nbeerbower/Llama3-Sapientia-70B
Full fine-tune vs Black-Ink-Guild/Pernicious_Prophecy_70B. Fine-tune: 100% of the weights changed, across 80 of 80 layers.
Tensors changed: Norms 161/161, MLP 240/240, Attention 320/320
2 derivatives checked
- mradermacher/Llama3-Sapientia-70B-i1-GGUF: Quantization
- mradermacher/Llama3-Sapientia-70B-GGUF: Quantization