chunk_config.json

{
    "input_dataset": "sergeipetrov/transformers-diffusers-docs-raw",
	"input_splits": ["train"],
	"input_text_col": "text",
	"output_dataset": "sergeipetrov/transformers-diffusers-docs-chunked",
	"strategy": "spacy",
	"split_seq": "\n\n",
	"chunk_len": 512,
	"private": "false"
}