tsilva/gradlab-breakout-checkpoint-evals

Checkpoint monitoring trajectories Each complete episode row links to immutable transition tables and lossless PNG chunks. The episode index retains Run, training seed, Checkpoint, evaluation, episode conditions, actions, rewards, boundaries and source/runtime provenance. Frames are full unmasked native RGB at the contracted action cadence. Chunk joins preserve the initial and true terminal image without duplicated transitions. Staged chunks without a complete episode index are… See the full description on the dataset page: https://huggingface.co/datasets/tsilva/gradlab-breakout-checkpoint-eva

<iframe src="https://silytics.si/embed/item/be9b2980-e758-4508-aa06-20c1a7b18d48" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.