DazzlingXeno/Westlake-Scribe
Full fine-tune (fewer layers) vs OmnicromsBrain/Eros_Scribe-10.7b-v3. OmnicromsBrain/Eros_Scribe-10.7b-v3 matches part of DazzlingXeno/Westlake-Scribe (435 of its 732 tensors; the rest of DazzlingXeno/Westlake-Scribe has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 97/97, MLP 144/144, Attention 192/192
2 derivatives checked
- mradermacher/Westlake-Scribe-GGUF: Quantization
- mradermacher/Westlake-Scribe-i1-GGUF: Quantization