A controlled 29-step replication of the Affine SN120 MiMo GRPO run (affine-mimo9b-grpo-20261013), started from the same initial checkpoint. It uses the official trainer and settings unchanged except for one switch: cappolicy = affine-unresolved-zero-reward-active-v1. Unresolved rollouts (hit the 8192-token cap, or no gradeable final answer) get reward 0 and stay in the loss, instead of being masked. A control arm, S0, is identical except it uses the live run's original affine-unresolved-zero-reward-mask-all-v1. S0 reproduces the live run's rise-then-decline on held-out data and its entropy collapse. S2 shows neither. S2 is the live run's new training policy without the new qualification…
Publisher
Alllwang
alllwang
Models in Library1
Datasets in Library0
Models on Hugging Face—
Followers—
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Reported licenses: mit. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.