poolside/Laguna-S-2.1-DFlash

Full fine-tune vs poolside/Laguna-S-2.1. Fine-tune: 100% of the weights changed, across 6 of 6 layers.

Tensors changed: Norms 25/25, MLP 3/3, Attention 8/8

1 derivatives checked

• dev7a/Laguna-S-2.1-DFlash-GGUF: Quantization

On the Hub · Downloads on Model Pulse