Sign in to view source links and access this dataset
Description
Approximately 600 million tokens and 405,000 examples make up this large-scale dataset for competitive exam preparation and reasoning AI. Generated using advanced distillation techniques, it covers over 25 Indian and international examinations including JEE, NEET, UPSC, Banking, GRE, and IELTS. The dataset was created by shravan01 and last updated on December 15, 2025.
Use Cases
Training language models for exam question answering based on coverage of over 25 competitive exams.
Developing chain-of-thought reasoning models based on the structured reasoning techniques mentioned.
Benchmarking AI performance on STEM and humanities subjects based on the described domain coverage.
Fine-tuning models for specific exam types like JEE or GRE based on the named examination categories.
Strengths
Contains approximately 600 million tokens, indicating a substantial text corpus.
Includes 405,000 examples, providing a large number of training instances.
Covers over 25 named competitive examinations from multiple countries.
Spans multiple domains including STEM, humanities, current affairs, and language as described.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count is unknown, which may limit suitability assessment.
Freshness should be verified as the last update was December 15, 2025.
Provenance
Source
huggingface
Collection Method
Generated using advanced distillation techniques.
Freshness
Last updated 2025-12-15 05:57:45.
Geography
Covers Indian and international examinations.
License is unknown; terms of use must be verified before application.