OmnicromsBrain/EverythingBagel-DPO-7B
Full fine-tune vs SanjiWatsuki/Silicon-Maid-7B. Fine-tune: 97% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 65/65, MLP 93/96, Attention 124/128, lm_head 1/1
4 derivatives checked
- mradermacher/EverythingBagel-DPO-7B-GGUF: Same weights, converted
- OmnicromsBrain/Eros_Scribe-7b-GGUF: Full fine-tune
- OmnicromsBrain/Eros_Scribe-7b: Full fine-tune
- OmnicromsBrain/Eros_Scribe-10.7b-V.02: Full fine-tune (fewer layers)