HuggingFaceTB/cosmopedia-100k

Dataset

stars

commits

linked in READMEs

Feb 19, 2024

updated

synthetic

README

Dataset description

This is a 100k subset of Cosmopedia dataset. A synthetic dataset of textbooks, blogposts, stories, posts and WikiHow articles generated by Mixtral-8x7B-Instruct-v0.1.

Here's how you can load the dataset

from datasets import load_dataset

ds = load_dataset("HuggingFaceTB/cosmopedia-100k", split="train")

Contributors

lvwerra

2 commits

loubnabnl

1 commits

Similar repos

HuggingFaceTB/cosmopedia

Dataset

743

HuggingFaceTB/smollm-corpus

Dataset

483

lapp0/hotpot_query_expansion_synthetic

Dataset

HuggingFaceH4/spin-ultrachat-prompts-qwen-1.5-0.5b-iter0-iter1

Dataset

HuggingFaceH4/Magpie-Pro-DPO-100K-v0.1-Prompts

Dataset

lapp0/hotpot_query_expansion_synthetic_cleaned

Dataset

obalcells/longfact-augmented-prompts

Dataset

HuggingFaceH4/h4-anthropic-hh-rlhf-helpful-base-gen

Dataset

Linked in READMEs

pranavktrpl/smolMoELM-custom

Upcycled smolLM to convert a Dense LM to an MoE. CPT'd the model to train the router. Two sub…

README badge

README

Dataset description

This is a 100k subset of Cosmopedia dataset. A synthetic dataset of textbooks, blogposts, stories, posts and WikiHow articles generated by Mixtral-8x7B-Instruct-v0.1.

Here's how you can load the dataset

from datasets import load_dataset

ds = load_dataset("HuggingFaceTB/cosmopedia-100k", split="train")