Dolci Think SFT Sources include a mixture of existing reasoning traces: OpenThoughts 3 (Apache 2.0): Extended to 32K context length and downsampled code prompts to 16X multiple, to 941,164 total prompts. Access our version, Dolci OpenThoughts 3 here. SYNTHETIC 2 (Apache 2.0) via the SFT Verified split, 104,568 prompts. Nemotron Post training dataset (CC BY 4), code split only, 113,777 prompts. New prompts and new reasoning traces from us (all ODC BY 1.0): Dolci Think Persona IF: New precise instruction following prompts and traces created with Nvidia's Nemotron Post training Personas. 220,530 prompts. Dolci Precise IF: New multi constraint instruction following data building off Pyatkin, Valentina, et al. "Generalizing Verifiable Instruction Following." (2025). 135,722 prompts. Dolci Think Python: 466,676 prompts (subsampled from larger mix). Existing prompts with new reasoning traces, largely repurposed from Tülu 3 / OLMo 2, with new traces generated by a mix of DeepSeek R1 and DeepSeek R1 0528: WildChat (ODC BY 1.0), 76,209 prompts. OpenAssistant Guanaco (Apache 2.0), 6,647 prompts. CoCoNot (ODC BY 1.0), 9,549 prompts. WildGuardMix (Apache 2.0), 36,673 prompts. WildJailbreak (ODC…
Runs entirely in your browser via DuckDB-Wasm — this dataset's real data file is loaded once, then queried locally. Nothing is sent to a server.
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy