nitky/Swallow-70b-RP-EX
Full fine-tune vs karakuri-ai/karakuri-lm-70b-chat-v0.1. 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/Swallow-70b-RP-EX-i1-GGUF: Quantization
- mradermacher/Swallow-70b-RP-EX-GGUF: Quantization