pmrccs/math-sft-mix-1024tok-v1 download history

pmrccs/math-sft-mix-1024tok-v1 is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 107 times (15 in the last 7 days), and 107 times in total. It ranks #143,410 among datasets by monthly downloads.

math-sft-mix-1024tok-v1 A mixed, English-only math SFT dataset of 142,316 problem/solution pairs, every one at most 1024 tokens (measured with the Qwen/Qwen2.5-1.5B tokenizer plus a 32-token template allowance). It is built for LoRA/QLoRA supervised fine-tuning of small (1B–3B) base m

Models trained on math-sft-mix-1024tok-v1

2 models list it as training data.

Open pmrccs/math-sft-mix-1024tok-v1 on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.