Pulse