Modelpublic

laion/tt-x0_verifier-binary-20-30B

sourceHugging Faceupdated 2mo agoView on Hugging Face

0likes137downloads

Model Card

tt-x0verifier-binary (TaskTrove X0 binary verifier, globalstep_20)

X0 verifier-ablation binary arm of the TaskTrove hyperparameter sweep. Completed its 20-step horizon. Base: Qwen/Qwen3-Coder-30B-A3B-Instruct; terminus-2 agentic RL on DCAgent/exp_rpt_multifile (pytest verifier, binary reward). Trailing-5 EMA reward at step 20 = 0.1703.

X0 verdict (owner 2026-08-04): the shaped arm wins on the post-#300 unshaped re-score (~3.7-pt gap), so pass_ratio shaping is the campaign verifier for X1-X5. See training_logs/report.md.

Training Traces

Not available. The trial-level trace trees were reclaimed during a 2026-08 GPFS inode-recovery pass before a dataset was uploaded to the Hub; no tt-x0_verifier-binary-traces dataset was published and the source trials no longer exist on GPFS. Per-trial unshaped verifier outcomes survive on Jupiter in /e/scratch/jureap59/feuer1/x0_unshaped_outcomes.tsv. The training_logs/ analysis here is derived from the per-step WANDB mirror, not the (deleted) trials.