datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
rm-dataset-v0.0.1
Verifiable Labs RM dataset v0.0.1
Reward-model training data for the Verifiable Labs SDK,
produced by the Phase 29 reward-distillation pipeline.
Stats
Rows: 840
With frontier judgment: 194
Frontier judge: anthropic/claude-sonnet-4 (when judged)
Source mix:
env: 646
judgment: 194
Schema
Each row is a JSON object with the following fields:
field
type
meaning
row_id
str
unique id
env_id
str
env that produced the row
prompt
str
task… See the full description on the dataset page: https://huggingface.co/datasets/verifiablelabs/rm-dataset-v0.0.1.ashercn97__a1-v0.0.1-details
Dataset Card for Evaluation run of ashercn97/a1-v0.0.1
Dataset automatically created during the evaluation run of model ashercn97/a1-v0.0.1
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An additional… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/ashercn97__a1-v0.0.1-details.