mediabiasgroup/magpie-pt-xlm
Full fine-tune vs FacebookAI/xlm-roberta-base. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: Embeddings 5/5, Norms 48/48, Attention 96/96, other 48/50
2 derivatives checked
- mediabiasgroup/magpie-babe-ft: Full fine-tune
- mediabiasgroup/magpie-babe-ft-xlm: Full fine-tune