What is neptune.ai?

Neptune is an experiment tracker purpose-built for foundation model training.

With Neptune, you can monitor thousands of per-layer metrics—losses, gradients, and activations—at any scale. Visualize them with no lag and no missed spikes. Drill down into logs and debug training issues fast. Keep your model training stable while reducing wasted GPU cycles.

Examples

This repo contains tutorials and examples of how to use Neptune.

	Docs	Neptune	GitHub	Colab
Quickstart
Log different types of metadata
Organize and filter runs
Resume run or other object
Use Neptune in HPO jobs
Debug training runs			[][debug-notebook]	[][debug-colab]

Integrations

	Docs	Neptune	GitHub	Colab
PyTorch Lightning			[][lightning-notebook]	[][lightning-colab]

Migration tools

	Docs	GitHub
Import runs from Weights & Biases
Migrate code from Neptune Legacy client

Monitoring tools

	Docs	Neptune	GitHub
Hardware monitoring
PyTorch model internals

Visualization tools

	GitHub
Compare media file-series

Can't find what you're looking for?

• Visit the documentation →
• Check out the blog →
• Visit our Support Center to get help or contact the Support team.

[lightning-notebook]: TODO: final notebook URL in GH [lightning-colab]: TODO: final colab URL [debug-notebook]: how-to-guides/debug-model-training-runs/debug_training_runs.ipynb [debug-colab]: https://colab.research.google.com/github/neptune-ai/scale-examples/blob/master/how-to-guides/debug-model-training-runs/notebooks/debug_training_runs.ipynb