francesca9805/urd-arab-100mb-after-ppt-mp-struct-100mb-ckpt500_seed455

Full fine-tune vs francesca9805/urd-arab-100mb-ppt-mp-struct-100mb_seed455. Fine-tune: 100% of the weights changed, across 12 of 12 layers.

Tensors changed: Attention 48/48, Norms 25/50, MLP 48/48, Embeddings 2/2

On the Hub · Downloads on Model Pulse