Dingdust/NeoHorse-1-9B-heretic
Partial fine-tune (fewer layers) vs Qwen/Qwen3.5-9B. Dingdust/NeoHorse-1-9B-heretic matches part of Qwen/Qwen3.5-9B (427 of its 775 tensors; the rest of Qwen/Qwen3.5-9B has no counterpart). On the shared part: Partial change: 89% of the weights changed (attention, lm_head, mlp, norms), in 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 0/1, MLP 96/96, Attention 182/224, Norms 70/105
2 derivatives checked
- mradermacher/NeoHorse-1-9B-heretic-i1-GGUF: Quantization
- mradermacher/NeoHorse-1-9B-heretic-GGUF: Same weights, converted