nvidia/OpenCodeReasoning-Nemotron-7B
Full fine-tune vs Qwen/Qwen2.5-7B-Instruct. Fine-tune: 93% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 57/57, MLP 84/84, Attention 196/196, lm_head 1/1
6 derivatives checked
- bartowski/nvidia_OpenCodeReasoning-Nemotron-7B-GGUF: Same weights, converted
- mradermacher/OpenCodeReasoning-Nemotron-7B-i1-GGUF: Quantization
- mradermacher/OpenCodeReasoning-Nemotron-7B-GGUF: Same weights, converted
- DevQuasar/nvidia.OpenCodeReasoning-Nemotron-7B-GGUF: Same weights, converted
- lmstudio-community/OpenCodeReasoning-Nemotron-7B-GGUF: Quantization
- ferdinandjasong/SuperCoder-7B-Qwen2.5-0525-peft-grpo-v2-merged: LoRA-style