usermma/Qwythos-9B-v2-Heretic-Abliterated
Partial fine-tune (fewer layers) vs empero-ai/Qwythos-9B-v2. usermma/Qwythos-9B-v2-Heretic-Abliterated matches part of empero-ai/Qwythos-9B-v2 (427 of its 775 tensors; the rest of empero-ai/Qwythos-9B-v2 has no counterpart). On the shared part: Partial change: 80% of the weights changed (attention, lm_head, mlp), in 30 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 0/105, Attention 76/224, MLP 90/96
1 derivatives checked
- maxylium/Qwythos-9B-v2-Heretic-Abliterated-GGUF: Quantization