schonsense/70B_Incisive_Vernacular
Full fine-tune vs schonsense/70B_Triage. Fine-tune: 100% of the weights changed, across 80 of 80 layers.
Tensors changed: Norms 161/161, MLP 240/240, Attention 320/320, lm_head 1/1, Embeddings 1/1
2 derivatives checked
- mradermacher/70B_Incisive_Vernacular-i1-GGUF: Quantization
- mradermacher/70B_Incisive_Vernacular-GGUF: Quantization