---
title: llama-nemotron-science-reasoning-on-le3000tok-100k
canonical_url: "https://www.modelscope.cn/datasets/laion/llama-nemotron-science-reasoning-on-le3000tok-100k"
md_url: "https://www.modelscope.cn/datasets/laion/llama-nemotron-science-reasoning-on-le3000tok-100k.md"
repository: laion/llama-nemotron-science-reasoning-on-le3000tok-100k
last_updated: 2026-06-09
license: other
storage_size: "256 MB"
downloads: 17
stars: 0
---

# llama-nemotron-science-reasoning-on-le3000tok-100k

> llama-nemotron-science-reasoning-on-le3000tok-100k - laion 在 ModelScope 开源的数据集。Llama-Nemotron science reasoning — Delphi cold-start CoT warmup (≤3000 tok, reasoning:on, 100k)

laion/llama-nemotron-science-reasoning-on-le3000tok-100k 是 ModelScope 魔搭社区上的数据集，存储大小 256 MB，采用 other 许可。

- **Repository**: laion/llama-nemotron-science-reasoning-on-le3000tok-100k
- **License**: other
- **Storage size**: 256 MB
- **Downloads**: 17
- **Stars**: 0
- **Last updated**: 2026-06-09

Source: https://www.modelscope.cn/datasets/laion/llama-nemotron-science-reasoning-on-le3000tok-100k

---

# Llama-Nemotron science reasoning — Delphi cold-start CoT warmup (≤3000 tok, reasoning:on, 100k)

A filtered subsample of **[nvidia/Llama-Nemotron-Post-Training-Dataset](https://huggingface.co/datasets/nvidia/Llama-Nemotron-Post-Training-Dataset)** built to
**warm up the Delphi chat-template reasoning tokens** (`<|start_think|>` / `<|end_think|>`) during
cold-start SFT (marin #6279). It is a *template/CoT warmup slice*, not a knowledge dataset.

## Filtering applied (this is the whole point of the repo)
Starting from `nvidia/Llama-Nemotron-Post-Training-Dataset`:
- **Subset:** `SFT/science` **only**. The `SFT/code` subset was excluded — ~98.6% of its reasoning
  traces exceed the 3000-token budget below, so it would contribute almost nothing under the filter.
  Math subsets were intentionally excluded to avoid adding math practice to the warmup.
- **`reasoning == "on"`** only — long chain-of-thought ("detailed thinking on") examples, so every
  row exercises the thinking region.
- **Length ≤ 3000 tokens**, measured with the **Delphi / Llama-3.1 tokenizer**
  (`marin-community/delphi-3e18-447Mparams-1.2Btokens` vocab) over `input + output`. Conservative cap: the warmup only needs to train the
  template tokens and the think→answer shape, not long reasoning.
- **Uniform random subsample** (reservoir, seed 0) to **100,000** rows.

## Format
- `messages`: ShareGPT-style `[{role, content}]`. The assistant turn contains an inline
  `<think>...</think>` block followed by the answer. The Delphi LLaMA-Factory template normalizes the
  inline `<think>` to the canonical `<|start_think|>`/`<|end_think|>` tokens at encode time.
- `num_tokens`: Delphi-tokenizer length of input+output (≤ 3000).
- `category` (`science`), `reasoning` (`on`), `generator`, `license`, `source`.

## Provenance / license
Derived from `nvidia/Llama-Nemotron-Post-Training-Dataset` (NVIDIA). Per-row `license` values present: ['cc-by-4.0', 'cc-by-sa']. Generators:
['DeepSeek-R1', 'DeepSeek-R1, Qwen-2.5-32B-Instruct', 'DeepSeek-R1, Qwen-2.5-72B-Instruct']. Refer to the source dataset for full licensing/terms; this subsample inherits them.

## Not for
Teaching new knowledge or math skills — it is a small, length-capped slice whose only job is to
install the chat-template reasoning tokens before RL.