mayflowergmbh/Wiederchat-7b-dpo-laser
Full fine-tune vs QuixiAI/openchat-3.5-0106-laser. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128
2 derivatives checked
- mradermacher/Wiederchat-7b-dpo-laser-i1-GGUF: Quantization
- mradermacher/Wiederchat-7b-dpo-laser-GGUF: Same weights, converted