ArmanAsq/AYA-8B

Full fine-tune vs CohereLabs/aya-23-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Norms 33/33, MLP 96/96, Attention 128/128

1 derivatives checked

• mradermacher/AYA-8B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse