AnmolNimmala0/

agri-slm-corpus-3

Pre-training corpus for a 300M parameter India-focused agriculture SLM.

	Contents




	

File Description Docs Tokens

corpus_india_downsampled.jsonl India-focused downsampled base corpus (PubMed, EuropePMC, Wikipedia, extension, ICAR, etc.) ~201K ~1,244M

krishikosh.jsonl ICAR KrishiKosh theses, articles, books, reports (DSpace 7 harvest) ~55K ~1,385M

{ "text": "...", "source": "krishikosh | ..."… See the full description on the dataset page: https://huggingface.co/datasets/AnmolNimmala0/agri-slm-corpus-3.