2.7T Math RL
Collection
Research artifacts from Snowball 2.7T math RL. Raw mutable-router exports may collapse; use only models marked repaired or verified frozen. • 17 items • Updated
Snowball-67B-A2B-Math-RL-E17b2-Step20-Repaired
This is the exact checkpoint used for a reported row in the Snowball 67B-A2B math-RL experiment.
Router state: SFT router-bias repaired; VERIFY OK. Closely named older laion/rl-snowball-* repositories may contain raw mutable-router exports that collapse under inference. Do not substitute them for this artifact. Models from this campaign are research checkpoints, not production releases; their utility is limited unless router-bias repair or frozen-router integrity is preserved.
- Arm:
E17b2 - Step:
20 - Held-out AIME24 / MATH-500 / OlympiadBench:
20.67 / 77.20 / 22.00 - Selection: Best E17 step-20 arm
- Source artifact:
s3://marin-us-east-02a/marin/exports/snowball-bias-repaired/rl-snowball-e17b2-rno2a-rlvrmath-interp015-sr-20260825-014631/global_step_20/policy/ - Experiment issue: https://github.com/marin-community/marin/issues/7786
- Evidence archive: https://huggingface.co/datasets/penfever/snowball-67b-a2b-math-rl-artifacts
The scores and evaluation caveats are recorded in MATH_EVALS.md in the evidence archive. Preserve config.json, the tokenizer files, and all shards named by model.safetensors.index.json together.
- Downloads last month
- -