Goedel-LM/Goedel-Formalizer-V2-32B
Full fine-tune vs Qwen/Qwen3-32B. Fine-tune: 98% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 0/1, Norms 256/257, MLP 192/192, Attention 256/256, lm_head 1/1
2 derivatives checked
- mradermacher/Goedel-Formalizer-V2-32B-i1-GGUF: Quantization
- mradermacher/Goedel-Formalizer-V2-32B-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.