YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
VLA fine-tuning-scope policies
Seven historical checkpoints retain their original bytes. All seven passed native loading and normalization-statistics checks. Original configs, statistics and training logs remain beside each checkpoint.
Training data have one shared owner: latency-sensitive-bench/latency-transfer-data@5305590625c28a6922e73f46d18d52738762c906, prefix flappy_fix_latency_2_200ep. Evaluation and historical-code evidence are in vla-finetuning-scope-data. See run_index.json for checkpoint identities, code commits, W&B runs and exact paper consumers.
Consolidated experiment sources
experiment_repository_index.json records each source repository, immutable revision, retained weight identity, original training configuration and paper use. Historical trials remain separate from formal paper runs. Missing checkpoints and unpublished local data remain explicit. Original configs are provenance; current training recipes use the fixed dataset references in the index. Full provenance and training code are in latency-sensitive-bench/vla-finetuning-scope-data@3d9e18dcde6fcc564af8bf292ff58a8f73ed616f.
Shared Flappy raw data now have one owner: latency-sensitive-bench/latency-transfer-data@5305590625c28a6922e73f46d18d52738762c906, prefix training/flappy_200ep/. Original repository names in provenance identify sources that were consolidated.