OmnicromsBrain/Eros_Scribe-10.7b-V.02
Full fine-tune (fewer layers) vs OmnicromsBrain/EverythingBagel-DPO-7B. OmnicromsBrain/EverythingBagel-DPO-7B matches part of OmnicromsBrain/Eros_Scribe-10.7b-V.02 (291 of its 435 tensors; the rest of OmnicromsBrain/Eros_Scribe-10.7b-V.02 has no counterpart). On the shared part: Fine-tune: 98% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 0/1, Embeddings 1/1, Norms 64/65, MLP 96/96, Attention 128/128
2 derivatives checked
- mradermacher/Eros_Scribe-10.7b-V.02-GGUF: Quantization
- mradermacher/Eros_Scribe-10.7b-V.02-i1-GGUF: Quantization