McClain/PlasmidGPT-RL
Full fine-tune vs UCL-CSSB/PlasmidGPT-SFT. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: Attention 48/48, Norms 50/50, MLP 48/48, Embeddings 2/2
Full fine-tune vs UCL-CSSB/PlasmidGPT-SFT. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: Attention 48/48, Norms 50/50, MLP 48/48, Embeddings 2/2