YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
NanoGPT ROCStories Model (Fine-tuned GPT-2)
- Experiment: Pre-trained GPT-2 fine-tuned on ROCStories
- Model size: 124M parameters (GPT-2 small)
- Best checkpoint: step 2400
- Validation loss: 2.6528
- Test PPL: 14.31
- Training steps: 3000
- Block size: 256
Generation Parameters
- temperature: 0.75 (recommended based on Qwen scoring)
- top_k: 40
- max_new_tokens: 150
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support