nvidia/OpenMath-Nemotron-1.5B
Full fine-tune vs Qwen/Qwen2.5-Math-1.5B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 196/196
8 derivatives checked
- bartowski/nvidia_OpenMath-Nemotron-1.5B-GGUF: Same weights, converted
- mradermacher/OpenMath-Nemotron-1.5B-i1-GGUF: Quantization
- mradermacher/OpenMath-Nemotron-1.5B-GGUF: Same weights, converted
- DevQuasar/nvidia.OpenMath-Nemotron-1.5B-GGUF: Same weights, converted
- lmstudio-community/OpenMath-Nemotron-1.5B-GGUF: Quantization
- hbx/JustRL-Nemotron-1.5B: Full fine-tune
- HasuerYu/KnowRL-Nemotron-1.5B: Full fine-tune
- zwhe99/DeepMath-Omn-1.5B: Partial fine-tune