paulml/DPOB-INMTOB-7B
Full fine-tune vs eren23/merged-dpo-binarized-NeutrixOmnibe-7B. Fine-tune: 97% of the weights changed, across 32 of 32 layers.
Tensors changed: MLP 93/96, Norms 64/65, Attention 124/128, lm_head 1/1, Embeddings 1/1
2 derivatives checked
- mradermacher/DPOB-INMTOB-7B-i1-GGUF: Quantization
- mradermacher/DPOB-INMTOB-7B-GGUF: Same weights, converted