fpadovani/arb-arab-10mb-after-ppt-Dp-10mb-ckpt500_seed3407

Full fine-tune vs fpadovani/arb-arab-10mb-ppt-Dp-10mb_seed3407. Fine-tune: 100% of the weights changed, across 4 of 4 layers.

Tensors changed: Attention 16/16, Norms 9/18, MLP 16/16, Embeddings 2/2

On the Hub · Downloads on Model Pulse