Finnish-NLP/Ahma-7B-Instruct
Full fine-tune vs Finnish-NLP/Ahma-7B. 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/Ahma-7B-Instruct-GGUF: Same weights, converted
- RASMUS/Ahma-7B-Instruct-SimPO: LoRA-style