nvidia/Ising-Calibration-1-35B-A3B
Partial fine-tune vs Qwen/Qwen3.5-35B-A3B. Partial change: 67% of the weights changed (attention, lm_head, mlp, norms), in 40 of 40 layers.
Tensors changed: lm_head 1/1, Embeddings 0/4, Attention 234/388, MLP 200/312, Norms 92/241
2 derivatives checked
- mradermacher/Ising-Calibration-1-35B-A3B-i1-GGUF: Quantization
- mradermacher/Ising-Calibration-1-35B-A3B-GGUF: Quantization
2 other repos hold byte-for-byte identical weight files.