instruction stringlengths 44 190 | input stringlengths 238 1k | output stringlengths 294 705 | source stringclasses 1
value | original_id stringlengths 8 44 |
|---|---|---|---|---|
Summarize the dataset for someone looking for community-curated prompt collections for LLM prompting and chatbot use. | Name: prompts.chat
Source: huggingface
ID: fka/prompts.chat
Description: a.k.a. Awesome ChatGPT Prompts. This is a Dataset Repository mirror of prompts.chat — a social platform for AI prompts. The platform lets users share, discover, and collect AI prompts from the community.
Tags: task_categories:question-answering, t... | prompts.chat is a community-driven prompt dataset mirrored on Hugging Face. It focuses on prompt sharing and discovery for LLMs and chatbots, with strong relevance to prompt engineering and conversational AI. The dataset is relatively small, text-based, distributed in CSV format, and licensed under CC0-1.0, making it e... | huggingface | fka/prompts.chat |
Provide a concise dataset overview for pretraining researchers evaluating large-scale English web corpora for LLM training. | Name: fineweb
Source: huggingface
ID: HuggingFaceFW/fineweb
Description: FineWeb contains more than 18.5T tokens of cleaned and deduplicated English web data from CommonCrawl. The processing pipeline is optimized for LLM performance and built with the datatrove large-scale data processing library.
Tags: task_categories... | FineWeb is a massive English web corpus for LLM pretraining, built from cleaned and deduplicated CommonCrawl data. With over 18.5 trillion tokens and a pipeline optimized for language model performance, it is well suited for large-scale text-generation research and foundation model training. Its scale and popularity ma... | huggingface | HuggingFaceFW/fineweb |
Explain what this dataset is useful for and note any limitation for supervised fine-tuning use. | Name: hh-rlhf
Source: huggingface
ID: Anthropic/hh-rlhf
Description: This repository provides human preference data about helpfulness and harmlessness from RLHF research. These data are meant to train preference or reward models for subsequent RLHF training and are not meant for supervised training of dialogue agents.
... | HH-RLHF is primarily a human preference dataset designed for reward modeling and RLHF research, especially around helpfulness and harmlessness. It is useful for training preference models rather than standard supervised chat assistants. A key limitation is that the creators explicitly caution against using it directly ... | huggingface | Anthropic/hh-rlhf |
Write a short description of this dataset for someone searching for instruction-tuning data for general-purpose LLMs. | Name: OpenOrca
Source: huggingface
ID: Open-Orca/OpenOrca
Description: OpenOrca is a rich collection of augmented FLAN data aligned as closely as possible with distributions described in the Orca paper. It has been used to generate strong model checkpoints and is intended as a resource for NLP researchers and developer... | OpenOrca is a large English instruction-style dataset built from augmented FLAN data and inspired by the Orca training setup. It supports a wide variety of NLP tasks, including question answering, summarization, classification, and text generation, making it a versatile resource for instruction tuning general-purpose L... | huggingface | Open-Orca/OpenOrca |
Summarize this dataset for multilingual alignment and assistant conversation fine-tuning use cases. | Name: oasst1
Source: huggingface
ID: OpenAssistant/oasst1
Description: OpenAssistant Conversations Dataset (OASST1) is a human-generated, human-annotated assistant-style conversation corpus with 161,443 messages in 35 languages, 461,292 quality ratings, and over 10,000 fully annotated conversation trees.
Tags: multiple... | OASST1 is a multilingual assistant conversation dataset designed for alignment research and chat model fine-tuning. It contains human-written, human-rated conversation trees across 35 languages, making it valuable for multilingual assistant training, response ranking, and quality-aware dialogue modeling. Its annotated ... | huggingface | OpenAssistant/oasst1 |
Summarize the dataset for supervised fine-tuning selection. | Name: gsm8k
Description: Dataset Card for GSM8K. GSM8K (Grade School Math 8K) is a dataset of 8.5K high quality linguistically diverse grade school math word problems created to support question answering on basic mathematical problems requiring multi-step reasoning. Problems typically take 2 to 8 steps and use element... | GSM8K is an English text dataset of grade school math word problems designed for multi-step reasoning and arithmetic question answering. It is a strong benchmark-style resource for SFT or evaluation on mathematical reasoning tasks. The dataset is MIT-licensed, medium-sized, and highly popular. | huggingface | openai/gsm8k |
Summarize the dataset for supervised fine-tuning selection. | Name: wikipedia
Description: Dataset Card for Wikimedia Wikipedia. Wikipedia dataset containing cleaned articles of all languages. Built from Wikipedia dumps with one subset per language, each containing a single train split. Each example contains the content of one full cleaned Wikipedia article.
Tags: task_categories... | Wikimedia Wikipedia is a very large multilingual text corpus of cleaned full Wikipedia articles across many languages. It is well suited for language modeling, masked language modeling, and general knowledge pretraining rather than narrow instruction tuning. Licensing includes CC-BY-SA-3.0 and GFDL, so downstream usage... | huggingface | wikimedia/wikipedia |
Summarize the dataset for supervised fine-tuning selection. | Name: EasyNegative
Description: Negative Embedding trained with Counterfeit. Intended for use in the stable-diffusion-webui embeddings folder. Can be used with other models, but effectiveness is uncertain.
Tags: license:other, size_categories:n<1K, format:imagefolder, modality:image
Downloads: 34931
Likes: 1174 | EasyNegative appears to be a small image-related asset centered on a negative embedding for Stable Diffusion workflows rather than a conventional text SFT dataset. It is likely unsuitable for LLM supervised fine-tuning in standard instruction-response format. The license is non-standard ('other'), so reuse terms should... | huggingface | gsdf/EasyNegative |
Summarize the dataset for supervised fine-tuning selection. | 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
Downloads: 2154
Likes: 1152 | RedPajama-Data-1T is a large English text corpus intended as an open-source reproduction of the data used for LLaMA-style model pretraining. It is best suited for foundation model pretraining or continued pretraining, not directly for instruction SFT unless additional prompt-response formatting is applied. Its value is... | huggingface | togethercomputer/RedPajama-Data-1T |
Summarize the dataset for supervised fine-tuning selection. | Name: medical-o1-reasoning-SFT
Description: Medical reasoning dataset for SFT built on medical verifiable problems and distilled reasoning data. News indicates the dataset now keeps only the medical SFT file and includes reasoning chains derived from Deepseek-R1-based distillation.
Tags: task_categories:question-answer... | medical-o1-reasoning-SFT is a bilingual English-Chinese medical reasoning dataset explicitly designed for supervised fine-tuning. It contains medical instruction and reasoning-oriented data based on verifiable problems, making it highly relevant for training domain-specific reasoning assistants. The Apache-2.0 license ... | huggingface | FreedomIntelligence/medical-o1-reasoning-SFT |
Summarize the dataset and explain why it is useful for language model training. | Name: fineweb-edu
Description: FineWeb-Edu is a 1.3 trillion token educational web dataset filtered from FineWeb using an educational quality classifier. It focuses on high-quality educational content from the web.
Tags: task_categories:text-generation, language:en, license:odc-by, size_categories:1B<n<10B, format:parq... | FineWeb-Edu is a large-scale English educational text dataset designed for text generation and language model pretraining. It contains 1.3 trillion tokens of web pages selected for educational quality from the broader FineWeb corpus. Because it emphasizes high-quality educational content and is distributed in scalable ... | huggingface | HuggingFaceFW/fineweb-edu |
Provide a concise overview of this dataset including its scale, purpose, and licensing. | Name: dolma
Description: Dolma is 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, language-modeling, casual-lm, llm
Downloads: 3153
Likes: 1022 | Dolma is a massive open English text corpus created for language model pretraining research. With roughly three trillion tokens, it is intended for large-scale language modeling and LLM development. The dataset is licensed under ODC-BY, making it suitable for open research workflows that require a very large pretrainin... | huggingface | allenai/dolma |
Describe this dataset in terms of content, scope, and ideal use cases. | Name: the-stack
Description: The Stack is a large code dataset. Its releases include many programming languages and permissive licenses, with near-deduplicated data totaling multiple terabytes.
Tags: task_categories:text-generation, multilinguality:multilingual, language:code, license:other, size_categories:100M<n<1B, ... | The Stack is a large multilingual code dataset built from source code across many programming languages and permissive software licenses. It is designed for code generation and related model training tasks, with near-deduplicated contents spanning terabytes of code data. This makes it well suited for training coding as... | huggingface | bigcode/the-stack |
Write a short dataset summary highlighting how the data was created and what it is best used for. | Name: TinyStories
Description: TinyStories contains synthetically generated short stories created by GPT-3.5 and GPT-4 using a small vocabulary. It is described in arXiv:2305.07759.
Tags: task_categories:text-generation, language:en, license:cdla-sharing-1.0, size_categories:1M<n<10M, format:parquet, modality:text
Down... | TinyStories is an English text generation dataset made up of short synthetic stories produced by GPT-3.5 and GPT-4 with intentionally simple vocabulary. It is particularly useful for studying small language models, language acquisition behavior, and efficient training on constrained narrative text. Its compact size and... | huggingface | roneneldan/TinyStories |
Summarize this dataset and identify the kinds of instruction-tuning tasks it supports. | Name: databricks-dolly-15k
Description: Databricks Dolly 15k is an open dataset of instruction-following records created by Databricks employees across categories such as brainstorming, classification, closed QA, generation, information extraction, open QA, and summarization.
Tags: task_categories:question-answering, t... | Databricks Dolly 15k is a human-created instruction-following dataset for supervised fine-tuning of language models. It includes a diverse set of tasks such as brainstorming, classification, question answering, information extraction, generation, and summarization. Because of its instruction-response structure and mode... | huggingface | databricks/databricks-dolly-15k |
Given a raw dataset metadata record, summarize the dataset for supervised fine-tuning use by identifying its domain, modality, purpose, scale, language, license, and notable characteristics. | {"source":"huggingface","id":"tatsu-lab/alpaca","name":"alpaca","description":"Dataset Card for Alpaca\n\nDataset Summary\n\nAlpaca 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 an... | Alpaca is an English text dataset for instruction fine-tuning and text generation. It contains about 52,000 instruction-response demonstrations generated with OpenAI's text-davinci-003 and based on the Self-Instruct pipeline. The dataset is suitable for training models to better follow instructions. It is distributed i... | huggingface | tatsu-lab/alpaca |
Given a raw dataset metadata record, summarize the dataset for supervised fine-tuning use by identifying its domain, modality, purpose, scale, language, license, and notable characteristics. | {"source":"huggingface","id":"Nerfgun3/bad_prompt","name":"bad_prompt","description":"Negative Embedding / Textual Inversion\n\nIdea\n\nThe idea behind this embedding was to somehow train the negative prompt as an embedding, thus unifying the basis of the negative prompt into one word or embedding. Side note: Embedding... | bad_prompt is a small English image-related dataset/resource associated with Stable Diffusion, textual inversion, and negative prompting. Its purpose is to provide a negative embedding that can be used in text-to-image and image-to-image workflows, including improving image generation details such as hands. It is tagge... | huggingface | Nerfgun3/bad_prompt |
Summarize the dataset and list its key characteristics for model training use. | 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. It is built through stringent filtering and large-scale deduplication of CommonCrawl. Models trained on RefinedWeb achie... | Falcon RefinedWeb is a large-scale English web text dataset from TII designed for text generation training. It is derived from CommonCrawl using strict filtering and large-scale deduplication, and is reported to support model performance comparable to or better than curated-data training. The dataset is licensed under ... | huggingface | tiiuae/falcon-refinedweb |
Summarize the dataset and list its key characteristics for conversational AI training use. | 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 conversational dataset for training and analyzing chat models. It contains one million conversations involving 25 state-of-the-art LLMs, collected from 210,000 unique IP addresses through the Vicuna demo and Chatbot Arena between April and August 2023. Each record includes structured conve... | huggingface | lmsys/lmsys-chat-1m |
Summarize the dataset metadata into a concise SFT-style description including purpose, source, notable characteristics, and basic popularity signals. | Name: ShareGPT_Vicuna_unfiltered
Source: huggingface
Dataset ID: anon8231489123/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: https://huggingface.co/AlekseyKorshuk/vicuna-7b/... | ShareGPT_Vicuna_unfiltered is a Hugging Face dataset focused on cleaned ShareGPT/Vicuna-style conversational data in English. The description highlights additional cleaning work and provides variants that either remove or retain instances of the phrase "I'm sorry, but," which may be relevant for instruction-tuning expe... | huggingface | anon8231489123/ShareGPT_Vicuna_unfiltered |
Summarize the dataset metadata into a concise SFT-style description including purpose, source, notable characteristics, and basic popularity signals. | Name: PhysicalAI-Autonomous-Vehicles
Source: huggingface
Dataset ID: nvidia/PhysicalAI-Autonomous-Vehicles
Description: PHYSICAL AI AUTONOMOUS VEHICLES The PhysicalAI-Autonomous-Vehicles dataset provides one of the largest, geographically diverse collections of multi-sensor data empowering AV researchers to build the n... | PhysicalAI-Autonomous-Vehicles is a Hugging Face dataset from NVIDIA for autonomous vehicle and physical AI research. It is described as one of the largest geographically diverse multi-sensor driving datasets, intended to support end-to-end driving systems. The dataset uses automatic or sensor-based data collection and... | huggingface | nvidia/PhysicalAI-Autonomous-Vehicles |
Summarize the dataset metadata into a concise SFT-style description highlighting what the dataset is, its scale, modality, licensing, and notable characteristics. | {"source":"huggingface","id":"HuggingFaceFW/finepdfs","name":"finepdfs","description":"Liberating 3T of the finest tokens from PDFs. FinePDFs is the largest publicly available corpus sourced exclusively from PDFs, containing about 3 trillion tokens across 475 million documents in 1733 languages.","tags":["task_categori... | finepdfs is a large-scale Hugging Face dataset built exclusively from PDF documents. It contains roughly 3 trillion tokens extracted from about 475 million documents spanning 1,733 languages, making it a major multilingual text resource for text generation and language model training. The dataset is distributed in Parq... | huggingface | HuggingFaceFW/finepdfs |
Summarize the dataset metadata into a concise SFT-style description highlighting what the dataset is, its scale, modality, licensing, and notable characteristics. | {"source":"huggingface","id":"open-thoughts/OpenThoughts-114k","name":"OpenThoughts-114k","description":"Open synthetic reasoning dataset with 114k high-quality examples covering math, science, code, and puzzles. Default subset contains ready-to-train data used to finetune the OpenThinker-7B and OpenThinker-32B models.... | OpenThoughts-114k is a synthetic reasoning dataset on Hugging Face containing 114,000 high-quality training examples. It covers domains such as mathematics, science, coding, and puzzles, and includes ready-to-train data used to finetune OpenThinker-7B and OpenThinker-32B. The dataset is text-based, provided in Parquet ... | huggingface | open-thoughts/OpenThoughts-114k |
Summarize the dataset for LLM supervised fine-tuning use, including what it is, its scale, modality, notable characteristics, and popularity signals. | Name: OpenHermes-2.5
Description: This is the dataset that made OpenHermes 2.5 and Nous Hermes 2 series of models. The Open Hermes 2/2.5 and Nous Hermes 2 models have made significant advancements of SOTA LLM's over recent months, and are underpinned by this exact compilation and curation of many open source datasets a... | OpenHermes-2.5 is a large English text dataset for LLM training distributed in JSON format. It is a compilation of open-source and synthetic data used to build the OpenHermes 2.5 and Nous Hermes 2 model series. The dataset is tagged as synthetic and associated with GPT-4 distillation and compilation workflows, suggesti... | huggingface | teknium/OpenHermes-2.5 |
Summarize the dataset for LLM supervised fine-tuning use, including what it is, its scale, modality, notable characteristics, and popularity signals. | Name: alpaca-cleaned
Description: This is a cleaned version of the original Alpaca Dataset released by Stanford. The following issues have been identified in the original release and fixed in this dataset: hallucinations caused by instructions referencing internet data.
Tags: task_categories:text-generation, language:e... | alpaca-cleaned is an English text-generation dataset designed for instruction fine-tuning. It is a cleaned version of Stanford's original Alpaca dataset, with fixes for problematic examples such as prompts that encouraged hallucinations by referencing unavailable internet information. The dataset is provided in JSON fo... | huggingface | yahma/alpaca-cleaned |
Summarize the dataset metadata into a concise SFT-style description. | {"source":"huggingface","id":"cais/hle","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 subjects.","tags... | HLE is an official Hugging Face benchmark dataset from cais/hle. It is a multimodal benchmark with image and text modalities, distributed in Parquet format under the MIT license. The dataset contains 2,500 frontier-level academic questions across many subjects, placing it in the 1K to 10K size category. It is available... | huggingface | cais/hle |
Summarize the dataset metadata into a concise SFT-style description. | {"source":"huggingface","id":"HuggingFaceFW/fineweb-2","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, released under the ODC-By 1.0 license, and validated through extensive ablation exp... | FineWeb2 is a Hugging Face dataset for text generation and large-scale multilingual pretraining from HuggingFaceFW/fineweb-2. It is the second iteration of FineWeb and offers high-quality, reproducible text data spanning more than 1000 languages. The dataset is licensed under ODC-By, includes text and tabular modalitie... | huggingface | HuggingFaceFW/fineweb-2 |
Summarize the dataset based on the provided metadata. | 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 synset. ImageNet aims to provide on ... | ImageNet-1k is a large-scale image classification dataset from the ILSVRC 2012 benchmark. It contains images organized by WordNet synsets and is commonly used for multi-class image classification tasks. The dataset is in English, image modality, distributed in parquet formats, and has a size between 1M and 10M examples... | huggingface | ILSVRC/imagenet-1k |
Summarize the dataset based on the provided metadata. | 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 supports empirical study on instruction-tuning datasets.
Tags: langu... | Alpaca-CoT is a collection of instruction-finetuning datasets standardized for direct use with Alpaca-style models. It focuses on instruction tuning and chain-of-thought style data, supports multiple languages including English and Chinese, and is released under the Apache-2.0 license. The dataset has gained strong com... | huggingface | QingyiSi/Alpaca-CoT |
Summarize the dataset based on its metadata. | Name: PersonaHub
Source: huggingface
ID: proj-persona/PersonaHub
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... | PersonaHub is a Hugging Face dataset focused on large-scale synthetic data creation through persona-driven generation. It is associated with the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas" and introduces PERSONA HUB, a collection of 1 billion diverse personas curated from web data to support var... | huggingface | proj-persona/PersonaHub |
Summarize the dataset for someone looking for community-curated prompt collections for LLM prompting and chatbot use. | Name: prompts.chat
Source: huggingface
ID: fka/prompts.chat
Description: a.k.a. Awesome ChatGPT Prompts. This is a Dataset Repository mirror of prompts.chat — a social platform for AI prompts. The platform lets users share, discover, and collect AI prompts from the community.
Tags: task_categories:question-answering, t... | prompts.chat is a community-driven prompt dataset mirrored on Hugging Face. It focuses on prompt sharing and discovery for LLMs and chatbots, with strong relevance to prompt engineering and conversational AI. The dataset is relatively small, text-based, distributed in CSV format, and licensed under CC0-1.0, making it e... | huggingface | fka/prompts.chat |
Provide a concise dataset overview for pretraining researchers evaluating large-scale English web corpora for LLM training. | Name: fineweb
Source: huggingface
ID: HuggingFaceFW/fineweb
Description: FineWeb contains more than 18.5T tokens of cleaned and deduplicated English web data from CommonCrawl. The processing pipeline is optimized for LLM performance and built with the datatrove large-scale data processing library.
Tags: task_categories... | FineWeb is a massive English web corpus for LLM pretraining, built from cleaned and deduplicated CommonCrawl data. With over 18.5 trillion tokens and a pipeline optimized for language model performance, it is well suited for large-scale text-generation research and foundation model training. Its scale and popularity ma... | huggingface | HuggingFaceFW/fineweb |
Explain what this dataset is useful for and note any limitation for supervised fine-tuning use. | Name: hh-rlhf
Source: huggingface
ID: Anthropic/hh-rlhf
Description: This repository provides human preference data about helpfulness and harmlessness from RLHF research. These data are meant to train preference or reward models for subsequent RLHF training and are not meant for supervised training of dialogue agents.
... | HH-RLHF is primarily a human preference dataset designed for reward modeling and RLHF research, especially around helpfulness and harmlessness. It is useful for training preference models rather than standard supervised chat assistants. A key limitation is that the creators explicitly caution against using it directly ... | huggingface | Anthropic/hh-rlhf |
Write a short description of this dataset for someone searching for instruction-tuning data for general-purpose LLMs. | Name: OpenOrca
Source: huggingface
ID: Open-Orca/OpenOrca
Description: OpenOrca is a rich collection of augmented FLAN data aligned as closely as possible with distributions described in the Orca paper. It has been used to generate strong model checkpoints and is intended as a resource for NLP researchers and developer... | OpenOrca is a large English instruction-style dataset built from augmented FLAN data and inspired by the Orca training setup. It supports a wide variety of NLP tasks, including question answering, summarization, classification, and text generation, making it a versatile resource for instruction tuning general-purpose L... | huggingface | Open-Orca/OpenOrca |
Summarize this dataset for multilingual alignment and assistant conversation fine-tuning use cases. | Name: oasst1
Source: huggingface
ID: OpenAssistant/oasst1
Description: OpenAssistant Conversations Dataset (OASST1) is a human-generated, human-annotated assistant-style conversation corpus with 161,443 messages in 35 languages, 461,292 quality ratings, and over 10,000 fully annotated conversation trees.
Tags: multiple... | OASST1 is a multilingual assistant conversation dataset designed for alignment research and chat model fine-tuning. It contains human-written, human-rated conversation trees across 35 languages, making it valuable for multilingual assistant training, response ranking, and quality-aware dialogue modeling. Its annotated ... | huggingface | OpenAssistant/oasst1 |
Summarize the dataset for supervised fine-tuning selection. | Name: gsm8k
Description: Dataset Card for GSM8K. GSM8K (Grade School Math 8K) is a dataset of 8.5K high quality linguistically diverse grade school math word problems created to support question answering on basic mathematical problems requiring multi-step reasoning. Problems typically take 2 to 8 steps and use element... | GSM8K is an English text dataset of grade school math word problems designed for multi-step reasoning and arithmetic question answering. It is a strong benchmark-style resource for SFT or evaluation on mathematical reasoning tasks. The dataset is MIT-licensed, medium-sized, and highly popular. | huggingface | openai/gsm8k |
Summarize the dataset for supervised fine-tuning selection. | Name: wikipedia
Description: Dataset Card for Wikimedia Wikipedia. Wikipedia dataset containing cleaned articles of all languages. Built from Wikipedia dumps with one subset per language, each containing a single train split. Each example contains the content of one full cleaned Wikipedia article.
Tags: task_categories... | Wikimedia Wikipedia is a very large multilingual text corpus of cleaned full Wikipedia articles across many languages. It is well suited for language modeling, masked language modeling, and general knowledge pretraining rather than narrow instruction tuning. Licensing includes CC-BY-SA-3.0 and GFDL, so downstream usage... | huggingface | wikimedia/wikipedia |
Summarize the dataset for supervised fine-tuning selection. | Name: EasyNegative
Description: Negative Embedding trained with Counterfeit. Intended for use in the stable-diffusion-webui embeddings folder. Can be used with other models, but effectiveness is uncertain.
Tags: license:other, size_categories:n<1K, format:imagefolder, modality:image
Downloads: 34931
Likes: 1174 | EasyNegative appears to be a small image-related asset centered on a negative embedding for Stable Diffusion workflows rather than a conventional text SFT dataset. It is likely unsuitable for LLM supervised fine-tuning in standard instruction-response format. The license is non-standard ('other'), so reuse terms should... | huggingface | gsdf/EasyNegative |
Summarize the dataset for supervised fine-tuning selection. | 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
Downloads: 2154
Likes: 1152 | RedPajama-Data-1T is a large English text corpus intended as an open-source reproduction of the data used for LLaMA-style model pretraining. It is best suited for foundation model pretraining or continued pretraining, not directly for instruction SFT unless additional prompt-response formatting is applied. Its value is... | huggingface | togethercomputer/RedPajama-Data-1T |
Summarize the dataset for supervised fine-tuning selection. | Name: medical-o1-reasoning-SFT
Description: Medical reasoning dataset for SFT built on medical verifiable problems and distilled reasoning data. News indicates the dataset now keeps only the medical SFT file and includes reasoning chains derived from Deepseek-R1-based distillation.
Tags: task_categories:question-answer... | medical-o1-reasoning-SFT is a bilingual English-Chinese medical reasoning dataset explicitly designed for supervised fine-tuning. It contains medical instruction and reasoning-oriented data based on verifiable problems, making it highly relevant for training domain-specific reasoning assistants. The Apache-2.0 license ... | huggingface | FreedomIntelligence/medical-o1-reasoning-SFT |
Summarize the dataset and explain why it is useful for language model training. | Name: fineweb-edu
Description: FineWeb-Edu is a 1.3 trillion token educational web dataset filtered from FineWeb using an educational quality classifier. It focuses on high-quality educational content from the web.
Tags: task_categories:text-generation, language:en, license:odc-by, size_categories:1B<n<10B, format:parq... | FineWeb-Edu is a large-scale English educational text dataset designed for text generation and language model pretraining. It contains 1.3 trillion tokens of web pages selected for educational quality from the broader FineWeb corpus. Because it emphasizes high-quality educational content and is distributed in scalable ... | huggingface | HuggingFaceFW/fineweb-edu |
Provide a concise overview of this dataset including its scale, purpose, and licensing. | Name: dolma
Description: Dolma is 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, language-modeling, casual-lm, llm
Downloads: 3153
Likes: 1022 | Dolma is a massive open English text corpus created for language model pretraining research. With roughly three trillion tokens, it is intended for large-scale language modeling and LLM development. The dataset is licensed under ODC-BY, making it suitable for open research workflows that require a very large pretrainin... | huggingface | allenai/dolma |
Describe this dataset in terms of content, scope, and ideal use cases. | Name: the-stack
Description: The Stack is a large code dataset. Its releases include many programming languages and permissive licenses, with near-deduplicated data totaling multiple terabytes.
Tags: task_categories:text-generation, multilinguality:multilingual, language:code, license:other, size_categories:100M<n<1B, ... | The Stack is a large multilingual code dataset built from source code across many programming languages and permissive software licenses. It is designed for code generation and related model training tasks, with near-deduplicated contents spanning terabytes of code data. This makes it well suited for training coding as... | huggingface | bigcode/the-stack |
Write a short dataset summary highlighting how the data was created and what it is best used for. | Name: TinyStories
Description: TinyStories contains synthetically generated short stories created by GPT-3.5 and GPT-4 using a small vocabulary. It is described in arXiv:2305.07759.
Tags: task_categories:text-generation, language:en, license:cdla-sharing-1.0, size_categories:1M<n<10M, format:parquet, modality:text
Down... | TinyStories is an English text generation dataset made up of short synthetic stories produced by GPT-3.5 and GPT-4 with intentionally simple vocabulary. It is particularly useful for studying small language models, language acquisition behavior, and efficient training on constrained narrative text. Its compact size and... | huggingface | roneneldan/TinyStories |
Summarize this dataset and identify the kinds of instruction-tuning tasks it supports. | Name: databricks-dolly-15k
Description: Databricks Dolly 15k is an open dataset of instruction-following records created by Databricks employees across categories such as brainstorming, classification, closed QA, generation, information extraction, open QA, and summarization.
Tags: task_categories:question-answering, t... | Databricks Dolly 15k is a human-created instruction-following dataset for supervised fine-tuning of language models. It includes a diverse set of tasks such as brainstorming, classification, question answering, information extraction, generation, and summarization. Because of its instruction-response structure and mode... | huggingface | databricks/databricks-dolly-15k |
Given a raw dataset metadata record, summarize the dataset for supervised fine-tuning use by identifying its domain, modality, purpose, scale, language, license, and notable characteristics. | {"source":"huggingface","id":"tatsu-lab/alpaca","name":"alpaca","description":"Dataset Card for Alpaca\n\nDataset Summary\n\nAlpaca 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 an... | Alpaca is an English text dataset for instruction fine-tuning and text generation. It contains about 52,000 instruction-response demonstrations generated with OpenAI's text-davinci-003 and based on the Self-Instruct pipeline. The dataset is suitable for training models to better follow instructions. It is distributed i... | huggingface | tatsu-lab/alpaca |
Given a raw dataset metadata record, summarize the dataset for supervised fine-tuning use by identifying its domain, modality, purpose, scale, language, license, and notable characteristics. | {"source":"huggingface","id":"Nerfgun3/bad_prompt","name":"bad_prompt","description":"Negative Embedding / Textual Inversion\n\nIdea\n\nThe idea behind this embedding was to somehow train the negative prompt as an embedding, thus unifying the basis of the negative prompt into one word or embedding. Side note: Embedding... | bad_prompt is a small English image-related dataset/resource associated with Stable Diffusion, textual inversion, and negative prompting. Its purpose is to provide a negative embedding that can be used in text-to-image and image-to-image workflows, including improving image generation details such as hands. It is tagge... | huggingface | Nerfgun3/bad_prompt |
Summarize the dataset and list its key characteristics for model training use. | 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. It is built through stringent filtering and large-scale deduplication of CommonCrawl. Models trained on RefinedWeb achie... | Falcon RefinedWeb is a large-scale English web text dataset from TII designed for text generation training. It is derived from CommonCrawl using strict filtering and large-scale deduplication, and is reported to support model performance comparable to or better than curated-data training. The dataset is licensed under ... | huggingface | tiiuae/falcon-refinedweb |
Summarize the dataset and list its key characteristics for conversational AI training use. | 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 conversational dataset for training and analyzing chat models. It contains one million conversations involving 25 state-of-the-art LLMs, collected from 210,000 unique IP addresses through the Vicuna demo and Chatbot Arena between April and August 2023. Each record includes structured conve... | huggingface | lmsys/lmsys-chat-1m |
Summarize the dataset metadata into a concise SFT-style description including purpose, source, notable characteristics, and basic popularity signals. | Name: ShareGPT_Vicuna_unfiltered
Source: huggingface
Dataset ID: anon8231489123/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: https://huggingface.co/AlekseyKorshuk/vicuna-7b/... | ShareGPT_Vicuna_unfiltered is a Hugging Face dataset focused on cleaned ShareGPT/Vicuna-style conversational data in English. The description highlights additional cleaning work and provides variants that either remove or retain instances of the phrase "I'm sorry, but," which may be relevant for instruction-tuning expe... | huggingface | anon8231489123/ShareGPT_Vicuna_unfiltered |
Summarize the dataset metadata into a concise SFT-style description including purpose, source, notable characteristics, and basic popularity signals. | Name: PhysicalAI-Autonomous-Vehicles
Source: huggingface
Dataset ID: nvidia/PhysicalAI-Autonomous-Vehicles
Description: PHYSICAL AI AUTONOMOUS VEHICLES The PhysicalAI-Autonomous-Vehicles dataset provides one of the largest, geographically diverse collections of multi-sensor data empowering AV researchers to build the n... | PhysicalAI-Autonomous-Vehicles is a Hugging Face dataset from NVIDIA for autonomous vehicle and physical AI research. It is described as one of the largest geographically diverse multi-sensor driving datasets, intended to support end-to-end driving systems. The dataset uses automatic or sensor-based data collection and... | huggingface | nvidia/PhysicalAI-Autonomous-Vehicles |
Summarize the dataset metadata into a concise SFT-style description highlighting what the dataset is, its scale, modality, licensing, and notable characteristics. | {"source":"huggingface","id":"HuggingFaceFW/finepdfs","name":"finepdfs","description":"Liberating 3T of the finest tokens from PDFs. FinePDFs is the largest publicly available corpus sourced exclusively from PDFs, containing about 3 trillion tokens across 475 million documents in 1733 languages.","tags":["task_categori... | finepdfs is a large-scale Hugging Face dataset built exclusively from PDF documents. It contains roughly 3 trillion tokens extracted from about 475 million documents spanning 1,733 languages, making it a major multilingual text resource for text generation and language model training. The dataset is distributed in Parq... | huggingface | HuggingFaceFW/finepdfs |
Summarize the dataset metadata into a concise SFT-style description highlighting what the dataset is, its scale, modality, licensing, and notable characteristics. | {"source":"huggingface","id":"open-thoughts/OpenThoughts-114k","name":"OpenThoughts-114k","description":"Open synthetic reasoning dataset with 114k high-quality examples covering math, science, code, and puzzles. Default subset contains ready-to-train data used to finetune the OpenThinker-7B and OpenThinker-32B models.... | OpenThoughts-114k is a synthetic reasoning dataset on Hugging Face containing 114,000 high-quality training examples. It covers domains such as mathematics, science, coding, and puzzles, and includes ready-to-train data used to finetune OpenThinker-7B and OpenThinker-32B. The dataset is text-based, provided in Parquet ... | huggingface | open-thoughts/OpenThoughts-114k |
Summarize the dataset for LLM supervised fine-tuning use, including what it is, its scale, modality, notable characteristics, and popularity signals. | Name: OpenHermes-2.5
Description: This is the dataset that made OpenHermes 2.5 and Nous Hermes 2 series of models. The Open Hermes 2/2.5 and Nous Hermes 2 models have made significant advancements of SOTA LLM's over recent months, and are underpinned by this exact compilation and curation of many open source datasets a... | OpenHermes-2.5 is a large English text dataset for LLM training distributed in JSON format. It is a compilation of open-source and synthetic data used to build the OpenHermes 2.5 and Nous Hermes 2 model series. The dataset is tagged as synthetic and associated with GPT-4 distillation and compilation workflows, suggesti... | huggingface | teknium/OpenHermes-2.5 |
Summarize the dataset for LLM supervised fine-tuning use, including what it is, its scale, modality, notable characteristics, and popularity signals. | Name: alpaca-cleaned
Description: This is a cleaned version of the original Alpaca Dataset released by Stanford. The following issues have been identified in the original release and fixed in this dataset: hallucinations caused by instructions referencing internet data.
Tags: task_categories:text-generation, language:e... | alpaca-cleaned is an English text-generation dataset designed for instruction fine-tuning. It is a cleaned version of Stanford's original Alpaca dataset, with fixes for problematic examples such as prompts that encouraged hallucinations by referencing unavailable internet information. The dataset is provided in JSON fo... | huggingface | yahma/alpaca-cleaned |
Summarize the dataset metadata into a concise SFT-style description. | {"source":"huggingface","id":"cais/hle","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 subjects.","tags... | HLE is an official Hugging Face benchmark dataset from cais/hle. It is a multimodal benchmark with image and text modalities, distributed in Parquet format under the MIT license. The dataset contains 2,500 frontier-level academic questions across many subjects, placing it in the 1K to 10K size category. It is available... | huggingface | cais/hle |
Summarize the dataset metadata into a concise SFT-style description. | {"source":"huggingface","id":"HuggingFaceFW/fineweb-2","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, released under the ODC-By 1.0 license, and validated through extensive ablation exp... | FineWeb2 is a Hugging Face dataset for text generation and large-scale multilingual pretraining from HuggingFaceFW/fineweb-2. It is the second iteration of FineWeb and offers high-quality, reproducible text data spanning more than 1000 languages. The dataset is licensed under ODC-By, includes text and tabular modalitie... | huggingface | HuggingFaceFW/fineweb-2 |
Summarize the dataset based on the provided metadata. | 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 synset. ImageNet aims to provide on ... | ImageNet-1k is a large-scale image classification dataset from the ILSVRC 2012 benchmark. It contains images organized by WordNet synsets and is commonly used for multi-class image classification tasks. The dataset is in English, image modality, distributed in parquet formats, and has a size between 1M and 10M examples... | huggingface | ILSVRC/imagenet-1k |
Summarize the dataset based on the provided metadata. | 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 supports empirical study on instruction-tuning datasets.
Tags: langu... | Alpaca-CoT is a collection of instruction-finetuning datasets standardized for direct use with Alpaca-style models. It focuses on instruction tuning and chain-of-thought style data, supports multiple languages including English and Chinese, and is released under the Apache-2.0 license. The dataset has gained strong com... | huggingface | QingyiSi/Alpaca-CoT |
Summarize the dataset based on its metadata. | Name: PersonaHub
Source: huggingface
ID: proj-persona/PersonaHub
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... | PersonaHub is a Hugging Face dataset focused on large-scale synthetic data creation through persona-driven generation. It is associated with the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas" and introduces PERSONA HUB, a collection of 1 billion diverse personas curated from web data to support var... | huggingface | proj-persona/PersonaHub |
SoloAI SFT Dataset: 20260427-2136
📊 数据集概览
| 属性 | 值 |
|---|---|
| 数据条数 | 30 条 |
| 数据源 | huggingface |
| 生成时间 | 2026-04-27 21:36 |
| 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': 'Summarize the dataset for someone looking for community-curated prompt collections for LLM prompting and chatbot use.', 'input': 'Name: prompts.chat\nSource: huggingface\nID: fka/prompts.chat\nDescription: a.k.a. Awesome ChatGPT Prompts. This is a Dataset Repository mirror of prompts.chat — a social platform for AI prompts. The platform lets users share, discover, and collect AI prompts from the community.\nTags: task_categories:question-answering, task_categories:text-generation, license:cc0-1.0, size_categories:1K<n<10K, format:csv, modality:text, ChatGPT, prompts, AI, GPT, Claude, Gemini, Llama, Mistral, LLM, prompt-engineering, conversational-ai, text-generation, chatbot, awesome-list\nDownloads: 53719\nLikes: 9680', 'output': 'prompts.chat is a community-driven prompt dataset mirrored on Hugging Face. It focuses on prompt sharing and discovery for LLMs and chatbots, with strong relevance to prompt engineering and conversational AI. The dataset is relatively small, text-based, distributed in CSV format, and licensed under CC0-1.0, making it easy to reuse for prompt collection, inspiration, and experimentation.', '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 | 2026-04-27 21:36 | 初始发布，30 条数据 |
- Downloads last month
- 32