A collection of datasets made to train general-purpose SLMs under 200M parameters
MoreThought
MoreThought
AI & ML interests
More. Thought.
More. Thought.
Recent Activity
liked a model about 2 hours ago
Muse-research/Muse2-230M-Base liked a model about 2 hours ago
Muse-research/Muse2-230M updated a Space about 4 hours ago
ArrowLabs/Smol-Intelligence-IndexOrganizations
SLMFineTrain
A collection of datasets made to train general-purpose SLMs under 200M parameters
DeepSWE-Gyms2
All the DeepSWE-Gym2 variants.
DeepSWE-Gyms
All of the DeepSWE-Gym variants.
OpenFineTrain
Very fine 55 trillion+ tokens worth of ungated huggingface datasets for LLM training across 80+ datasets.