0xKitkat/Ornith-1.5-35B-A3B-Uncensored
Full fine-tune (fewer layers) vs Qwen/Qwen3.6-35B-A3B. Qwen/Qwen3.6-35B-A3B matches part of 0xKitkat/Ornith-1.5-35B-A3B-Uncensored (1043 of its 1811 tensors; the rest of 0xKitkat/Ornith-1.5-35B-A3B-Uncensored has no counterpart). On the shared part: Fine-tune: 99% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 2/5, Attention 268/392, MLP 285/397, Norms 115/247, lm_head 1/1, other 1/1
6 derivatives checked
- zerodigest/Ornith-1.5-35B-Uncensored-YMQ-MTP-GGUF: Quantization
- mradermacher/Ornith-1.5-35B-A3B-Uncensored-i1-GGUF: Quantization
- mradermacher/Ornith-1.5-35B-A3B-Uncensored-GGUF: Quantization
- dopaemon/Ornith-1.5-35B-Uncensored-YMQ-MTP-GGUF: Quantization
- abhaymin/Ornith-1.5-Cyclos-1M-Uncensored-GGUF: Quantization
- abhaymin/Ornith-1.5-Cyclos-1M-Uncensored: Full fine-tune