Weights biases
Weights & Biases ML experiment tracking. Use for ML monitoring.From its SKILL.md
npx -y skills add G1Joshi/Agent-Skills --skill weights-biasesAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 12 stars12 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.0 KB, 223 tokens by cl100k_base, as published. Nobody here has run it
Weights & Biases (W&B)
W&B is the "Github for ML models". It tracks every run, hyperparameter, and artifact. 2025 brings W&B Inference and Weave.
When to Use
- Visualization: Live loss curves during training.
- Collaboration: Sharing reports ("The loss spiked here") with teammates.
- Model Registry: Managing lifecycle.
Core Concepts
W&B Run
A single experiment.
Sweeps
Hyperparameter optimization engine.
Weave
A toolkit for logging and debugging interactive GenAI applications (like Tracing).
Best Practices (2025)
Do:
- Use Weave: For LLM apps. It replaces simple text logging.
- Log Model Artifacts: Save
best_model.ptto W&B, not local disk.
Don't:
- Don't leak secrets: W&B logs configs. Ensure API keys aren't in your
configobject.
References
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.