0

Sentence Repeater RL Env (Prime Intellect)

Fresh

Interactively repeat sentences from a paragraph in a random order

Type
RL Env
Runtime
multi-turn
License
unknown
Size
v0.1.1
Published
Dec 2025

Cite

Notes

Only stored in your browser.

sentence-repeater

Source Code

Overview

  • Environment ID: sentence-repeater
  • Short description: Multi-turn QA over a paragraph to retrieve specific sentences in random order.
  • Tags: retrieval, multi-turn, text, similarity

Datasets

  • Primary dataset(s): agentlans/wikipedia-paragraphs processed into 5 Q/A turns per paragraph
  • Source links: Hugging Face Datasets
  • Split sizes: Train split filtered to paragraphs with 5 adequate-length sentences

Task

  • Type: multi-turn
  • Rubric overview: Sequence similarity against the 5 target answers across turns

Quickstart

Run an evaluation with default settings:

prime eval run sentence-repeater

Configure model and sampling:

prime eval run sentence-repeater \
  -m gpt-4.1-mini \
  -n 20 -r 3 -t 1024 -T 0.7

Environment Arguments

This loader does not expose custom arguments.

Metrics

MetricMeaning
rewardMean SequenceMatcher ratio across the 5 repeated sentences