timpal0l/dolphin-2.9-llama3-8b-flashback

Full fine-tune vs dphn/dolphin-2.9-llama3-8b. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128

1 derivatives checked

• mradermacher/dolphin-2.9-llama3-8b-flashback-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse