anthracite-org/magnum-v1-32b
Full fine-tune vs Qwen/Qwen1.5-32B. Fine-tune: 98% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 0/1, Norms 99/129, MLP 192/192, Attention 448/448, lm_head 1/1
5 derivatives checked
- anthracite-org/magnum-v1-32b-gguf: Quantization
- bartowski/magnum-32b-v1-GGUF: Same weights, converted
- mradermacher/magnum-32b-v1-i1-GGUF: Quantization
- mradermacher/magnum-32b-v1-GGUF: Quantization
- featherless-ai-quants/anthracite-org-magnum-v1-32b-GGUF: Quantization