Experiment logs, synced 2026-10-17 20:46:24 every few minutes. tdprmresults.tar (5946 MB, tar -xf tdprmresults.tar -> tdprmresults/) holds every metric, trace, evaluation, state, queue and service log of the run; no model weights, environments or dataset text. Updated 2026-10-17 20:46:22 · refreshes automatically Training queue (finished): training runs (97) done 97; other jobs (data, states, base evals) done 44 Step-credit evaluation queue (finished): 10 jobs, done 10 (see Step-credit evaluation) Greedy accuracy, v1 and v2 methods in separate tables. Δ: v1 methods vs grpo, v2 methods vs grpov2 (same cell), GRPO rows vs the base model. pass@8 on the in-domain set; mean@8 / pass@8 for every…
Independent publisher
Junda Wu
JoshuaW1997
Models in Library0
Datasets in Library1
Models on Hugging Face10
Followers4
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.
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.