instruction stringlengths 45 176 | input stringlengths 249 1.12k | output stringlengths 382 884 | source stringclasses 1
value | original_id stringlengths 8 44 |
|---|---|---|---|---|
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"fka/prompts.chat","name":"prompts.chat","description":"a.k.a. Awesome ChatGPT Prompts. This is a Dataset Repository mirror of prompts.chat — a social platform for AI prompts. prompts.chat is an open-source platform where users can share, discover, and collect AI prompts from the community.... | prompts.chat is a community-curated prompt collection mirrored from the prompts.chat platform and the Awesome ChatGPT Prompts project. It is best suited for prompt engineering research, instruction template discovery, prompt retrieval, and analysis of popular prompting patterns across major LLM ecosystems such as ChatG... | huggingface | fka/prompts.chat |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"HuggingFaceFW/fineweb","name":"fineweb","description":"FineWeb: 15 trillion tokens of the finest data the web has to offer. The dataset consists of more than 18.5T tokens of cleaned and deduplicated English web data from CommonCrawl. The data processing pipeline is optimized for LLM perfor... | FineWeb is a massive English web-scale corpus derived from CommonCrawl and processed for high-quality LLM pretraining. Its primary use case is large-scale base model training for text generation, especially next-token prediction. The dataset emphasizes cleaning, deduplication, and pipeline quality, making it suitable f... | huggingface | HuggingFaceFW/fineweb |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"Anthropic/hh-rlhf","name":"hh-rlhf","description":"Dataset Card for HH-RLHF. This repository provides access to human preference data about helpfulness and harmlessness for training preference or reward models for subsequent RLHF training. These data are not meant for supervised training o... | HH-RLHF is a human feedback dataset designed for alignment research, especially reward modeling and preference learning around helpfulness and harmlessness. Its core use case is RLHF pipelines, where models learn from ranked or preferred responses rather than standard supervised targets. The dataset card explicitly war... | huggingface | Anthropic/hh-rlhf |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"Open-Orca/OpenOrca","name":"OpenOrca","description":"The OpenOrca dataset is a rich collection of augmented FLAN data aligned as closely as possible with the distributions outlined in the Orca paper. It has been instrumental in generating high-performing model checkpoints and serves as a r... | OpenOrca is a large-scale instruction-style dataset built from augmented FLAN data and inspired by the Orca family of training distributions. It is well suited for supervised fine-tuning of general-purpose assistants because it covers many NLP task types, including question answering, summarization, classification, tab... | huggingface | Open-Orca/OpenOrca |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"OpenAssistant/oasst1","name":"oasst1","description":"OpenAssistant Conversations Dataset (OASST1). A human-generated, human-annotated assistant-style conversation corpus consisting of 161,443 messages in 35 different languages, annotated with 461,292 quality ratings and over 10,000 fully a... | OASST1 is a multilingual assistant conversation dataset created through human generation and annotation, with conversation trees and extensive quality ratings. It is highly suitable for supervised fine-tuning of chat assistants, multilingual dialogue modeling, response ranking, and research on conversational alignment.... | huggingface | OpenAssistant/oasst1 |
Summarize the dataset in a concise SFT-friendly format, including domain, modality, main task, language, size, license, and notable characteristics. | Name: gsm8k
Description: Dataset Card for GSM8K
Dataset Summary
GSM8K (Grade School Math 8K) is a dataset of 8.5K high quality linguistically diverse grade school math word problems. The dataset was created to support the task of question answering on basic mathematical problems that require multi-step reasoning.
Thes... | GSM8K is an English text dataset of grade school math word problems designed for multi-step reasoning and question answering. It contains around 8.5K high-quality problems, generally requiring 2 to 8 reasoning steps using basic arithmetic. The dataset is original, crowdsourced, monolingual, MIT-licensed, and commonly u... | huggingface | openai/gsm8k |
Summarize the dataset in a concise SFT-friendly format, including domain, modality, main task, language coverage, size, license, and notable characteristics. | Name: wikipedia
Description: Dataset Card for Wikimedia Wikipedia
Dataset Summary
Wikipedia dataset containing cleaned articles of all languages. The dataset is built from the Wikipedia dumps with one subset per language, each containing a single train split. Each example contains the content of one full Wikipedia art... | Wikimedia Wikipedia is a massive multilingual text dataset consisting of cleaned full Wikipedia articles across a very large number of languages. It is intended for language modeling and masked language modeling tasks, with one language subset per language and a train split containing processed article text. The datase... | huggingface | wikimedia/wikipedia |
Summarize the dataset or asset in a concise SFT-friendly format, including modality, intended use, scale, license, and notable limitations if apparent. | Name: EasyNegative
Description: Negative Embedding
This is a Negative Embedding trained with Counterfeit. Please use it in the "stable-diffusion-webui/embeddings" folder. It can be used with other models, but the effectiveness is not certain.
Referenced models: Counterfeit-V2.0.safetensors, AbyssOrangeMix2_sfw.safetens... | EasyNegative is an image-related negative embedding asset intended for use with Stable Diffusion-style workflows, especially via the embeddings folder in stable-diffusion-webui. It was trained with Counterfeit and may also work with other models, though effectiveness outside the original setup is uncertain. The asset i... | huggingface | gsdf/EasyNegative |
Summarize the dataset in a concise SFT-friendly format, including domain, modality, main task, language, scale, and notable characteristics. | Name: RedPajama-Data-1T
Description: RedPajama is a clean-room, fully open-source implementation of the LLaMa dataset.
Tags: task_categories:text-generation, language:en, size_categories:1M<n<10M, modality:text, region:us
Downloads: 2123
Likes: 1151 | RedPajama-Data-1T is a large English text dataset created as a clean-room, fully open-source reproduction of the data mixture used for LLaMA-style model training. It is designed for text generation and large-scale language model pretraining. The dataset is text-only, large in scale, and notable for its open-source posi... | huggingface | togethercomputer/RedPajama-Data-1T |
Summarize the dataset in a concise SFT-friendly format, including domain, modality, tasks, languages, license, size, and notable characteristics. | Name: medical-o1-reasoning-SFT
Description: News
[2025/04/22] We split the data and kept only the medical SFT dataset (medical_o1_sft.json). The file medical_o1_sft_mix.json contains a mix of medical and general instruction data.
[2025/02/22] We released the distilled dataset from Deepseek-R1 based on medical verifiabl... | medical-o1-reasoning-SFT is a bilingual English-Chinese medical text dataset for supervised fine-tuning, focused on medical reasoning over verifiable problems. It supports both question answering and text generation, and includes reasoning-oriented supervision distilled from DeepSeek-R1. The dataset is distributed in J... | huggingface | FreedomIntelligence/medical-o1-reasoning-SFT |
Summarize the dataset for LLM pretraining use, including its scale, domain, language, license, and notable metadata. | Name: fineweb-edu
Description: FineWeb-Edu is a 1.3 trillion token educational web dataset filtered from FineWeb using an educational quality classifier. It is the 1.3T token version and is intended for high-quality educational text collection.
Tags: task_categories:text-generation, language:en, license:odc-by, size_ca... | FineWeb-Edu is a large-scale English educational text dataset from Hugging Face intended for text generation and LLM pretraining. It contains about 1.3 trillion tokens of educational web content filtered from FineWeb using a quality classifier informed by Llama3-70B-Instruct annotations. The dataset is distributed in p... | huggingface | HuggingFaceFW/fineweb-edu |
Provide a concise dataset overview covering purpose, token scale, language, license, and relevance for language model training. | Name: dolma
Description: Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research
Tags: task_categories:text-generation, language:en, license:odc-by, size_categories:n>1T, arxiv:2402.00159, arxiv:2301.13688, region:us, language-modeling, casual-lm, llm
Downloads: 3087
Likes: 1021 | Dolma is an open English corpus designed for language model pretraining research. It contains approximately three trillion tokens, making it one of the largest openly available text-generation datasets. The dataset is licensed under ODC-BY and is explicitly positioned for language modeling, causal language modeling, an... | huggingface | allenai/dolma |
Summarize this dataset with emphasis on content type, multilingual coverage, licensing characteristics, and suitability for code model training. | Name: the-stack
Description: The Stack is a large code dataset. Initial release included 30 programming languages and 18 permissive licenses, producing a near-deduplicated 3TB dataset. Later versions excluded weak copyleft licenses and expanded the permissive license list to 193 licenses.
Tags: task_categories:text-gen... | The Stack is a large multilingual code dataset built for text generation and code model training. It aggregates source code from many programming languages, originally covering 30 languages and later expanding licensing coverage while removing weak copyleft licenses. The dataset is near-deduplicated, distributed in par... | huggingface | bigcode/the-stack |
Create a short summary describing the dataset’s content, generation method, scale, license, and practical use in model training. | Name: TinyStories
Description: Dataset containing synthetically generated short stories created by GPT-3.5 and GPT-4 using a small vocabulary. It is described in arXiv:2305.07759 and includes train and validation files for model training.
Tags: task_categories:text-generation, language:en, license:cdla-sharing-1.0, siz... | TinyStories is an English text-generation dataset of synthetically generated short stories created with GPT-3.5 and GPT-4. The stories use a constrained vocabulary, making the dataset useful for studying small language models, language acquisition behavior, and efficient training setups. It is relatively small in size,... | huggingface | roneneldan/TinyStories |
Summarize the dataset as an instruction-tuning resource, covering task types, data origin, size, format, and license. | Name: databricks-dolly-15k
Description: databricks-dolly-15k is an open source dataset of instruction-following records generated by thousands of Databricks employees across behavioral categories such as brainstorming, classification, closed QA, generation, information extraction, open QA, and summarization.
Tags: task... | databricks-dolly-15k is an English instruction-following dataset designed for supervised fine-tuning of assistant models. It contains around 15,000 examples created by Databricks employees and spans tasks such as brainstorming, classification, question answering, information extraction, generation, and summarization. T... | huggingface | databricks/databricks-dolly-15k |
Summarize the dataset and highlight its main purpose, modality, language, license, scale, and popularity. | Name: alpaca
Source: huggingface
ID: tatsu-lab/alpaca
Description: Dataset Card for Alpaca. Alpaca is a dataset of 52,000 instructions and demonstrations generated by OpenAI's text-davinci-003 engine. This instruction data can be used to conduct instruction-tuning for language models and make the language model follow ... | Alpaca is an English text dataset for instruction tuning and text generation. It contains about 52,000 instruction-response examples generated with OpenAI's text-davinci-003 and is designed to improve how well language models follow instructions. The dataset is licensed under CC-BY-NC-4.0, distributed in Parquet format... | huggingface | tatsu-lab/alpaca |
Summarize the dataset and highlight its main purpose, modality, language, license, scale, and popularity. | Name: bad_prompt
Source: huggingface
ID: Nerfgun3/bad_prompt
Description: Negative Embedding / Textual Inversion. The idea behind this embedding was to train the negative prompt as an embedding, unifying the basis of the negative prompt into one word or embedding. The embedding is described as helpful for image generat... | bad_prompt is a small English image-related dataset/resource associated with Stable Diffusion and textual inversion. Its purpose is to package a negative prompt concept into a reusable embedding that can be applied in text-to-image and image-to-image workflows, with claimed benefits for image quality such as hand gener... | huggingface | Nerfgun3/bad_prompt |
Summarize the dataset in a concise SFT-friendly format, including what it is, source, scale, modality, language, license, and notable characteristics. | Name: falcon-refinedweb
Source: huggingface
Dataset ID: tiiuae/falcon-refinedweb
Description: Falcon RefinedWeb is a massive English web dataset built by TII and released under an ODC-By 1.0 license. RefinedWeb is built through stringent filtering and large-scale deduplication of CommonCrawl; models trained on RefinedW... | Falcon RefinedWeb is a large-scale English text dataset hosted on Hugging Face under the ID tiiuae/falcon-refinedweb. It is a web-derived corpus built from CommonCrawl using stringent filtering and large-scale deduplication. The dataset is licensed under ODC-By 1.0, stored in Parquet format, and categorized as text gen... | huggingface | tiiuae/falcon-refinedweb |
Summarize the dataset in a concise SFT-friendly format, including what it is, source, scale, modality, language if available, license if available, and notable characteristics. | Name: lmsys-chat-1m
Source: huggingface
Dataset ID: lmsys/lmsys-chat-1m
Description: LMSYS-Chat-1M is a large-scale real-world LLM conversation dataset containing one million conversations with 25 state-of-the-art LLMs. It was collected from 210K unique IP addresses on the Vicuna demo and Chatbot Arena website from Apr... | LMSYS-Chat-1M is a real-world LLM conversation dataset hosted on Hugging Face under the ID lmsys/lmsys-chat-1m. It contains about one million conversations involving 25 state-of-the-art language models, collected from 210,000 unique IP addresses through the Vicuna demo and Chatbot Arena between April and August 2023. E... | huggingface | lmsys/lmsys-chat-1m |
Summarize the dataset metadata into a concise description for model training data curation. | {"source":"huggingface","id":"anon8231489123/ShareGPT_Vicuna_unfiltered","name":"ShareGPT_Vicuna_unfiltered","description":"Further cleaning done. Please look through the dataset and ensure that I didn't miss anything. Update: Confirmed working method for training the model. Two choices: one version removes instances o... | ShareGPT_Vicuna_unfiltered is a Hugging Face dataset of English conversational training data derived from ShareGPT/Vicuna-style dialogues. The description notes additional cleaning and provides two variants: one that removes occurrences of the phrase "I'm sorry, but" and one that retains them. It is tagged with Apache-... | huggingface | anon8231489123/ShareGPT_Vicuna_unfiltered |
Summarize the dataset metadata into a concise description for model training data curation. | {"source":"huggingface","id":"HuggingFaceFW/finepdfs","name":"finepdfs","description":"Liberating 3T of the finest tokens from PDFs. FinePDFs is a publicly available corpus sourced exclusively from PDFs, containing about 3 trillion tokens across 475 million documents in 1733 languages.","tags":["task_categories:text-ge... | finepdfs is a large-scale Hugging Face text generation dataset built exclusively from PDF sources. According to the description, it contains roughly 3 trillion tokens drawn from 475 million documents spanning 1,733 languages, making it a highly multilingual corpus. The dataset is distributed in Parquet format with both... | huggingface | HuggingFaceFW/finepdfs |
Summarize the dataset for a machine learning practitioner, including its purpose, data characteristics, and notable metadata. | Name: PhysicalAI-Autonomous-Vehicles
Description: The PhysicalAI-Autonomous-Vehicles dataset provides one of the largest, geographically diverse collections of multi-sensor data empowering AV researchers to build the next generation of Physical AI based end-to-end driving systems. This dataset is ready for commercial/n... | PhysicalAI-Autonomous-Vehicles is a large-scale autonomous driving dataset intended for AV research and development, especially end-to-end Physical AI driving systems. It contains geographically diverse multi-sensor driving data and includes about 1700 hours of driving. Both data collection and labeling are performed a... | huggingface | nvidia/PhysicalAI-Autonomous-Vehicles |
Summarize the dataset for a machine learning practitioner, including its purpose, content coverage, and notable metadata. | Name: OpenThoughts-114k
Description: Open synthetic reasoning dataset with 114k high-quality examples covering math, science, code, and puzzles. The default subset contains ready-to-train data used to finetune the OpenThinker-7B and OpenThinker-32B models.
Tags: license:apache-2.0, size_categories:100K<n<1M, format:par... | OpenThoughts-114k is a synthetic text reasoning dataset designed for training and fine-tuning language models. It contains 114k high-quality examples spanning math, science, code, and puzzles. The dataset includes a default ready-to-train subset that was used to finetune the OpenThinker-7B and OpenThinker-32B models. I... | huggingface | open-thoughts/OpenThoughts-114k |
Summarize the dataset for instruction tuning. | Name: OpenHermes-2.5
Source: huggingface
Dataset ID: teknium/OpenHermes-2.5
Description: This dataset made OpenHermes 2.5 and Nous Hermes 2 series of models. It is an exact compilation and curation of many open source datasets and custom created synthetic datasets, supporting recent state-of-the-art LLM advancements.
T... | OpenHermes-2.5 is a large English text dataset on Hugging Face used to train the OpenHermes 2.5 and Nous Hermes 2 model series. It consists of a curated compilation of open-source and synthetic instruction data, with signals such as GPT-4, distillation, and dataset compilation. The dataset is distributed in JSON format... | huggingface | teknium/OpenHermes-2.5 |
Summarize the dataset for instruction tuning. | Name: alpaca-cleaned
Source: huggingface
Dataset ID: yahma/alpaca-cleaned
Description: This is a cleaned version of the original Alpaca dataset released by Stanford. It fixes issues in the original release, including hallucination-prone instructions that referenced internet data.
Tags: task_categories:text-generation, ... | alpaca-cleaned is an English instruction-tuning dataset on Hugging Face derived from Stanford's original Alpaca release. It is a cleaned version designed to remove problematic samples, especially instructions that encouraged hallucinations by referencing unavailable internet data. The dataset is intended for text gener... | huggingface | yahma/alpaca-cleaned |
Summarize the dataset metadata into a concise description suitable for supervised fine-tuning. | Dataset source: huggingface
Dataset id: cais/hle
Dataset name: hle
Description: Humanity's Last Exam (HLE) is a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. It consists of 2,500 questions across dozens of subj... | {"name":"hle","full_id":"cais/hle","source":"huggingface","summary":"Humanity's Last Exam is an official multimodal benchmark dataset covering frontier-level academic questions across many subjects. It contains 2,500 closed-ended questions and includes both text and image modalities.","license":"mit","modalities":["ima... | huggingface | cais/hle |
Summarize the dataset metadata into a concise description suitable for supervised fine-tuning. | Dataset source: huggingface
Dataset id: HuggingFaceFW/fineweb-2
Dataset name: fineweb-2
Description: FineWeb2 is the second iteration of the FineWeb dataset, providing high-quality pretraining data for over 1000 languages. It is fully reproducible, available under the permissive ODC-By 1.0 license, and validated throug... | {"name":"fineweb-2","full_id":"HuggingFaceFW/fineweb-2","source":"huggingface","summary":"FineWeb2 is a large-scale multilingual pretraining dataset spanning over 1000 languages. It is designed for text generation, is fully reproducible, and has been extensively validated through ablation studies.","license":"odc-by","... | huggingface | HuggingFaceFW/fineweb-2 |
Summarize the dataset based on its metadata, including its purpose, modality, likely use cases, language coverage, license, and notable characteristics. | Name: imagenet-1k
Source: huggingface
ID: ILSVRC/imagenet-1k
Description: ILSVRC 2012, commonly known as 'ImageNet' is an image dataset organized according to the WordNet hierarchy. Each meaningful concept in WordNet, possibly described by multiple words or word phrases, is called a synonym set or synset. There are mor... | ImageNet-1k is a large-scale image classification dataset derived from the ILSVRC 2012 benchmark. It contains image data organized according to the WordNet hierarchy, where each class corresponds to a synset, making it suitable for multi-class image classification tasks. The dataset is monolingual in English and was cr... | huggingface | ILSVRC/imagenet-1k |
Summarize the dataset based on its metadata, including its purpose, data type, language coverage, license, and likely machine learning applications. | Name: Alpaca-CoT
Source: huggingface
ID: QingyiSi/Alpaca-CoT
Description: This repository continuously collects various instruction tuning datasets and standardizes them into the same format so they can be directly loaded by Alpaca model code. It also references empirical study on various instruction-tuning datasets ba... | Alpaca-CoT is an instruction-finetuning dataset collection designed to aggregate and standardize multiple instruction-tuning sources into a unified format compatible with Alpaca-style training pipelines. Its content is intended for instruction-following and chain-of-thought style learning, as suggested by the tags 'Ins... | huggingface | QingyiSi/Alpaca-CoT |
Summarize the dataset and list its key characteristics for synthetic data and persona-driven LLM research. | Dataset name: PersonaHub
Source: huggingface
Description: Scaling Synthetic Data Creation with 1,000,000,000 Personas. This repo releases data introduced in the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas." It proposes a persona-driven data synthesis methodology that leverages various perspective... | PersonaHub is a Hugging Face dataset focused on persona-driven synthetic data creation for LLM research. It accompanies the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas" and introduces PERSONA HUB, a large collection of automatically curated personas derived from web data to support diverse data s... | huggingface | proj-persona/PersonaHub |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"fka/prompts.chat","name":"prompts.chat","description":"a.k.a. Awesome ChatGPT Prompts. This is a Dataset Repository mirror of prompts.chat — a social platform for AI prompts. prompts.chat is an open-source platform where users can share, discover, and collect AI prompts from the community.... | prompts.chat is a community-curated prompt collection mirrored from the prompts.chat platform and the Awesome ChatGPT Prompts project. It is best suited for prompt engineering research, instruction template discovery, prompt retrieval, and analysis of popular prompting patterns across major LLM ecosystems such as ChatG... | huggingface | fka/prompts.chat |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"HuggingFaceFW/fineweb","name":"fineweb","description":"FineWeb: 15 trillion tokens of the finest data the web has to offer. The dataset consists of more than 18.5T tokens of cleaned and deduplicated English web data from CommonCrawl. The data processing pipeline is optimized for LLM perfor... | FineWeb is a massive English web-scale corpus derived from CommonCrawl and processed for high-quality LLM pretraining. Its primary use case is large-scale base model training for text generation, especially next-token prediction. The dataset emphasizes cleaning, deduplication, and pipeline quality, making it suitable f... | huggingface | HuggingFaceFW/fineweb |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"Anthropic/hh-rlhf","name":"hh-rlhf","description":"Dataset Card for HH-RLHF. This repository provides access to human preference data about helpfulness and harmlessness for training preference or reward models for subsequent RLHF training. These data are not meant for supervised training o... | HH-RLHF is a human feedback dataset designed for alignment research, especially reward modeling and preference learning around helpfulness and harmlessness. Its core use case is RLHF pipelines, where models learn from ranked or preferred responses rather than standard supervised targets. The dataset card explicitly war... | huggingface | Anthropic/hh-rlhf |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"Open-Orca/OpenOrca","name":"OpenOrca","description":"The OpenOrca dataset is a rich collection of augmented FLAN data aligned as closely as possible with the distributions outlined in the Orca paper. It has been instrumental in generating high-performing model checkpoints and serves as a r... | OpenOrca is a large-scale instruction-style dataset built from augmented FLAN data and inspired by the Orca family of training distributions. It is well suited for supervised fine-tuning of general-purpose assistants because it covers many NLP task types, including question answering, summarization, classification, tab... | huggingface | Open-Orca/OpenOrca |
Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details. | {"source":"huggingface","id":"OpenAssistant/oasst1","name":"oasst1","description":"OpenAssistant Conversations Dataset (OASST1). A human-generated, human-annotated assistant-style conversation corpus consisting of 161,443 messages in 35 different languages, annotated with 461,292 quality ratings and over 10,000 fully a... | OASST1 is a multilingual assistant conversation dataset created through human generation and annotation, with conversation trees and extensive quality ratings. It is highly suitable for supervised fine-tuning of chat assistants, multilingual dialogue modeling, response ranking, and research on conversational alignment.... | huggingface | OpenAssistant/oasst1 |
Summarize the dataset in a concise SFT-friendly format, including domain, modality, main task, language, size, license, and notable characteristics. | Name: gsm8k
Description: Dataset Card for GSM8K
Dataset Summary
GSM8K (Grade School Math 8K) is a dataset of 8.5K high quality linguistically diverse grade school math word problems. The dataset was created to support the task of question answering on basic mathematical problems that require multi-step reasoning.
Thes... | GSM8K is an English text dataset of grade school math word problems designed for multi-step reasoning and question answering. It contains around 8.5K high-quality problems, generally requiring 2 to 8 reasoning steps using basic arithmetic. The dataset is original, crowdsourced, monolingual, MIT-licensed, and commonly u... | huggingface | openai/gsm8k |
Summarize the dataset in a concise SFT-friendly format, including domain, modality, main task, language coverage, size, license, and notable characteristics. | Name: wikipedia
Description: Dataset Card for Wikimedia Wikipedia
Dataset Summary
Wikipedia dataset containing cleaned articles of all languages. The dataset is built from the Wikipedia dumps with one subset per language, each containing a single train split. Each example contains the content of one full Wikipedia art... | Wikimedia Wikipedia is a massive multilingual text dataset consisting of cleaned full Wikipedia articles across a very large number of languages. It is intended for language modeling and masked language modeling tasks, with one language subset per language and a train split containing processed article text. The datase... | huggingface | wikimedia/wikipedia |
Summarize the dataset or asset in a concise SFT-friendly format, including modality, intended use, scale, license, and notable limitations if apparent. | Name: EasyNegative
Description: Negative Embedding
This is a Negative Embedding trained with Counterfeit. Please use it in the "stable-diffusion-webui/embeddings" folder. It can be used with other models, but the effectiveness is not certain.
Referenced models: Counterfeit-V2.0.safetensors, AbyssOrangeMix2_sfw.safetens... | EasyNegative is an image-related negative embedding asset intended for use with Stable Diffusion-style workflows, especially via the embeddings folder in stable-diffusion-webui. It was trained with Counterfeit and may also work with other models, though effectiveness outside the original setup is uncertain. The asset i... | huggingface | gsdf/EasyNegative |
Summarize the dataset in a concise SFT-friendly format, including domain, modality, main task, language, scale, and notable characteristics. | Name: RedPajama-Data-1T
Description: RedPajama is a clean-room, fully open-source implementation of the LLaMa dataset.
Tags: task_categories:text-generation, language:en, size_categories:1M<n<10M, modality:text, region:us
Downloads: 2123
Likes: 1151 | RedPajama-Data-1T is a large English text dataset created as a clean-room, fully open-source reproduction of the data mixture used for LLaMA-style model training. It is designed for text generation and large-scale language model pretraining. The dataset is text-only, large in scale, and notable for its open-source posi... | huggingface | togethercomputer/RedPajama-Data-1T |
Summarize the dataset in a concise SFT-friendly format, including domain, modality, tasks, languages, license, size, and notable characteristics. | Name: medical-o1-reasoning-SFT
Description: News
[2025/04/22] We split the data and kept only the medical SFT dataset (medical_o1_sft.json). The file medical_o1_sft_mix.json contains a mix of medical and general instruction data.
[2025/02/22] We released the distilled dataset from Deepseek-R1 based on medical verifiabl... | medical-o1-reasoning-SFT is a bilingual English-Chinese medical text dataset for supervised fine-tuning, focused on medical reasoning over verifiable problems. It supports both question answering and text generation, and includes reasoning-oriented supervision distilled from DeepSeek-R1. The dataset is distributed in J... | huggingface | FreedomIntelligence/medical-o1-reasoning-SFT |
Summarize the dataset for LLM pretraining use, including its scale, domain, language, license, and notable metadata. | Name: fineweb-edu
Description: FineWeb-Edu is a 1.3 trillion token educational web dataset filtered from FineWeb using an educational quality classifier. It is the 1.3T token version and is intended for high-quality educational text collection.
Tags: task_categories:text-generation, language:en, license:odc-by, size_ca... | FineWeb-Edu is a large-scale English educational text dataset from Hugging Face intended for text generation and LLM pretraining. It contains about 1.3 trillion tokens of educational web content filtered from FineWeb using a quality classifier informed by Llama3-70B-Instruct annotations. The dataset is distributed in p... | huggingface | HuggingFaceFW/fineweb-edu |
Provide a concise dataset overview covering purpose, token scale, language, license, and relevance for language model training. | Name: dolma
Description: Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research
Tags: task_categories:text-generation, language:en, license:odc-by, size_categories:n>1T, arxiv:2402.00159, arxiv:2301.13688, region:us, language-modeling, casual-lm, llm
Downloads: 3087
Likes: 1021 | Dolma is an open English corpus designed for language model pretraining research. It contains approximately three trillion tokens, making it one of the largest openly available text-generation datasets. The dataset is licensed under ODC-BY and is explicitly positioned for language modeling, causal language modeling, an... | huggingface | allenai/dolma |
Summarize this dataset with emphasis on content type, multilingual coverage, licensing characteristics, and suitability for code model training. | Name: the-stack
Description: The Stack is a large code dataset. Initial release included 30 programming languages and 18 permissive licenses, producing a near-deduplicated 3TB dataset. Later versions excluded weak copyleft licenses and expanded the permissive license list to 193 licenses.
Tags: task_categories:text-gen... | The Stack is a large multilingual code dataset built for text generation and code model training. It aggregates source code from many programming languages, originally covering 30 languages and later expanding licensing coverage while removing weak copyleft licenses. The dataset is near-deduplicated, distributed in par... | huggingface | bigcode/the-stack |
Create a short summary describing the dataset’s content, generation method, scale, license, and practical use in model training. | Name: TinyStories
Description: Dataset containing synthetically generated short stories created by GPT-3.5 and GPT-4 using a small vocabulary. It is described in arXiv:2305.07759 and includes train and validation files for model training.
Tags: task_categories:text-generation, language:en, license:cdla-sharing-1.0, siz... | TinyStories is an English text-generation dataset of synthetically generated short stories created with GPT-3.5 and GPT-4. The stories use a constrained vocabulary, making the dataset useful for studying small language models, language acquisition behavior, and efficient training setups. It is relatively small in size,... | huggingface | roneneldan/TinyStories |
Summarize the dataset as an instruction-tuning resource, covering task types, data origin, size, format, and license. | Name: databricks-dolly-15k
Description: databricks-dolly-15k is an open source dataset of instruction-following records generated by thousands of Databricks employees across behavioral categories such as brainstorming, classification, closed QA, generation, information extraction, open QA, and summarization.
Tags: task... | databricks-dolly-15k is an English instruction-following dataset designed for supervised fine-tuning of assistant models. It contains around 15,000 examples created by Databricks employees and spans tasks such as brainstorming, classification, question answering, information extraction, generation, and summarization. T... | huggingface | databricks/databricks-dolly-15k |
Summarize the dataset and highlight its main purpose, modality, language, license, scale, and popularity. | Name: alpaca
Source: huggingface
ID: tatsu-lab/alpaca
Description: Dataset Card for Alpaca. Alpaca is a dataset of 52,000 instructions and demonstrations generated by OpenAI's text-davinci-003 engine. This instruction data can be used to conduct instruction-tuning for language models and make the language model follow ... | Alpaca is an English text dataset for instruction tuning and text generation. It contains about 52,000 instruction-response examples generated with OpenAI's text-davinci-003 and is designed to improve how well language models follow instructions. The dataset is licensed under CC-BY-NC-4.0, distributed in Parquet format... | huggingface | tatsu-lab/alpaca |
Summarize the dataset and highlight its main purpose, modality, language, license, scale, and popularity. | Name: bad_prompt
Source: huggingface
ID: Nerfgun3/bad_prompt
Description: Negative Embedding / Textual Inversion. The idea behind this embedding was to train the negative prompt as an embedding, unifying the basis of the negative prompt into one word or embedding. The embedding is described as helpful for image generat... | bad_prompt is a small English image-related dataset/resource associated with Stable Diffusion and textual inversion. Its purpose is to package a negative prompt concept into a reusable embedding that can be applied in text-to-image and image-to-image workflows, with claimed benefits for image quality such as hand gener... | huggingface | Nerfgun3/bad_prompt |
Summarize the dataset in a concise SFT-friendly format, including what it is, source, scale, modality, language, license, and notable characteristics. | Name: falcon-refinedweb
Source: huggingface
Dataset ID: tiiuae/falcon-refinedweb
Description: Falcon RefinedWeb is a massive English web dataset built by TII and released under an ODC-By 1.0 license. RefinedWeb is built through stringent filtering and large-scale deduplication of CommonCrawl; models trained on RefinedW... | Falcon RefinedWeb is a large-scale English text dataset hosted on Hugging Face under the ID tiiuae/falcon-refinedweb. It is a web-derived corpus built from CommonCrawl using stringent filtering and large-scale deduplication. The dataset is licensed under ODC-By 1.0, stored in Parquet format, and categorized as text gen... | huggingface | tiiuae/falcon-refinedweb |
Summarize the dataset in a concise SFT-friendly format, including what it is, source, scale, modality, language if available, license if available, and notable characteristics. | Name: lmsys-chat-1m
Source: huggingface
Dataset ID: lmsys/lmsys-chat-1m
Description: LMSYS-Chat-1M is a large-scale real-world LLM conversation dataset containing one million conversations with 25 state-of-the-art LLMs. It was collected from 210K unique IP addresses on the Vicuna demo and Chatbot Arena website from Apr... | LMSYS-Chat-1M is a real-world LLM conversation dataset hosted on Hugging Face under the ID lmsys/lmsys-chat-1m. It contains about one million conversations involving 25 state-of-the-art language models, collected from 210,000 unique IP addresses through the Vicuna demo and Chatbot Arena between April and August 2023. E... | huggingface | lmsys/lmsys-chat-1m |
Summarize the dataset metadata into a concise description for model training data curation. | {"source":"huggingface","id":"anon8231489123/ShareGPT_Vicuna_unfiltered","name":"ShareGPT_Vicuna_unfiltered","description":"Further cleaning done. Please look through the dataset and ensure that I didn't miss anything. Update: Confirmed working method for training the model. Two choices: one version removes instances o... | ShareGPT_Vicuna_unfiltered is a Hugging Face dataset of English conversational training data derived from ShareGPT/Vicuna-style dialogues. The description notes additional cleaning and provides two variants: one that removes occurrences of the phrase "I'm sorry, but" and one that retains them. It is tagged with Apache-... | huggingface | anon8231489123/ShareGPT_Vicuna_unfiltered |
Summarize the dataset metadata into a concise description for model training data curation. | {"source":"huggingface","id":"HuggingFaceFW/finepdfs","name":"finepdfs","description":"Liberating 3T of the finest tokens from PDFs. FinePDFs is a publicly available corpus sourced exclusively from PDFs, containing about 3 trillion tokens across 475 million documents in 1733 languages.","tags":["task_categories:text-ge... | finepdfs is a large-scale Hugging Face text generation dataset built exclusively from PDF sources. According to the description, it contains roughly 3 trillion tokens drawn from 475 million documents spanning 1,733 languages, making it a highly multilingual corpus. The dataset is distributed in Parquet format with both... | huggingface | HuggingFaceFW/finepdfs |
Summarize the dataset for a machine learning practitioner, including its purpose, data characteristics, and notable metadata. | Name: PhysicalAI-Autonomous-Vehicles
Description: The PhysicalAI-Autonomous-Vehicles dataset provides one of the largest, geographically diverse collections of multi-sensor data empowering AV researchers to build the next generation of Physical AI based end-to-end driving systems. This dataset is ready for commercial/n... | PhysicalAI-Autonomous-Vehicles is a large-scale autonomous driving dataset intended for AV research and development, especially end-to-end Physical AI driving systems. It contains geographically diverse multi-sensor driving data and includes about 1700 hours of driving. Both data collection and labeling are performed a... | huggingface | nvidia/PhysicalAI-Autonomous-Vehicles |
Summarize the dataset for a machine learning practitioner, including its purpose, content coverage, and notable metadata. | Name: OpenThoughts-114k
Description: Open synthetic reasoning dataset with 114k high-quality examples covering math, science, code, and puzzles. The default subset contains ready-to-train data used to finetune the OpenThinker-7B and OpenThinker-32B models.
Tags: license:apache-2.0, size_categories:100K<n<1M, format:par... | OpenThoughts-114k is a synthetic text reasoning dataset designed for training and fine-tuning language models. It contains 114k high-quality examples spanning math, science, code, and puzzles. The dataset includes a default ready-to-train subset that was used to finetune the OpenThinker-7B and OpenThinker-32B models. I... | huggingface | open-thoughts/OpenThoughts-114k |
Summarize the dataset for instruction tuning. | Name: OpenHermes-2.5
Source: huggingface
Dataset ID: teknium/OpenHermes-2.5
Description: This dataset made OpenHermes 2.5 and Nous Hermes 2 series of models. It is an exact compilation and curation of many open source datasets and custom created synthetic datasets, supporting recent state-of-the-art LLM advancements.
T... | OpenHermes-2.5 is a large English text dataset on Hugging Face used to train the OpenHermes 2.5 and Nous Hermes 2 model series. It consists of a curated compilation of open-source and synthetic instruction data, with signals such as GPT-4, distillation, and dataset compilation. The dataset is distributed in JSON format... | huggingface | teknium/OpenHermes-2.5 |
Summarize the dataset for instruction tuning. | Name: alpaca-cleaned
Source: huggingface
Dataset ID: yahma/alpaca-cleaned
Description: This is a cleaned version of the original Alpaca dataset released by Stanford. It fixes issues in the original release, including hallucination-prone instructions that referenced internet data.
Tags: task_categories:text-generation, ... | alpaca-cleaned is an English instruction-tuning dataset on Hugging Face derived from Stanford's original Alpaca release. It is a cleaned version designed to remove problematic samples, especially instructions that encouraged hallucinations by referencing unavailable internet data. The dataset is intended for text gener... | huggingface | yahma/alpaca-cleaned |
Summarize the dataset metadata into a concise description suitable for supervised fine-tuning. | Dataset source: huggingface
Dataset id: cais/hle
Dataset name: hle
Description: Humanity's Last Exam (HLE) is a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. It consists of 2,500 questions across dozens of subj... | {"name":"hle","full_id":"cais/hle","source":"huggingface","summary":"Humanity's Last Exam is an official multimodal benchmark dataset covering frontier-level academic questions across many subjects. It contains 2,500 closed-ended questions and includes both text and image modalities.","license":"mit","modalities":["ima... | huggingface | cais/hle |
Summarize the dataset metadata into a concise description suitable for supervised fine-tuning. | Dataset source: huggingface
Dataset id: HuggingFaceFW/fineweb-2
Dataset name: fineweb-2
Description: FineWeb2 is the second iteration of the FineWeb dataset, providing high-quality pretraining data for over 1000 languages. It is fully reproducible, available under the permissive ODC-By 1.0 license, and validated throug... | {"name":"fineweb-2","full_id":"HuggingFaceFW/fineweb-2","source":"huggingface","summary":"FineWeb2 is a large-scale multilingual pretraining dataset spanning over 1000 languages. It is designed for text generation, is fully reproducible, and has been extensively validated through ablation studies.","license":"odc-by","... | huggingface | HuggingFaceFW/fineweb-2 |
Summarize the dataset based on its metadata, including its purpose, modality, likely use cases, language coverage, license, and notable characteristics. | Name: imagenet-1k
Source: huggingface
ID: ILSVRC/imagenet-1k
Description: ILSVRC 2012, commonly known as 'ImageNet' is an image dataset organized according to the WordNet hierarchy. Each meaningful concept in WordNet, possibly described by multiple words or word phrases, is called a synonym set or synset. There are mor... | ImageNet-1k is a large-scale image classification dataset derived from the ILSVRC 2012 benchmark. It contains image data organized according to the WordNet hierarchy, where each class corresponds to a synset, making it suitable for multi-class image classification tasks. The dataset is monolingual in English and was cr... | huggingface | ILSVRC/imagenet-1k |
Summarize the dataset based on its metadata, including its purpose, data type, language coverage, license, and likely machine learning applications. | Name: Alpaca-CoT
Source: huggingface
ID: QingyiSi/Alpaca-CoT
Description: This repository continuously collects various instruction tuning datasets and standardizes them into the same format so they can be directly loaded by Alpaca model code. It also references empirical study on various instruction-tuning datasets ba... | Alpaca-CoT is an instruction-finetuning dataset collection designed to aggregate and standardize multiple instruction-tuning sources into a unified format compatible with Alpaca-style training pipelines. Its content is intended for instruction-following and chain-of-thought style learning, as suggested by the tags 'Ins... | huggingface | QingyiSi/Alpaca-CoT |
Summarize the dataset and list its key characteristics for synthetic data and persona-driven LLM research. | Dataset name: PersonaHub
Source: huggingface
Description: Scaling Synthetic Data Creation with 1,000,000,000 Personas. This repo releases data introduced in the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas." It proposes a persona-driven data synthesis methodology that leverages various perspective... | PersonaHub is a Hugging Face dataset focused on persona-driven synthetic data creation for LLM research. It accompanies the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas" and introduces PERSONA HUB, a large collection of automatically curated personas derived from web data to support diverse data s... | huggingface | proj-persona/PersonaHub |
SoloAI SFT Dataset: 20260426-0138
📊 数据集概览
| 属性 | 值 |
|---|---|
| 数据条数 | 30 条 |
| 数据源 | huggingface |
| 生成时间 | 2026-04-26 01:38 |
| 格式 | Instruction-Input-Output |
| 主要用途 | SFT 微调 / 指令调优 |
🎯 适用场景
- LLM 指令微调 (Instruction Tuning): 适用于训练对话型 AI 助手
- Prompt Engineering 研究: 可用于分析和优化提示词设计
- 多语言支持: 包含英文和中文数据
📝 数据格式
每条数据包含三个字段：
instruction: 任务指令（描述需要完成的任务）input: 输入上下文（提供任务相关的背景信息）output: 期望输出（模型应生成的回答）
示例
{'instruction': 'Given a dataset metadata record, summarize its purpose, likely LLM training use cases, notable characteristics, and key licensing or scale details.', 'input': '{"source":"huggingface","id":"fka/prompts.chat","name":"prompts.chat","description":"a.k.a. Awesome ChatGPT Prompts. This is a Dataset Repository mirror of prompts.chat — a social platform for AI prompts. prompts.chat is an open-source platform where users can share, discover, and collect AI prompts from the community.","tags":["task_categories:question-answering","task_categories:text-generation","license:cc0-1.0","size_categories:1K<n<10K","format:csv","modality:text","library:datasets","library:pandas","library:polars","library:mlcroissant","region:us","ChatGPT","prompts","AI","GPT","Claude","Gemini","Llama","Mistral","LLM","prompt-engineering","conversational-ai","text-generation","chatbot","awesome-list"],"downloads":47783,"likes":9680,"created_at":""}', 'output': 'prompts.chat is a community-curated prompt collection mirrored from the prompts.chat platform and the Awesome ChatGPT Prompts project. It is best suited for prompt engineering research, instruction template discovery, prompt retrieval, and analysis of popular prompting patterns across major LLM ecosystems such as ChatGPT, Claude, Gemini, Llama, and Mistral. The dataset is relatively small, text-based, distributed as CSV, and tagged for question answering and text generation. Its CC0-1.0 license makes it easy to reuse commercially and academically. This dataset is more useful for studying prompts and building prompt libraries than for direct supervised fine-tuning of assistant responses, because it appears focused on prompt artifacts rather than full prompt-response conversations.', 'source': 'huggingface', 'original_id': 'fka/prompts.chat'}
🤖 数据来源
本数据集由 SoloAI 自动化数据管道生成：
- 从 HuggingFace Datasets Hub 发现高质量数据集
- AI 清洗为 SFT 格式（Instruction-Input-Output）
- 质量过滤后发布
⚠️ 使用说明 & 📬 商务联系
- 本数据集仅供研究和实验用途
- 请遵守原始数据的许可证要求
- 商业用途 / 定制数据 / 深度合作:
- 📧 请联系:
379744050@qq.com - 🤖 SoloAI 提供高质量 SFT 数据定制服务。
- 建议邮件标题:
【数据定制咨询】行业 + 数据类型 + 规模 - 建议正文包含: 目标用途、需要条数、语言、字段格式、预算、交付时间
- 📧 请联系:
💰 商业合作报价
| 套餐 | 价格 | 说明 |
|---|---|---|
| Starter | $199 / 1000条高质量 SFT 数据 | 适合个人开发者 / 小团队 |
| Growth | $499 / 5000条行业数据 | 适合垂直行业训练数据 |
| Enterprise | $1499 / 定制领域数据管道 | 适合长期定制与数据管道 |
💳 支付方式
- 中国客户: 支付宝, 微信支付
- 海外客户: PayPal, USDT (TRC20)
- 下单方式: 邮件联系后 24 小时内提供交付方案与付款指引
🚀 为什么现在联系 SoloAI
- 24 小时内响应有效询盘
- 报价前可免费给出需求范围建议
- 支持中文 / English 项目合作
- 可从单次交付升级为长期数据管道合作
📈 更新日志
| 版本 | 日期 | 说明 |
|---|---|---|
| v1.0 | 2026-04-26 01:38 | 初始发布，30 条数据 |
- Downloads last month
- 13