Sign in to view source links and access this dataset
Description
119,446 samples from Claude Sonnet 4.6 and 9,633 samples from Claude Opus form a curated collection of open-source Claude distillation datasets. The repository, created by clzoro, provides unified formatting and deduplication of data originally created by others. It was last updated on May 12, 2026.
Use Cases
Fine-tuning smaller models based on the described general, code, math, and psychology data
Benchmarking model performance on tasks distilled from Claude models
Studying knowledge transfer and distillation methodologies using the unified datasets
Strengths
Contains 119,446 samples from Claude Sonnet 4.6
Includes 9,633 samples from Claude Opus
Data has been deduplicated and uniformly formatted
Limitations
Column-level documentation is absent; field semantics must be inferred after download
Row count for the full collection is unknown, which may limit suitability assessment
Provenance
Source
Multiple open-source Claude distillation datasets
Collection Method
Curated, unified, and deduplicated from original sources
Freshness
Last updated 2026-05-12 12:33:06
License is unknown; users should verify licensing terms for the original data sources before use.