DavidAU/Qwen3.6-21B-IQ-Ultra-Heretic-Uncensored-Thinking
Partial fine-tune vs trohrbaugh/Qwen3.6-27B-heretic-ara. Partial change: 86% of the weights changed (attention, mlp, norms), in 48 of 48 layers.
Tensors changed: lm_head 0/1, Embeddings 0/4, Norms 104/267, Attention 300/444, MLP 144/256
3 derivatives checked
- mradermacher/Qwen3.6-21B-IQ-Ultra-Heretic-Uncensored-Thinking-i1-GGUF: Quantization
- mradermacher/Qwen3.6-21B-IQ-Ultra-Heretic-Uncensored-Thinking-GGUF: Quantization
- sealad886/DavidAU-Qwen3.6-21B-IQ-Ultra-Heretic-Uncensored-Thinking-mixed_2_6-MLX: Quantization
1 other repos hold byte-for-byte identical weight files.