instruction stringlengths 53 153 | input stringlengths 212 1.03k | output stringlengths 293 744 | source stringclasses 1
value | original_id stringlengths 8 44 |
|---|---|---|---|---|
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | 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_categ... | prompts.chat is a community prompt dataset mirrored from the prompts.chat platform, also known as Awesome ChatGPT Prompts. It focuses on user-created prompts for conversational AI and text generation systems such as ChatGPT, Claude, Gemini, Llama, and Mistral. The dataset is relatively small, in the 1K to 10K range, st... | huggingface | fka/prompts.chat |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | Name: fineweb
Description: FineWeb consists of more than 18.5T tokens of cleaned and deduplicated English web data from CommonCrawl. The processing pipeline is optimized for LLM performance and ran on the datatrove library. It was originally meant to be a fully open replication of RefinedWeb.
Tags: task_categories:text... | FineWeb is a massive English web corpus designed for large-scale language model pretraining. It contains over 18.5 trillion cleaned and deduplicated tokens derived from CommonCrawl, with a processing pipeline optimized for LLM performance using the datatrove library. The dataset is licensed under ODC-BY and is extremel... | huggingface | HuggingFaceFW/fineweb |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | Name: hh-rlhf
Description: This repository provides access to human preference data about helpfulness and harmlessness from Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback. These data are meant to train preference or reward models for subsequent RLHF training and are not meant ... | HH-RLHF is a human feedback dataset created for training preference or reward models in RLHF pipelines, with a focus on helpfulness and harmlessness. It is not intended for standard supervised fine-tuning of chat assistants, since doing so may produce undesirable behaviors. The dataset is text-based, provided in JSON f... | huggingface | Anthropic/hh-rlhf |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | 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 is intended as a resource for NLP researchers and developers.
Tags: ... | OpenOrca is a large English instruction-tuning dataset built from augmented FLAN-style data and designed to approximate the distributions described in the Orca work. It is intended to support training high-performing general-purpose assistant models and has already been used for strong model checkpoints. The dataset is... | huggingface | Open-Orca/OpenOrca |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | Name: oasst1
Description: OpenAssistant Conversations Dataset (OASST1) is 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 annotated conversation trees. It was created through a w... | OASST1 is a multilingual assistant conversation dataset created through large-scale crowdsourcing for open alignment research. It contains 161,443 messages across 35 languages, along with hundreds of thousands of quality ratings and more than 10,000 annotated conversation trees. The dataset is licensed under Apache-2.0... | huggingface | OpenAssistant/oasst1 |
Summarize the dataset in a concise way for a machine learning practitioner. | Name: gsm8k
Description: 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 that require multi-step reasoning. Problems typically take 2 to 8 steps and use elementary arithmetic operat... | GSM8K is an English benchmark dataset of about 8.5K grade-school math word problems designed for multi-step reasoning. It is commonly used for evaluating question answering and text-generation models on arithmetic reasoning tasks. The dataset is MIT-licensed and focused on math word problems. | huggingface | openai/gsm8k |
Summarize the dataset in a concise way for a machine learning practitioner. | Name: wikipedia
Description: Wikipedia dataset containing cleaned articles of all languages. The dataset is built from Wikipedia dumps with one subset per language, each containing a single train split. Each example contains the content of one full Wikipedia article with cleaning to strip markdown and unwanted sections... | Wikimedia Wikipedia is a large multilingual corpus of cleaned full Wikipedia articles derived from official dumps. It supports language modeling and masked language modeling, with one subset per language. The dataset is useful for pretraining and general NLP research, and is distributed under CC-BY-SA 3.0 and GFDL-rela... | huggingface | wikimedia/wikipedia |
Summarize the dataset or asset in a concise way for a machine learning practitioner. | Name: EasyNegative
Description: Negative Embedding trained with Counterfeit. Intended for use in the stable-diffusion-webui embeddings folder. It may work with other models, though effectiveness is uncertain.
Tags: license:other, format:imagefolder, modality:image
Downloads: 31720
Likes: 1174 | EasyNegative is an image-related negative embedding asset intended for Stable Diffusion workflows, especially via the stable-diffusion-webui embeddings directory. It is designed to be used with certain image generation models, though cross-model effectiveness may vary. Its licensing is marked as other. | huggingface | gsdf/EasyNegative |
Summarize the dataset in a concise way for a machine learning practitioner. | 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, modality:text
Downloads: 2169
Likes: 1147 | RedPajama-Data-1T is a large English text dataset intended for text generation and language model training. It was created as a clean-room, fully open-source reproduction of the data used for LLaMA-style model development. It is primarily useful for large-scale pretraining and open LLM research. | huggingface | togethercomputer/RedPajama-Data-1T |
Summarize the dataset in a concise way for a machine learning practitioner. | Name: medical-o1-reasoning-SFT
Description: Medical reasoning dataset for supervised fine-tuning built on medical verifiable problems and distilled reasoning data. The dataset includes medical SFT data and references releases focused on reasoning chains derived from stronger models.
Tags: task_categories:question-answe... | medical-o1-reasoning-SFT is a bilingual English and Chinese medical reasoning dataset for supervised fine-tuning. It focuses on medical verifiable problems and reasoning-oriented QA or generation tasks, making it suitable for training models on structured medical reasoning. The dataset is Apache-2.0 licensed and releva... | huggingface | FreedomIntelligence/medical-o1-reasoning-SFT |
Summarize the dataset for LLM pretraining use, including its scale, domain, language, license, format, and notable metadata. | {"source":"huggingface","id":"HuggingFaceFW/fineweb-edu","name":"fineweb-edu","description":"📚 FineWeb-Edu\n1.3 trillion tokens of the finest educational data the web has to offer. FineWeb-Edu dataset consists of 1.3T tokens and 5.4T tokens (FineWeb-Edu-score-2) of educational web pages filtered from FineWeb dataset. ... | fineweb-edu is a large English educational web-text dataset from Hugging Face, intended for text generation and LLM pretraining. It contains about 1.3 trillion tokens of educational web pages filtered from FineWeb, with quality improved using a classifier trained from Llama3-70B-Instruct-generated annotations. The data... | huggingface | HuggingFaceFW/fineweb-edu |
Provide a concise dataset description suitable for supervised fine-tuning metadata, covering purpose, scale, language, license, and popularity signals. | {"source":"huggingface","id":"allenai/dolma","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-m... | dolma is an open English corpus designed for language model pretraining research. It contains roughly three trillion tokens and is positioned as a large-scale resource for language modeling and LLM development. The dataset is licensed under ODC-BY and is associated with research including arXiv:2402.00159. Based on its... | huggingface | allenai/dolma |
Summarize the dataset with emphasis on content type, scale, licensing characteristics, multilingual/code aspects, and intended use in model training. | {"source":"huggingface","id":"bigcode/the-stack","name":"the-stack","description":"Dataset Card for The Stack. Initial release included 30 programming languages and 18 permissive licenses, resulting in a near-deduplicated dataset of 3TB. Later versions excluded weak copyleft licenses (MPL/EPL/LGPL) and extended the lis... | the-stack is a large multilingual code dataset created for text generation and code model training. It consists of source code collected across many programming languages, with a near-deduplicated release reported at roughly 3TB. The dataset emphasizes permissive licensing, with later versions removing weak copyleft li... | huggingface | bigcode/the-stack |
Create a short structured summary of the dataset, mentioning data generation method, task relevance, scale, language, license, and common training usage. | {"source":"huggingface","id":"roneneldan/TinyStories","name":"TinyStories","description":"Dataset containing synthetically generated (by GPT-3.5 and GPT-4) short stories that only use a small vocabulary. Described in arXiv:2305.07759. The models referred to in the paper were trained on TinyStories-train.txt, with tinys... | TinyStories is an English text-generation dataset of synthetically generated short stories created by GPT-3.5 and GPT-4. The stories are constrained to a small vocabulary, making the dataset especially useful for studying small language models and language acquisition-like settings. It is relatively small compared with... | huggingface | roneneldan/TinyStories |
Summarize this instruction-tuning dataset, including task types, data origin, size, license, format, and how it can be used for SFT. | {"source":"huggingface","id":"databricks/databricks-dolly-15k","name":"databricks-dolly-15k","description":"databricks-dolly-15k is an open source dataset of instruction-following records generated by thousands of Databricks employees in several behavioral categories outlined in the InstructGPT paper, including brainst... | databricks-dolly-15k is an English instruction-following dataset designed for supervised fine-tuning. It contains about 15,000 human-generated records created by Databricks employees across tasks such as brainstorming, classification, closed and open question answering, generation, information extraction, and summariza... | huggingface | databricks/databricks-dolly-15k |
Summarize the dataset metadata into a concise SFT training example. | {"source":"huggingface","id":"tatsu-lab/alpaca","name":"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 m... | {"name":"alpaca","summary":"A popular English instruction-tuning dataset with 52K instruction-response examples generated using OpenAI text-davinci-003 and inspired by the Self-Instruct pipeline.","modality":"text","language":"en","license":"cc-by-nc-4.0","size":"10K<n<100K","primary_use":"instruction-finetuning","popu... | huggingface | tatsu-lab/alpaca |
Summarize the dataset metadata into a concise SFT training example. | {"source":"huggingface","id":"Nerfgun3/bad_prompt","name":"bad_prompt","description":"Negative Embedding / Textual Inversion. The 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.","tags":["language:en","licens... | {"name":"bad_prompt","summary":"A small image-related Hugging Face dataset/resource associated with negative embedding or textual inversion for Stable Diffusion workflows, intended to improve generation quality via a reusable negative prompt embedding.","modality":"image","language":"en","license":"creativeml-openrail-... | huggingface | Nerfgun3/bad_prompt |
Summarize the dataset metadata into a concise SFT training example. | {"source":"huggingface","id":"tiiuae/falcon-refinedweb","name":"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 Refi... | {"name":"falcon-refinedweb","summary":"A massive English web-scale text dataset derived from CommonCrawl, heavily filtered and deduplicated for large language model pretraining, with multimodal-friendly web content references.","modality":"text","language":"en","license":"odc-by","size":"100M<n<1B","primary_use":"LLM p... | huggingface | tiiuae/falcon-refinedweb |
Summarize the dataset metadata into a concise SFT training example. | {"source":"huggingface","id":"lmsys/lmsys-chat-1m","name":"lmsys-chat-1m","description":"LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset. This dataset contains one million real-world conversations with 25 state-of-the-art LLMs, collected from 210K unique IP addresses on the Vicuna demo and Chatbot Aren... | {"name":"lmsys-chat-1m","summary":"A large-scale real-world conversation dataset containing around one million chats with 25 LLMs, sourced from public demo and arena platforms and useful for conversational AI analysis and training.","modality":"text","language":"unknown/varied","license":"not specified","size":"1M<n<10... | huggingface | lmsys/lmsys-chat-1m |
Summarize the dataset metadata into a concise SFT training example. | {"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 includes confirmed working method for training the model and multiple cleaned dataset variants... | {"name":"ShareGPT_Vicuna_unfiltered","summary":"An English conversational dataset derived from ShareGPT/Vicuna-style data with additional cleaning and variants intended for chat model training and experimentation.","modality":"text","language":"en","license":"apache-2.0","size":"not specified","primary_use":"chat fine-... | huggingface | anon8231489123/ShareGPT_Vicuna_unfiltered |
Summarize the dataset for supervised fine-tuning metadata extraction. | 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_categories:text-generation, license:odc-by, size_categories:100M<n<1B, ... | {"name":"finepdfs","provider":"Hugging Face","summary":"A massive multilingual text corpus extracted exclusively from PDFs, designed for large-scale language modeling and text generation research.","license":"odc-by","task_category":"text-generation","languages":"1733 languages","formats":["parquet"],"modalities":["tab... | huggingface | HuggingFaceFW/finepdfs |
Summarize the dataset for supervised fine-tuning metadata extraction. | 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 end-to-end driving systems. It contains a total of 1700 hours of driving and is available for commercial/no... | {"name":"PhysicalAI-Autonomous-Vehicles","provider":"Hugging Face","summary":"A large-scale, geographically diverse autonomous vehicle dataset with multi-sensor driving data for end-to-end driving research and deployment.","license":"other","data_collection":"automatic/sensor","labeling_method":"automatic/sensor","dura... | huggingface | nvidia/PhysicalAI-Autonomous-Vehicles |
Summarize the dataset for supervised fine-tuning metadata extraction. | 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 OpenThinker models.
Tags: license:apache-2.0, size_categories:100K<n<1M, format:parquet, modality:text, region... | {"name":"OpenThoughts-114k","provider":"Hugging Face","summary":"A synthetic reasoning dataset with 114k training examples spanning math, science, coding, and puzzle tasks for model finetuning.","license":"apache-2.0","domains":["math","science","code","puzzles"],"synthetic":true,"formats":["parquet"],"modalities":["te... | huggingface | open-thoughts/OpenThoughts-114k |
Summarize the dataset for supervised fine-tuning metadata extraction. | Name: alpaca-cleaned
Description: This is a cleaned version of the original Alpaca Dataset released by Stanford. It fixes issues in the original release such as hallucination-prone instructions referencing internet data.
Tags: task_categories:text-generation, language:en, license:cc-by-4.0, size_categories:10K<n<100K, ... | {"name":"alpaca-cleaned","provider":"Hugging Face","summary":"A cleaned English instruction-tuning dataset based on Stanford Alpaca, with problematic and hallucination-inducing examples corrected or removed.","license":"cc-by-4.0","task_category":"text-generation","language":"en","use_case":"instruction-finetuning","fo... | huggingface | yahma/alpaca-cleaned |
Summarize the dataset for supervised fine-tuning metadata extraction. | Name: OpenHermes-2.5
Description: This dataset underpins OpenHermes 2.5 and Nous Hermes 2 models. It is a compilation and curation of many open source datasets and custom synthetic datasets.
Tags: language:eng, size_categories:1M<n<10M, format:json, modality:text, region:us, synthetic, GPT-4, Distillation, Compilation
... | {"name":"OpenHermes-2.5","provider":"Hugging Face","summary":"A large compiled instruction-tuning corpus used for OpenHermes and Nous Hermes models, combining open-source and synthetic data with GPT-4 distillation.","language":"eng","synthetic":true,"techniques":["GPT-4 distillation","dataset compilation"],"formats":["... | huggingface | teknium/OpenHermes-2.5 |
Summarize the dataset fineweb-2 for SFT metadata ingestion. | Source: huggingface
ID: HuggingFaceFW/fineweb-2
Name: fineweb-2
Description: FineWeb2 is the second iteration of the FineWeb dataset, providing high-quality reproducible pretraining data across 1000+ languages under the ODC-By 1.0 license, validated through extensive ablation experiments.
Tags: task_categories:text-gen... | {"name":"fineweb-2","summary":"A large-scale multilingual pretraining dataset covering over 1000 languages, designed for high-quality and reproducible text generation research.","task_categories":["text-generation"],"languages":"1000+ languages","license":"odc-by","modalities":["tabular","text"],"size_category":"1B<n<1... | huggingface | HuggingFaceFW/fineweb-2 |
Summarize the dataset hle for SFT metadata ingestion. | Source: huggingface
ID: cais/hle
Name: hle
Description: Humanity's Last Exam (HLE) is a multimodal benchmark at the frontier of human knowledge, designed as a closed-ended academic benchmark with broad subject coverage and 2,500 questions.
Tags: benchmark:official; license:mit; size_categories:1K<n<10K; format:parquet;... | {"name":"hle","summary":"Humanity's Last Exam is an official multimodal benchmark with 2,500 closed-ended questions spanning broad academic subjects at the frontier of human knowledge.","benchmark":true,"modalities":["image","text"],"license":"mit","format":"parquet","size_category":"1K<n<10K","downloads":47598,"likes"... | huggingface | cais/hle |
Summarize the dataset imagenet-1k for SFT metadata ingestion. | Source: huggingface
ID: ILSVRC/imagenet-1k
Name: imagenet-1k
Description: ImageNet (ILSVRC 2012) is a large image dataset organized according to the WordNet hierarchy, with roughly 1000 images per synset and widely used for image classification.
Tags: task_categories:image-classification; task_ids:multi-class-image-cla... | {"name":"imagenet-1k","summary":"A canonical large-scale image classification dataset from ILSVRC 2012, organized by WordNet synsets and widely used for multi-class image classification.","task_categories":["image-classification"],"task_id":"multi-class-image-classification","language":"en","license":"other","modalitie... | huggingface | ILSVRC/imagenet-1k |
Summarize the dataset Alpaca-CoT for SFT metadata ingestion. | Source: huggingface
ID: QingyiSi/Alpaca-CoT
Name: Alpaca-CoT
Description: Alpaca-CoT is a continuously collected instruction-finetuning dataset collection that standardizes multiple instruction tuning datasets into a unified format compatible with Alpaca-style model training.
Tags: language:en; language:zh; language:ml... | {"name":"Alpaca-CoT","summary":"A standardized collection of instruction-tuning datasets for Alpaca-style finetuning, including chain-of-thought oriented data in multiple languages.","use_case":"instruction-finetuning","languages":["en","zh","ml"],"license":"apache-2.0","keywords":["Instruction","Cot"],"downloads":7554... | huggingface | QingyiSi/Alpaca-CoT |
Summarize the dataset PersonaHub for SFT metadata ingestion. | Source: huggingface
ID: proj-persona/PersonaHub
Name: PersonaHub
Description: PersonaHub releases data from the paper Scaling Synthetic Data Creation with 1,000,000,000 Personas, introducing a persona-driven synthetic data methodology and a collection of 1 billion diverse personas curated from web data.
Tags: task_cate... | {"name":"PersonaHub","summary":"A synthetic text dataset resource built around persona-driven data generation, releasing large-scale diverse personas for creating instruction, reasoning, math, and tool-use data.","task_categories":["text-generation","text-classification","token-classification","fill-mask","table-questi... | huggingface | proj-persona/PersonaHub |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | 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_categ... | prompts.chat is a community prompt dataset mirrored from the prompts.chat platform, also known as Awesome ChatGPT Prompts. It focuses on user-created prompts for conversational AI and text generation systems such as ChatGPT, Claude, Gemini, Llama, and Mistral. The dataset is relatively small, in the 1K to 10K range, st... | huggingface | fka/prompts.chat |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | Name: fineweb
Description: FineWeb consists of more than 18.5T tokens of cleaned and deduplicated English web data from CommonCrawl. The processing pipeline is optimized for LLM performance and ran on the datatrove library. It was originally meant to be a fully open replication of RefinedWeb.
Tags: task_categories:text... | FineWeb is a massive English web corpus designed for large-scale language model pretraining. It contains over 18.5 trillion cleaned and deduplicated tokens derived from CommonCrawl, with a processing pipeline optimized for LLM performance using the datatrove library. The dataset is licensed under ODC-BY and is extremel... | huggingface | HuggingFaceFW/fineweb |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | Name: hh-rlhf
Description: This repository provides access to human preference data about helpfulness and harmlessness from Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback. These data are meant to train preference or reward models for subsequent RLHF training and are not meant ... | HH-RLHF is a human feedback dataset created for training preference or reward models in RLHF pipelines, with a focus on helpfulness and harmlessness. It is not intended for standard supervised fine-tuning of chat assistants, since doing so may produce undesirable behaviors. The dataset is text-based, provided in JSON f... | huggingface | Anthropic/hh-rlhf |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | 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 is intended as a resource for NLP researchers and developers.
Tags: ... | OpenOrca is a large English instruction-tuning dataset built from augmented FLAN-style data and designed to approximate the distributions described in the Orca work. It is intended to support training high-performing general-purpose assistant models and has already been used for strong model checkpoints. The dataset is... | huggingface | Open-Orca/OpenOrca |
Summarize the dataset for LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics. | Name: oasst1
Description: OpenAssistant Conversations Dataset (OASST1) is 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 annotated conversation trees. It was created through a w... | OASST1 is a multilingual assistant conversation dataset created through large-scale crowdsourcing for open alignment research. It contains 161,443 messages across 35 languages, along with hundreds of thousands of quality ratings and more than 10,000 annotated conversation trees. The dataset is licensed under Apache-2.0... | huggingface | OpenAssistant/oasst1 |
Summarize the dataset in a concise way for a machine learning practitioner. | Name: gsm8k
Description: 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 that require multi-step reasoning. Problems typically take 2 to 8 steps and use elementary arithmetic operat... | GSM8K is an English benchmark dataset of about 8.5K grade-school math word problems designed for multi-step reasoning. It is commonly used for evaluating question answering and text-generation models on arithmetic reasoning tasks. The dataset is MIT-licensed and focused on math word problems. | huggingface | openai/gsm8k |
Summarize the dataset in a concise way for a machine learning practitioner. | Name: wikipedia
Description: Wikipedia dataset containing cleaned articles of all languages. The dataset is built from Wikipedia dumps with one subset per language, each containing a single train split. Each example contains the content of one full Wikipedia article with cleaning to strip markdown and unwanted sections... | Wikimedia Wikipedia is a large multilingual corpus of cleaned full Wikipedia articles derived from official dumps. It supports language modeling and masked language modeling, with one subset per language. The dataset is useful for pretraining and general NLP research, and is distributed under CC-BY-SA 3.0 and GFDL-rela... | huggingface | wikimedia/wikipedia |
Summarize the dataset or asset in a concise way for a machine learning practitioner. | Name: EasyNegative
Description: Negative Embedding trained with Counterfeit. Intended for use in the stable-diffusion-webui embeddings folder. It may work with other models, though effectiveness is uncertain.
Tags: license:other, format:imagefolder, modality:image
Downloads: 31720
Likes: 1174 | EasyNegative is an image-related negative embedding asset intended for Stable Diffusion workflows, especially via the stable-diffusion-webui embeddings directory. It is designed to be used with certain image generation models, though cross-model effectiveness may vary. Its licensing is marked as other. | huggingface | gsdf/EasyNegative |
Summarize the dataset in a concise way for a machine learning practitioner. | 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, modality:text
Downloads: 2169
Likes: 1147 | RedPajama-Data-1T is a large English text dataset intended for text generation and language model training. It was created as a clean-room, fully open-source reproduction of the data used for LLaMA-style model development. It is primarily useful for large-scale pretraining and open LLM research. | huggingface | togethercomputer/RedPajama-Data-1T |
Summarize the dataset in a concise way for a machine learning practitioner. | Name: medical-o1-reasoning-SFT
Description: Medical reasoning dataset for supervised fine-tuning built on medical verifiable problems and distilled reasoning data. The dataset includes medical SFT data and references releases focused on reasoning chains derived from stronger models.
Tags: task_categories:question-answe... | medical-o1-reasoning-SFT is a bilingual English and Chinese medical reasoning dataset for supervised fine-tuning. It focuses on medical verifiable problems and reasoning-oriented QA or generation tasks, making it suitable for training models on structured medical reasoning. The dataset is Apache-2.0 licensed and releva... | huggingface | FreedomIntelligence/medical-o1-reasoning-SFT |
Summarize the dataset for LLM pretraining use, including its scale, domain, language, license, format, and notable metadata. | {"source":"huggingface","id":"HuggingFaceFW/fineweb-edu","name":"fineweb-edu","description":"📚 FineWeb-Edu\n1.3 trillion tokens of the finest educational data the web has to offer. FineWeb-Edu dataset consists of 1.3T tokens and 5.4T tokens (FineWeb-Edu-score-2) of educational web pages filtered from FineWeb dataset. ... | fineweb-edu is a large English educational web-text dataset from Hugging Face, intended for text generation and LLM pretraining. It contains about 1.3 trillion tokens of educational web pages filtered from FineWeb, with quality improved using a classifier trained from Llama3-70B-Instruct-generated annotations. The data... | huggingface | HuggingFaceFW/fineweb-edu |
Provide a concise dataset description suitable for supervised fine-tuning metadata, covering purpose, scale, language, license, and popularity signals. | {"source":"huggingface","id":"allenai/dolma","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-m... | dolma is an open English corpus designed for language model pretraining research. It contains roughly three trillion tokens and is positioned as a large-scale resource for language modeling and LLM development. The dataset is licensed under ODC-BY and is associated with research including arXiv:2402.00159. Based on its... | huggingface | allenai/dolma |
Summarize the dataset with emphasis on content type, scale, licensing characteristics, multilingual/code aspects, and intended use in model training. | {"source":"huggingface","id":"bigcode/the-stack","name":"the-stack","description":"Dataset Card for The Stack. Initial release included 30 programming languages and 18 permissive licenses, resulting in a near-deduplicated dataset of 3TB. Later versions excluded weak copyleft licenses (MPL/EPL/LGPL) and extended the lis... | the-stack is a large multilingual code dataset created for text generation and code model training. It consists of source code collected across many programming languages, with a near-deduplicated release reported at roughly 3TB. The dataset emphasizes permissive licensing, with later versions removing weak copyleft li... | huggingface | bigcode/the-stack |
Create a short structured summary of the dataset, mentioning data generation method, task relevance, scale, language, license, and common training usage. | {"source":"huggingface","id":"roneneldan/TinyStories","name":"TinyStories","description":"Dataset containing synthetically generated (by GPT-3.5 and GPT-4) short stories that only use a small vocabulary. Described in arXiv:2305.07759. The models referred to in the paper were trained on TinyStories-train.txt, with tinys... | TinyStories is an English text-generation dataset of synthetically generated short stories created by GPT-3.5 and GPT-4. The stories are constrained to a small vocabulary, making the dataset especially useful for studying small language models and language acquisition-like settings. It is relatively small compared with... | huggingface | roneneldan/TinyStories |
Summarize this instruction-tuning dataset, including task types, data origin, size, license, format, and how it can be used for SFT. | {"source":"huggingface","id":"databricks/databricks-dolly-15k","name":"databricks-dolly-15k","description":"databricks-dolly-15k is an open source dataset of instruction-following records generated by thousands of Databricks employees in several behavioral categories outlined in the InstructGPT paper, including brainst... | databricks-dolly-15k is an English instruction-following dataset designed for supervised fine-tuning. It contains about 15,000 human-generated records created by Databricks employees across tasks such as brainstorming, classification, closed and open question answering, generation, information extraction, and summariza... | huggingface | databricks/databricks-dolly-15k |
Summarize the dataset metadata into a concise SFT training example. | {"source":"huggingface","id":"tatsu-lab/alpaca","name":"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 m... | {"name":"alpaca","summary":"A popular English instruction-tuning dataset with 52K instruction-response examples generated using OpenAI text-davinci-003 and inspired by the Self-Instruct pipeline.","modality":"text","language":"en","license":"cc-by-nc-4.0","size":"10K<n<100K","primary_use":"instruction-finetuning","popu... | huggingface | tatsu-lab/alpaca |
Summarize the dataset metadata into a concise SFT training example. | {"source":"huggingface","id":"Nerfgun3/bad_prompt","name":"bad_prompt","description":"Negative Embedding / Textual Inversion. The 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.","tags":["language:en","licens... | {"name":"bad_prompt","summary":"A small image-related Hugging Face dataset/resource associated with negative embedding or textual inversion for Stable Diffusion workflows, intended to improve generation quality via a reusable negative prompt embedding.","modality":"image","language":"en","license":"creativeml-openrail-... | huggingface | Nerfgun3/bad_prompt |
Summarize the dataset metadata into a concise SFT training example. | {"source":"huggingface","id":"tiiuae/falcon-refinedweb","name":"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 Refi... | {"name":"falcon-refinedweb","summary":"A massive English web-scale text dataset derived from CommonCrawl, heavily filtered and deduplicated for large language model pretraining, with multimodal-friendly web content references.","modality":"text","language":"en","license":"odc-by","size":"100M<n<1B","primary_use":"LLM p... | huggingface | tiiuae/falcon-refinedweb |
Summarize the dataset metadata into a concise SFT training example. | {"source":"huggingface","id":"lmsys/lmsys-chat-1m","name":"lmsys-chat-1m","description":"LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset. This dataset contains one million real-world conversations with 25 state-of-the-art LLMs, collected from 210K unique IP addresses on the Vicuna demo and Chatbot Aren... | {"name":"lmsys-chat-1m","summary":"A large-scale real-world conversation dataset containing around one million chats with 25 LLMs, sourced from public demo and arena platforms and useful for conversational AI analysis and training.","modality":"text","language":"unknown/varied","license":"not specified","size":"1M<n<10... | huggingface | lmsys/lmsys-chat-1m |
Summarize the dataset metadata into a concise SFT training example. | {"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 includes confirmed working method for training the model and multiple cleaned dataset variants... | {"name":"ShareGPT_Vicuna_unfiltered","summary":"An English conversational dataset derived from ShareGPT/Vicuna-style data with additional cleaning and variants intended for chat model training and experimentation.","modality":"text","language":"en","license":"apache-2.0","size":"not specified","primary_use":"chat fine-... | huggingface | anon8231489123/ShareGPT_Vicuna_unfiltered |
Summarize the dataset for supervised fine-tuning metadata extraction. | 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_categories:text-generation, license:odc-by, size_categories:100M<n<1B, ... | {"name":"finepdfs","provider":"Hugging Face","summary":"A massive multilingual text corpus extracted exclusively from PDFs, designed for large-scale language modeling and text generation research.","license":"odc-by","task_category":"text-generation","languages":"1733 languages","formats":["parquet"],"modalities":["tab... | huggingface | HuggingFaceFW/finepdfs |
Summarize the dataset for supervised fine-tuning metadata extraction. | 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 end-to-end driving systems. It contains a total of 1700 hours of driving and is available for commercial/no... | {"name":"PhysicalAI-Autonomous-Vehicles","provider":"Hugging Face","summary":"A large-scale, geographically diverse autonomous vehicle dataset with multi-sensor driving data for end-to-end driving research and deployment.","license":"other","data_collection":"automatic/sensor","labeling_method":"automatic/sensor","dura... | huggingface | nvidia/PhysicalAI-Autonomous-Vehicles |
Summarize the dataset for supervised fine-tuning metadata extraction. | 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 OpenThinker models.
Tags: license:apache-2.0, size_categories:100K<n<1M, format:parquet, modality:text, region... | {"name":"OpenThoughts-114k","provider":"Hugging Face","summary":"A synthetic reasoning dataset with 114k training examples spanning math, science, coding, and puzzle tasks for model finetuning.","license":"apache-2.0","domains":["math","science","code","puzzles"],"synthetic":true,"formats":["parquet"],"modalities":["te... | huggingface | open-thoughts/OpenThoughts-114k |
Summarize the dataset for supervised fine-tuning metadata extraction. | Name: alpaca-cleaned
Description: This is a cleaned version of the original Alpaca Dataset released by Stanford. It fixes issues in the original release such as hallucination-prone instructions referencing internet data.
Tags: task_categories:text-generation, language:en, license:cc-by-4.0, size_categories:10K<n<100K, ... | {"name":"alpaca-cleaned","provider":"Hugging Face","summary":"A cleaned English instruction-tuning dataset based on Stanford Alpaca, with problematic and hallucination-inducing examples corrected or removed.","license":"cc-by-4.0","task_category":"text-generation","language":"en","use_case":"instruction-finetuning","fo... | huggingface | yahma/alpaca-cleaned |
Summarize the dataset for supervised fine-tuning metadata extraction. | Name: OpenHermes-2.5
Description: This dataset underpins OpenHermes 2.5 and Nous Hermes 2 models. It is a compilation and curation of many open source datasets and custom synthetic datasets.
Tags: language:eng, size_categories:1M<n<10M, format:json, modality:text, region:us, synthetic, GPT-4, Distillation, Compilation
... | {"name":"OpenHermes-2.5","provider":"Hugging Face","summary":"A large compiled instruction-tuning corpus used for OpenHermes and Nous Hermes models, combining open-source and synthetic data with GPT-4 distillation.","language":"eng","synthetic":true,"techniques":["GPT-4 distillation","dataset compilation"],"formats":["... | huggingface | teknium/OpenHermes-2.5 |
Summarize the dataset fineweb-2 for SFT metadata ingestion. | Source: huggingface
ID: HuggingFaceFW/fineweb-2
Name: fineweb-2
Description: FineWeb2 is the second iteration of the FineWeb dataset, providing high-quality reproducible pretraining data across 1000+ languages under the ODC-By 1.0 license, validated through extensive ablation experiments.
Tags: task_categories:text-gen... | {"name":"fineweb-2","summary":"A large-scale multilingual pretraining dataset covering over 1000 languages, designed for high-quality and reproducible text generation research.","task_categories":["text-generation"],"languages":"1000+ languages","license":"odc-by","modalities":["tabular","text"],"size_category":"1B<n<1... | huggingface | HuggingFaceFW/fineweb-2 |
Summarize the dataset hle for SFT metadata ingestion. | Source: huggingface
ID: cais/hle
Name: hle
Description: Humanity's Last Exam (HLE) is a multimodal benchmark at the frontier of human knowledge, designed as a closed-ended academic benchmark with broad subject coverage and 2,500 questions.
Tags: benchmark:official; license:mit; size_categories:1K<n<10K; format:parquet;... | {"name":"hle","summary":"Humanity's Last Exam is an official multimodal benchmark with 2,500 closed-ended questions spanning broad academic subjects at the frontier of human knowledge.","benchmark":true,"modalities":["image","text"],"license":"mit","format":"parquet","size_category":"1K<n<10K","downloads":47598,"likes"... | huggingface | cais/hle |
Summarize the dataset imagenet-1k for SFT metadata ingestion. | Source: huggingface
ID: ILSVRC/imagenet-1k
Name: imagenet-1k
Description: ImageNet (ILSVRC 2012) is a large image dataset organized according to the WordNet hierarchy, with roughly 1000 images per synset and widely used for image classification.
Tags: task_categories:image-classification; task_ids:multi-class-image-cla... | {"name":"imagenet-1k","summary":"A canonical large-scale image classification dataset from ILSVRC 2012, organized by WordNet synsets and widely used for multi-class image classification.","task_categories":["image-classification"],"task_id":"multi-class-image-classification","language":"en","license":"other","modalitie... | huggingface | ILSVRC/imagenet-1k |
Summarize the dataset Alpaca-CoT for SFT metadata ingestion. | Source: huggingface
ID: QingyiSi/Alpaca-CoT
Name: Alpaca-CoT
Description: Alpaca-CoT is a continuously collected instruction-finetuning dataset collection that standardizes multiple instruction tuning datasets into a unified format compatible with Alpaca-style model training.
Tags: language:en; language:zh; language:ml... | {"name":"Alpaca-CoT","summary":"A standardized collection of instruction-tuning datasets for Alpaca-style finetuning, including chain-of-thought oriented data in multiple languages.","use_case":"instruction-finetuning","languages":["en","zh","ml"],"license":"apache-2.0","keywords":["Instruction","Cot"],"downloads":7554... | huggingface | QingyiSi/Alpaca-CoT |
Summarize the dataset PersonaHub for SFT metadata ingestion. | Source: huggingface
ID: proj-persona/PersonaHub
Name: PersonaHub
Description: PersonaHub releases data from the paper Scaling Synthetic Data Creation with 1,000,000,000 Personas, introducing a persona-driven synthetic data methodology and a collection of 1 billion diverse personas curated from web data.
Tags: task_cate... | {"name":"PersonaHub","summary":"A synthetic text dataset resource built around persona-driven data generation, releasing large-scale diverse personas for creating instruction, reasoning, math, and tool-use data.","task_categories":["text-generation","text-classification","token-classification","fill-mask","table-questi... | huggingface | proj-persona/PersonaHub |
SoloAI SFT Dataset: 20260422-1657
📊 数据集概览
| 属性 | 值 |
|---|---|
| 数据条数 | 30 条 |
| 数据源 | huggingface |
| 生成时间 | 2026-04-22 16:57 |
| 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 LLM practitioners, including its purpose, likely use cases, scale, licensing, and notable characteristics.', 'input': 'Name: prompts.chat\nDescription: 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.\nTags: task_categories:question-answering, task_categories:text-generation, license:cc0-1.0, size_categories:1K<n<10K, format:csv, modality:text, region:us, ChatGPT, prompts, AI, GPT, Claude, Gemini, Llama, Mistral, LLM, prompt-engineering, conversational-ai, text-generation, chatbot, awesome-list\nDownloads: 38777\nLikes: 9679', 'output': 'prompts.chat is a community prompt dataset mirrored from the prompts.chat platform, also known as Awesome ChatGPT Prompts. It focuses on user-created prompts for conversational AI and text generation systems such as ChatGPT, Claude, Gemini, Llama, and Mistral. The dataset is relatively small, in the 1K to 10K range, stored in CSV format, and licensed under CC0-1.0, which makes it highly permissive for reuse. Its main value is for prompt engineering research, prompt discovery, chatbot prototyping, and building prompt libraries or recommendation systems. Because it is a mirror of a live social platform, users may want to consult the original site or GitHub repository for the latest community contributions.', '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:
307809343@qq.com - 🤖 SoloAI 提供高质量 SFT 数据定制服务。
- 📧 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 | 2026-04-22 16:57 | 初始发布，30 条数据 |
- Downloads last month
- 12