eval_run_idstring required
Unique evaluation run identifier.
bench_idstring required
Benchmark evaluated by this run.
display_namestring nullable required
User-facing name for the evaluation run; null for legacy rows.
model_slugstring nullable
Model evaluated by this run.
status'pending' | 'registered' | 'running' | 'completed' | 'failed' | 'cancelled' required
checkpoint_idstring nullable
Checkpoint evaluated by this run, when applicable.
training_run_idstring nullable
Training run that produced the evaluated checkpoint.
stepinteger nullable
Evaluated checkpoint training step.
reward_meannumber nullable
Completed evaluation mean reward.
created_atstring date-time required
Time when the evaluation run was created.
updated_atstring date-time required
Time when the evaluation run was last updated.
completed_atstring date-time nullable
Time when the evaluation reached a terminal status.