Vortex5/Violet-Starlight-12B

Full fine-tune vs DreadPoor/Strawberry_Smoothie-12B-Model_Stock. Fine-tune: 100% of the weights changed, across 40 of 40 layers.

Tensors changed: Norms 71/81, MLP 120/120, Attention 160/160

2 derivatives checked

On the Hub · Downloads on Model Pulse