thaddickson/Delphi-7B-v1
Full fine-tune vs Orion-zhen/Qwen2.5-7B-Instruct-Uncensored. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 196/196, lm_head 1/1
2 derivatives checked
- mradermacher/Delphi-7B-v1-i1-GGUF: Full fine-tune
- mradermacher/Delphi-7B-v1-GGUF: LoRA-style