Aleteian/saiga-vikhr-dostoevsky-ru-mn-12b

Full fine-tune vs IlyaGusev/saiga_nemo_12b. Fine-tune: 98% of the weights changed, across 40 of 40 layers.

Tensors changed: Embeddings 1/1, Norms 0/81, MLP 117/120, Attention 156/160, lm_head 1/1

6 derivatives checked

On the Hub · Downloads on Model Pulse