text stringlengths 323 957k | uuid stringlengths 9 127 | reasoning_effort stringclasses 3
values | source stringclasses 13
values | license stringclasses 11
values |
|---|---|---|---|---|
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2028-05-07
Reasoning effort: high<|im_end|>
<|im_start|>user
A large pharmaceutical company is considering acquiring a smaller biotech firm that has a promising new drug in late-stage clini... | nemotron_post_v2_stem_reasoning_high_136ac269-9050-44d1-9f3c-04de5cda8574 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-01-15
Reasoning effort: high<|im_end|>
<|im_start|>user
七和駅の歴史年表から、すべての日付とその日の主要な出来事内容を抽出してください。特に、駅の移転、事業譲渡、設備の変更などの重要な事項に注目してください。箇条書き形式に整えて出力してください。
七和駅(ななわえき)は、三重県桑名市大字芳ヶ崎にある、三岐鉄道... | llmjp_extraction_wiki_ja_v0.3_reasoning_high_extraction_wiki202501300506_06-8711 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2028-09-11
Reasoning effort: high<|im_end|>
<|im_start|>user
In this task, you are given two strings A, B. Find the longest common substring in the strings A and B.
One example is below.
Q:... | flan_reasoning_high_flan-NIv2-task600_find_the_longest_common_substring_in_two_strings-fs_opt-19 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2026-07-03
Reasoning effort: medium<|im_end|>
<|im_start|>user
最近、インターネット上の記事で「ESG投資」という言葉をよく見かけますが、これは一体何なのでしょうか?<|im_end|>
<|im_start|>assistant
## ESG投資とは?
**ESG投資(Environment・Social・Go... | llmjp_magpie_sft_v1.0_reasoning_medium_calm3_22b_chat_20241018083433-99105--Qwen2.5_32B_Instruct_20241022115410_2596922 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "bash", "description": "runs the given command directly in bash", "parameters": {"type": "object", "properties": {"command": {"type": "string", "description": "The bash command to execute."}}, "req... | 8140707c-4527-4a03-987e-9608afa1a493 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | MIT |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2024-11-24
Reasoning effort: medium<|im_end|>
<|im_start|>user
Compare and contrast the challenges and opportunities that developing country artists face in the global music industry, consi... | daring_anteater_reasoning_medium_daring_anteater_en-9184 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2026-06-05
Reasoning effort: medium<|im_end|>
<|im_start|>user
説明書によると、南京の雨が強いため、風雨に強い錠を使用することが最も安全です。<|im_end|>
<|im_start|>assistant
<think>
The user says (in Japanese): "According to the... | multiturn_calm3_reasoning_medium_multiturn_calm3-48839 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2028-02-13
Reasoning effort: high<|im_end|>
<|im_start|>user
これから提示する日本語の文章を英語に翻訳してください。必ず英語の訳文のみを出力し、それ以外には何も含めないことを厳守してください。
### 入力:
応神天皇・仲哀天皇・神功皇后(八幡神)を主祭神とし、左相殿に仲恭天皇、右相殿に椎根津彦大神(陶祖神)を配祀... | jaster_v1.4.1_reasoning_high_wikicorpus-j-to-e-17235 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-10-01
Reasoning effort: medium<|im_end|>
<|im_start|>user
文脈と選択肢を入力として受け取り、選択肢から文脈の後に続く文として最も適切なものを選択してください。なお、回答は選択肢の番号(例:0)でするものとします。 回答となる数値をint型で返し、他には何も含めないことを厳守してください。
### 入力:
文... | jaster_v1.4.1_reasoning_medium_kuci-64746 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-04-21
Reasoning effort: high<|im_end|>
<|im_start|>user
これから提示する日本語の文章を英語に翻訳してください。必ず英語の訳文のみを出力し、それ以外には何も含めないことを厳守してください。
### 入力:
澤村宗十郎 (4代目) - 三代目の長男。<|im_end|>
<|im_start|>assistant... | jaster_v1.4.1_reasoning_high_wikicorpus-j-to-e-5172 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "bash", "description": "runs the given command directly in bash", "parameters": {"type": "object", "properties": {"command": {"type": "string", "description": "The bash command to execute."}}, "req... | d1c0ea87-5ce9-4645-b58c-1a089f2b9b6d | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | Apache-2.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2028-07-19
Reasoning effort: medium<|im_end|>
<|im_start|>user
この文章から、各種紀年法で表記されている年を抽出してください。特に、西暦に相当する年、干支、皇紀、修正ユリウス日、リリウス日の情報に注目してください。出力形式:JSON形式
アフリカにおいて当時西欧諸国の植民地であった地域の多数が独立を達成した年であ... | llmjp_extraction_wiki_ja_v0.3_reasoning_medium_extraction_wiki202501300506_03-1133 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-09-05
Reasoning effort: medium<|im_end|>
<|im_start|>user
In this task you are given a sentence pair from wikipedia that have high lexical overlap. If the sentences have the same meani... | flan_reasoning_medium_flan-NIv2-task400_paws_paraphrase_classification-fs_opt-0 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2023-12-31
Reasoning effort: high<|im_end|>
<|im_start|>user
オブジェクト指向プログラミングにおいて、特定の属性に修飾を加える方法について詳述してください。特に、属性のゲッターおよびセッターを作成し、それらを実装する具体的な手順について、PythonやJavaといった代表的なプログラミング言語の例を用いて説明してくだ... | synthetic_jp_en_coding_reasoning_high_synthetic_jp_en_coding-38300 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
生態学における古典的な仮説の一つに、「多様性は安定性を生み出す」というものがあります。食物網の複雑さ、すなわち種の多様性とそれらの間の相互作用の複雑さが、生態系全体の安定性(種の個体数変動の小ささや絶滅への抵抗力など)にどのような影響を与えるのか、その理論的根拠と実証研究に基づいて論じてください。<|im_end|>
<|im_start|>assistant
## 1. ... | e6d03852-e4ee-4ba9-9a87-c198b81cce43 | medium | https://huggingface.co/datasets/nri-ai/nri-fin-reasoning | cc-by-4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "browser.search", "description": "Search the web for information", "parameters": {"type": "object", "properties": {"query": {"type": "string"}, "topn": {"type": "integer"}, "source": {"type": "stri... | seed_42_7840 | high | https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Dataset | mit |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-04-16
Reasoning effort: high<|im_end|>
<|im_start|>user
セルゲイ・パラジャーノフの生涯に関する文章から、彼が生まれた場所と出生・死亡年を抽出してください。出力形式:YAML形式
セルゲイ・パラジャーノフ (ラテン文字転写:Sargis Hovsepi Parajanyan, , , 1924年1月9日 - 1... | llmjp_extraction_wiki_ja_v0.3_reasoning_high_extraction_wiki202501300506_06-3595 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2028-12-24
Reasoning effort: low<|im_end|>
<|im_start|>user
In this task, you're given a statement, and three sentences as choices. Your job is to determine which sentence clearly disagrees... | flan_reasoning_low_flan-NIv2-task202_mnli_contradiction_classification-zs_opt-2 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | 8dc17d14-74ad-44e8-81b1-43de4e915885 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | MIT |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | 669f010c-254b-47b9-9297-456e242207f5 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | BSD-2-Clause |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | 15edd233-0a88-490a-8708-999929055e09 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | Apache-2.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2028-02-04
Reasoning effort: medium<|im_end|>
<|im_start|>user
Formulate a Sustainable Food System Initiative for Local Communities, emphasizing environmentally responsible agriculture prac... | daring_anteater_reasoning_medium_daring_anteater_en-64965 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-08-05
Reasoning effort: medium<|im_end|>
<|im_start|>user
What strategies can be employed to handle the trade-off between simplicity and complexity in large-scale programming projects ... | synthetic_jp_en_coding_reasoning_medium_synthetic_jp_en_coding-221093 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Draft an executive summary for a new 'Green Champions' employee engagement program, detailing the proposed rollout strategy across our global offices. The summary must be under 250 words a... | 2cfa4713-75c7-4b2a-9ec8-a22de14de47d | medium | https://huggingface.co/datasets/nri-ai/nri-fin-reasoning | cc-by-4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "authenticate_by_member", "description": "Authenticate member using Member ID and last name for account actions", "parameters": {"properties": {"member_id": {"description": "Member's unique identif... | d6715469-0529-b0b0-8748-6bd12e18ba04 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Agentic-v2 | cc-by-4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "request_authentication_code", "description": "Send authentication code via SMS/email using either Customer ID + registered email OR Full name + service address + ZIP code", "parameters": {"propert... | 23f1dcc1-81c2-8a01-e985-eea6b0cb65c9 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Agentic-v2 | cc-by-4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "arc_python_executor", "description": "Executes Python code to transform grids. Code is tested on all training examples, then applied to test input. Use this for algorithmic solutions involving rot... | b20e6436-9c05-40a0-973c-ab2e28d6a58c | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-ARC-AGI-v1 | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-02-17
Reasoning effort: high<|im_end|>
<|im_start|>user
Reexamine the aforementioned C++ function that iteratively determines the sequence of Lucas numbers:
```cpp
int lucas(int n) {
... | synthetic_jp_en_coding_reasoning_high_synthetic_jp_en_coding-521274 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-12-14
Reasoning effort: medium<|im_end|>
<|im_start|>user
Solve the following math problem. Make sure to put the answer (and only answer) inside \boxed{}.
Given discrete variables \(a... | nemotron_post_v3_math_reasoning_medium_7c52986d-2e1f-4cd4-8a43-24d2ceb54d21 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Answer the following multiple choice question. The last line of your response should be in the following format: 'Answer: A/B/C/D' (e.g. 'Answer: A').
What is the minimum force required t... | a6f9423a3b0946d9b198915fc6658d18 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Science-v2 | CC BY-SA 4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: high<|im_end|>
<|im_start|>user
次の数学の問題を解いてください。答え(答えのみ)を\boxed{}の中に入れてください。
直角三角形 \(ABC\) において、\(AC = b\)、\(BC = a\) とします。点 \(C\) を中心とする円が斜辺 \(AB\) に点 \(D\) で接しています。点 \(A\) と点 \(B\) を通る円の接線がそれぞれ円と交わる点を \(E\)、\(F\) とします。線... | 272cfc92-9f15-5bb2-9360-57a951f87ae2 | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Multilingual-v2 | cc-by-sa-4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | 3f51b69e-9bcc-44a7-aeac-8e9b3cbda0b1 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | Apache-2.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: high<|im_end|>
<|im_start|>user
与えられた正N角形の頂点は(cos(2πi / N), sin(2πi / N)), ∀ i ∈ [0,N-1]に位置しています。このN角形は直径100 mmの回転する鋼鉄シャフトの断面を表しており、250 kNの軸方向荷重を伝達します。シャフトは切欠き感度指数0.5、理論的応力集中係数2.1の鋼鉄で作られています。設置時の不整列により、完全に逆転した曲げ応力が加わります。鋼鉄... | 2fbed541-a87f-5e3d-8e36-d9e59180cd2c | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Multilingual-v2 | cc-by-4.0 |
<|im_start|>system
You are a helpful and harmless assistant.
You are not allowed to use any tools.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Assist me in drafting a letter to my relatives in Havana, emphasizing the importance of financial literacy and budgeting in today's economy:
1. Your response should co... | 4be936ac-0159-42b9-aeab-9de5c32ec2de | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Instruction-Following-Chat-v3 | cc-by-4.0 AND odc-by |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2026-01-15
Reasoning effort: low<|im_end|>
<|im_start|>user
高校のコンピュータサイエンスクラブでは、アリス、ボブ、キャロル、デイブの四人の生徒がエラトステネスの篩アルゴリズムを使用して素数リストを生成するPythonプログラムを最適化するという課題に取り組んでいます。彼らはコードの効率を向上させるために作業を複数の段... | logical_math_coding_wizard8x22b_reasoning_low_logical_math_coding_wizard8x22b-59417 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are a helpful and harmless assistant.
You are not allowed to use any tools.
Reasoning effort: medium<|im_end|>
<|im_start|>user
As a Brazilian football coach who is passionate about the tactical aspects of the game and the unique style of South American football, create a comprehensive analysis... | 10cb6f97-871a-4b93-9807-feb3e9b46378 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Instruction-Following-Chat-v3 | cc-by-4.0 AND odc-by |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "browser.search", "description": "Search the web for information", "parameters": {"type": "object", "properties": {"query": {"type": "string"}, "topn": {"type": "integer"}, "source": {"type": "stri... | seed_48_2035 | high | https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Dataset | mit |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
金融機関Aは、TCFDの「リスク管理とガバナンス」開示に際し、気候関連リスクの統合プロセスを以下のように設計した。あなたは、提供された情報をもとに、最も不適切な統合手法を特定し、その理由を論理的に説明してください(最低300字)。
- ポートフォリオ全体のリスクエクスポージャーを算出する際、スコープ1・2・3のCO₂排出量をベースに、シナリオ分析(2℃シナリオ、4℃シナ... | 5a223eae-ec85-425d-8018-4583826c4ef6 | medium | https://huggingface.co/datasets/nri-ai/nri-fin-reasoning | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2028-02-23
Reasoning effort: high<|im_end|>
<|im_start|>user
文脈と選択肢を入力として受け取り、選択肢から文脈の後に続く文として最も適切なものを選択してください。なお、回答は選択肢の番号(例:0)でするものとします。 回答となる数値をint型で返し、他には何も含めないことを厳守してください。
### 入力:
文脈:... | jaster_v1.4.1_reasoning_high_kuci-48573 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Answer the following multiple choice question. The last line of your response should be in the following format: 'Answer: A/B/C/D' (e.g. 'Answer: A').
What term is missing in the derivati... | 006edb9f82e946bfada3e6b547d4c646 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Science-v2 | CC BY-SA 4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "browser.search", "description": "Search the web for information", "parameters": {"type": "object", "properties": {"query": {"type": "string"}, "topn": {"type": "integer"}, "source": {"type": "stri... | seed_55_6449 | high | https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Dataset | mit |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | 79936f05-47d7-468a-8930-66aac2449c59 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | Apache-2.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2026-12-27
Reasoning effort: high<|im_end|>
<|im_start|>user
Solve the following math problem. Make sure to put the answer (and only answer) inside \boxed{}.
Calculate the average roll of ... | nemotron_post_v3_math_reasoning_high_a9314390-4e10-42f9-9ab9-0ac7b8c303a6 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "arc_python_executor", "description": "Executes Python code to transform grids. Code is tested on all training examples, then applied to test input. Use this for algorithmic solutions involving rot... | 6eb0055b-5b30-4a0f-9bd7-50f90e7b3688 | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-ARC-AGI-v1 | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Financial statement analysis is a cornerstone of fundamental analysis for long-term equity investing. Elaborate on its primary objectives, which include evaluating past performance, curren... | 6f386797-ac89-4c52-97e9-4736f313cf7b | medium | https://huggingface.co/datasets/nri-ai/nri-fin-reasoning | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2029-07-21
Reasoning effort: medium<|im_end|>
<|im_start|>user
自分のブログやサイトを作るためには、どんな知識やスキルが必要ですか?<|im_end|>
<|im_start|>assistant
<think>
The user asks in Japanese: "What knowledge and skil... | llmjp_magpie_sft_v1.0_reasoning_medium_calm3_22b_chat_20241022155627--22943--Qwen2.5_32B_Instruct_20241024144245_671773 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "arc_python_executor", "description": "Executes Python code to transform grids. Code is tested on all training examples, then applied to test input. Use this for algorithmic solutions involving rot... | 7df468e4-c3f6-428f-a9a0-dcd5fd9250e3 | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-ARC-AGI-v1 | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2028-02-11
Reasoning effort: low<|im_end|>
<|im_start|>user
Here’s a brief recap of the most recent developments:
- The team completed the core backend integration two days ahead of schedu... | nemotron_post_v3_if_reasoning_low_0f465c9e-17dd-41d7-aaa1-a4b7e87c258d | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2029-09-26
Reasoning effort: high<|im_end|>
<|im_start|>user
あなたは、一節ごとに次の韻を踏んだ構成の5連詩を書く: [A] 私は猫が好きです。 [B} 私は海外へ行ってしまいました。 [A] 私は愛する人を見失いました。 [B] 私は寂しい夜の星を祈ります。 上記の詩は、ABAB韻スキーム(非常に簡単な言葉で)とラ... | multiturn_calm3_reasoning_high_multiturn_calm3-13912 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: high<|im_end|>
<|im_start|>user
コスチャ博士は多段階の結晶化プロセスを最適化しています。目的は、最小時間内に少なくともs個の精製された結晶を得ることです。時間0で0個の結晶から始まり、アクティブなプロセスはありませんが、n個の実験プロトコルから選択できます。i番目のプロトコルは初期化にci個の結晶を消費し(溶媒、試薬、装置の組み立てを含む)、その後毎秒vi個の結晶を生成します。実験室の制約により、一度にアクテ... | c2232026-6cf8-5dda-acbf-e9d16fb16c9b | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Multilingual-v2 | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2025-08-01
Reasoning effort: medium<|im_end|>
<|im_start|>user
村上春樹の『ノルウェイの森』は、彼の他の作品と比べてどのような特徴があるかを教えてください。<|im_end|>
<|im_start|>assistant
<think>
The user asks in Japanese: "Tell me wha... | llmjp_magpie_sft_v1.0_reasoning_medium_calm3_22b_chat_20241022155627--99849--Qwen2.5_32B_Instruct_20241024144245_2965519 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Answer the following multiple choice question. The last line of your response should be in the following format: 'Answer: A/B/C/D' (e.g. 'Answer: A').
Does an object moving at a constant ... | 88201558076045e3bb8e4365de73718d | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Science-v2 | CC BY-SA 4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2024-08-05
Reasoning effort: medium<|im_end|>
<|im_start|>user
アナ、今日の天気はどうですか?<|im_end|>
<|im_start|>assistant
<think>
The user asks in Japanese: "Ana, what's the weather today?" They addre... | multiturn_calm3_reasoning_medium_multiturn_calm3-44517 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "bash", "description": "runs the given command directly in bash", "parameters": {"type": "object", "properties": {"command": {"type": "string", "description": "The bash command to execute."}}, "req... | 8fb72260-3fee-4ec8-aff4-e1545949daf4 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | Apache-2.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2026-05-05
Reasoning effort: medium<|im_end|>
<|im_start|>user
Is Great Pyramid of Giza the last wonder of its kind? Given the above question, please answer with reasoning first!<|im_end|>
... | flan_reasoning_medium_flan-CoT-cot_strategyqa-zs_opt-7 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "bash", "description": "runs the given command directly in bash", "parameters": {"type": "object", "properties": {"command": {"type": "string", "description": "The bash command to execute."}}, "req... | e4d98794-1b43-4b8e-a74a-52480d63ec12 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | MIT |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "search_adoptable_animals", "description": "Search adoptable animals based on filters such as species, breed, age, or animal ID", "parameters": {"properties": {"animal_id": {"description": "Specifi... | a20b2d55-7f48-b109-0eaa-3e4ca0ae1769 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Agentic-v2 | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-05-21
Reasoning effort: low<|im_end|>
<|im_start|>user
与えられた文章から固有表現で書かれたターゲットの名前を抽出し、それに対する極性をpositive、neutral、negativeの中から選択して下さい。固有表現で書かれたターゲットの名前と、それに対する極性(positive、neutral、negativ... | jaster_v1.4.1_reasoning_low_chabsa-662 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: high<|im_end|>
<|im_start|>user
次の数学の問題を解きなさい。答え(答えのみ)を\boxed{}の中に入れること。
代数数 \(a \ge 0\) が与えられたとき、次の等式を満たす有理数 \(b\) を求めよ。
\[
\arctan(a) = \int_0^a \frac{dx}{1+x^2} = \pi b.
\]<|im_end|>
<|im_start|>assistant
<think>
We ne... | e6e35d8e-90af-5d91-9a73-dac709f9215b | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Multilingual-v2 | cc-by-sa-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2026-09-01
Reasoning effort: medium<|im_end|>
<|im_start|>user
Please answer the following question: The exercise is to decide whether the question accepts the proposed suggestion as a corr... | flan_reasoning_medium_flan-P3-wiki_qa_exercise-zs_noopt-10 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-05-27
Reasoning effort: low<|im_end|>
<|im_start|>user
この文章から、埼玉縣信用金庫の正式名称、略称、本部・本店の所在地、および2016年に移転した一部本部機能の新所在地を抽出してください。出力形式:リスト形式
埼玉縣信用金庫(さいたまけんしんようきんこ、英語:The Saitamaken Shinkin Ba... | llmjp_extraction_wiki_ja_v0.3_reasoning_low_extraction_wiki202501300506_02-2733 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
A monocationic gold complex with a single water ligand, Au⁺·(H₂O), is investigated using a relativistic effective core potential (RECP) for Au combined with a large, polarized basis set fo... | b52df3bfce3a4345bfbb6aadf5a60fb2 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Science-v2 | CC BY-SA 4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2024-07-03
Reasoning effort: low<|im_end|>
<|im_start|>user
筆を執るたびに、手に汗をかくことが気になっているのですが、何か良い対策はないでしょうか?<|im_end|>
<|im_start|>assistant
### 手が汗ばむ(多汗症)への対策まとめ
---
## 1. 基本的な生活習慣の見直し
|... | llmjp_magpie_sft_v1.0_reasoning_low_calm3_22b_chat_20241022133932--11701--Qwen2.5_32B_Instruct_20241024100350_323636 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "bash", "description": "runs the given command directly in bash", "parameters": {"type": "object", "properties": {"command": {"type": "string", "description": "The bash command to execute."}}, "req... | 2bde227a-5a37-48f6-bb6b-ab5ba3893269 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | MIT |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Answer the following multiple choice question. The last line of your response should be in the following format: 'Answer: A/B/C/D/E/F/G/H/I/J' (e.g. 'Answer: A').
Does an increase in curr... | 70fd677b23104764be58032477cc8e9d | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Science-v2 | CC BY-SA 4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "bash", "description": "runs the given command directly in bash", "parameters": {"type": "object", "properties": {"command": {"type": "string", "description": "The bash command to execute."}}, "req... | a2aa9596-1956-48d0-aac9-2e1d9142af55 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | Apache-2.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-07-28
Reasoning effort: medium<|im_end|>
<|im_start|>user
ボルチモア・オリオールズの歴史的な優勝情報を抽出してください。具体的に、ワールドシリーズ優勝年、リーグ優勝年、地区優勝年、ワイルドカード進出年をリストアップしてください。
{{MLBのチーム
| チーム名 = ボルチモア・オリオールズ
| 英語表記... | llmjp_extraction_wiki_ja_v0.1_reasoning_medium_extraction_wiki_mSV8LRgpus_20250127--UTVcHZUpK8_20250127-5640 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: high<|im_end|>
<|im_start|>user
以下の数学の問題を解いてください。答え(答えのみ)を \boxed{} の中に入れてください。
整数 \(a \geq b \geq c > 0\) を見つけて、
\[ a + b + c + ab + bc + ca = abc + 1 \]
を満たすようにしてください。<|im_end|>
<|im_start|>assistant
<think>
We need t... | 197632f5-b71c-50fb-8d8f-69920b4f025c | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Multilingual-v2 | cc-by-sa-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Answer the following multiple choice question. The last line of your response should be in the following format: 'Answer: A/B/C/D' (e.g. 'Answer: A').
What happens to the electrical field... | f47d97f7985d4af49859e0415c24de55 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Science-v2 | CC BY-SA 4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Answer the following multiple choice question. The last line of your response should be in the following format: 'Answer: A/B/C/D/E/F/G/H/I/J' (e.g. 'Answer: A').
Which book is recommende... | 42af3b48449f429bbc5b119b2979f30e | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Science-v2 | CC BY-SA 4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2024-11-02
Reasoning effort: low<|im_end|>
<|im_start|>user
超音波装置を利用した病態のスクリーニングを行っていますが、検査結果に誤差が生じる原因としてはどのようなものが考えられますか?<|im_end|>
<|im_start|>assistant
## 超音波(エコー)装置で得られるスクリーニング結果に「誤差」が入... | llmjp_magpie_sft_v1.0_reasoning_low_calm3_22b_chat_20241022155627--22398--Qwen2.5_32B_Instruct_20241024144245_655422 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2029-06-23
Reasoning effort: medium<|im_end|>
<|im_start|>user
Schema:
{
"name": "JewelrySelectionProfile",
"schema": {
"type": "object",
"required": [
"personalStyle",
... | nemotron_post_v3_if_reasoning_medium_dfc8368c-a838-4db0-b5c8-3f6b7bd676cd | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-12-03
Reasoning effort: medium<|im_end|>
<|im_start|>user
Solve the following math problem. Make sure to put the answer (and only answer) inside \boxed{}.
One zero of the polynomial \... | nemotron_post_v3_math_reasoning_medium_c2f5be57-3798-490e-ad56-4a51e822de99 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-05-20
Reasoning effort: low<|im_end|>
<|im_start|>user
質問:無向グラフが与えられ、エッジのリストと正の整数 `k` がある場合、そのグラフ内にちょうど `k` 本のエッジを持つサイクルが存在するかどうかを判断するPython関数を書いてください。関数はそのようなサイクルが存在すれば `True` を返し、存在し... | logical_math_coding_wizard8x22b_reasoning_low_logical_math_coding_wizard8x22b-26515 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "authenticate_user", "description": "Authenticate user using company ID and employee email. Required before accessing or modifying company, employee, or enrollment data.", "parameters": {"propertie... | 19e88837-8099-f196-5535-5d671faad0f8 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Agentic-v2 | cc-by-4.0 |
<|im_start|>system
You are a helpful and harmless assistant.
You are not allowed to use any tools.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Draft a 6-paragraph legal memorandum addressing the implications of a recent Supreme Court ruling on biotech patent eligibility. Ensure that the third paragraph begins... | 559fc65b-06af-49bc-8cd2-14c032323ebd | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Instruction-Following-Chat-v3 | cc-by-4.0 AND odc-by |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2024-11-30
Reasoning effort: medium<|im_end|>
<|im_start|>user
過去一時間におけるChatGPTが送信したメッセージの数を表すリストがあるとします。このリストは次の通りです:[34, 12, 45, 7, 89, 23, 67]。マージソートアルゴリズムを使用して、ChatGPTがこのリストを昇順に並べ替える手順を... | logical_math_coding_wizard8x22b_reasoning_medium_logical_math_coding_wizard8x22b-134740 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | 18d14ad5-c423-428f-9aec-50fc5bec4005 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | Apache-2.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | f8c6caea-90a3-4ba3-98e4-c84ac07f09b5 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | MIT |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: high<|im_end|>
<|im_start|>user
以下の数学の問題を解きなさい。解答(解答のみ)を \boxed{} の中に入れることを忘れないでください。
\( n \) 人の小人がおり、それぞれが \( k \) 種類の可能な色の帽子をかぶっています。各小人は、自分の帽子の色を除き、自分と同じ色の帽子の数を知っています。各小人は自分の名前と色を紙に書いて箱に入れます。"訂正小人"は紙だけを使って、各小人を自分の帽子の色と... | 59923fc1-f585-555b-8444-749823c74b6b | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Multilingual-v2 | cc-by-sa-4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | 4aa1c40b-529c-4615-8bb5-4e71a6629b87 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | Apache-2.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2025-12-07
Reasoning effort: medium<|im_end|>
<|im_start|>user
いじめを受けていると感じるとき、その理由や背景についてもう少し具体的に知りたいと思っています。いじめられる側に問題があるという意見もありますが、それは正しいと思いますか?<|im_end|>
<|im_start|>assistant
## 1. 「い... | llmjp_magpie_sft_v1.0_reasoning_medium_calm3_22b_chat_20241022155627--30988--Qwen2.5_32B_Instruct_20241024144245_905265 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "bash", "description": "runs the given command directly in bash", "parameters": {"type": "object", "properties": {"command": {"type": "string", "description": "The bash command to execute."}}, "req... | 2a6aefbb-1b60-47fa-b18d-f677111fad71 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | MIT |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-03-17
Reasoning effort: low<|im_end|>
<|im_start|>user
コーディング問題:公平なキャンディ分けゲーム
アリスとボブはテーブルに置かれたキャンディの列でゲームをしています。列には合計で \( n \) 個のキャンディがあり、それぞれのキャンディはその風味の良さを表す整数値で表されています。ゲームは左から右に並ん... | logical_math_coding_wizard8x22b_reasoning_low_logical_math_coding_wizard8x22b-136060 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-05-04
Reasoning effort: high<|im_end|>
<|im_start|>user
前提と仮説の関係をentailment、contradiction、neutralの中から回答してください。それ以外には何も含めないことを厳守してください。
制約:
- 前提から仮説が、論理的知識や常識的知識を用いて導出可能である場合はentailmen... | jaster_v1.4.1_reasoning_high_jnli-15670 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Reasoning effort: medium<|im_end|>
<|im_start|>user
経営陣に対して、金利上昇リスクに関する理解を深め、効果的な対策を議論するための資料を作成してほしい。当社の負債ポートフォリオの金利感応性(デュレーション/コンベキシティ)を計算し、ベンチマーク金利が2%上昇した場合の財務諸表(特に利息費用と純利益)への影響を定量化した上で、ヘッジ手法を含む実行可能なリスク軽減オプションを提案する報告書を提出してください。<|im_end|>... | 961fbc21-88a8-450c-83ba-bb7e53fa95c5 | medium | https://huggingface.co/datasets/nri-ai/nri-fin-reasoning | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2022-09-14
Reasoning effort: medium<|im_end|>
<|im_start|>user
逆光補正で有用なアルゴリズムはある?<|im_end|>
<|im_start|>assistant
## 逆光(バックライト)画像の補正に使える代表的アルゴリズムまとめ
| カテゴリ | アルゴリズム名 | 主なアイデア / 手順 | 長所 |... | multiturn_calm3_reasoning_medium_multiturn_calm3-56980 | medium | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-08-31
Reasoning effort: low<|im_end|>
<|im_start|>user
前提と仮説の関係をentailment、contradiction、neutralの中から回答してください。それ以外には何も含めないことを厳守してください。
制約:
- 前提から仮説が、論理的知識や常識的知識を用いて導出可能である場合はentailment... | jaster_v1.4.1_reasoning_low_jnli-1287 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "arc_python_executor", "description": "Executes Python code to transform grids. Code is tested on all training examples, then applied to test input. Use this for algorithmic solutions involving rot... | bb7b10da-a67b-4549-84c8-0db55770cc69 | high | https://huggingface.co/datasets/nvidia/Nemotron-SFT-ARC-AGI-v1 | cc-by-4.0 |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2027-02-26
Reasoning effort: high<|im_end|>
<|im_start|>user
How much funding will the Washington Apple Commission be receiving from the USDA Agricultural Trade Promotion (ATP) Program? Ans... | daring_anteater_reasoning_high_daring_anteater_en-83088 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "authenticate_customer", "description": "Authenticates customer using Customer ID and last four SSN digits. Required before any account-sensitive operations.", "parameters": {"properties": {"custom... | b3ef2261-60e4-abee-09f2-04f8649592bf | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Agentic-v2 | cc-by-4.0 |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "browser.search", "description": "Search the web for information", "parameters": {"type": "object", "properties": {"query": {"type": "string"}, "topn": {"type": "integer"}, "source": {"type": "stri... | seed_56_940 | high | https://huggingface.co/datasets/OpenResearcher/OpenResearcher-Dataset | mit |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "bash", "description": "runs the given command directly in bash", "parameters": {"type": "object", "properties": {"command": {"type": "string", "description": "The bash command to execute."}}, "req... | aa0605e7-b939-4ac5-9ed0-3f3bb50493a5 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | MIT |
<|im_start|>system
You are a helpful and harmless assistant.
You are not allowed to use any tools.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Help me write a 3-section feature article about the surprising moments in the retired player's career, capturing the essence of their personality and achievements. The... | e993ff23-71a1-431c-b1bb-52eaa7db16d0 | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Instruction-Following-Chat-v3 | cc-by-4.0 AND odc-by |
<|im_start|>system
# ツール
タスクを解決するために、1つまたは複数のツールを呼び出すことができます。
ツールのシグネチャは、<tools></tools> XMLタグ内に記載されています。
<tools>
{"name": "execute_bash", "description": "Execute a bash command in the terminal within a persistent shell session.\n\n\n### Command Execution\n* One command at a time: You can only execute one bash command ... | 5ea5576e-2adf-465d-8bd2-5162294cdcf6 | medium | https://huggingface.co/datasets/nvidia/Open-SWE-Traces | MIT |
<|im_start|>system
You are a helpful and harmless assistant.
You are not allowed to use any tools.
Reasoning effort: medium<|im_end|>
<|im_start|>user
Draft a detailed negotiation guide for contract discussions, formatted into 5 sections, each marked with Section N. The guide should include essential negotiation stra... | 454ad273-9e34-4ec3-b3ab-ee23e9b8693d | medium | https://huggingface.co/datasets/nvidia/Nemotron-SFT-Instruction-Following-Chat-v3 | cc-by-4.0 AND odc-by |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-08-22
Reasoning effort: high<|im_end|>
<|im_start|>user
Given a text passage, come up with an appropriate title for it. The title should accurately represent what the main focus of the... | flan_reasoning_high_flan-NIv2-task769_qed_summarization-zs_opt-2 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2022-07-18
Reasoning effort: low<|im_end|>
<|im_start|>user
色とりどりのブロック(赤、青、緑)が混ざった山があり、サイズもさまざま(小、中、大)であると想像してください。まずサイズで分け、その後各サイズのリスト内で色でさらに分類したいと思います。手でこれらのブロックを並べ替えるような簡単な方法で、Pythonプログラ... | logical_math_coding_wizard8x22b_reasoning_low_logical_math_coding_wizard8x22b-100447 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2030-01-29
Reasoning effort: low<|im_end|>
<|im_start|>user
Which of the following strategies is most effective in enhancing customer satisfaction in a highly competitive market environment... | nemotron_post_v2_stem_reasoning_low_6bb32332-1b03-4f83-9b08-e5325ddd9485 | low | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |
<|im_start|>system
You are LLM-jp-4, a large language model trained by LLM-jp.
Knowledge cutoff: 2025-12
Conversation start date: 2031-01-22
Reasoning effort: high<|im_end|>
<|im_start|>user
孝元皇后に関する記述から、各孝元皇后の時代(例えば、「前漢」「後漢」「明の泰昌帝」など)とその名前(例えば、「王政君」「孝元貞皇后」など)を抽出してください。
孝元皇后(こうげんこうごう)は、中国の皇后の諡号。
一覧
王政君 - 前漢の元帝の皇后。
... | llmjp_extraction_wiki_ja_v0.2_reasoning_high_extraction_wiki202501300506_02-7113 | high | https://huggingface.co/datasets/llm-jp/llm-jp-4-thinking-sft-data | multiple; see dataset card |