Download tokenizer.json from subrace/NepBERTa-pytorch: direct link, hf CLI and curl.
- Browser
- Download file 1.03 MB
-
https://huggingface.co/subrace/NepBERTa-pytorch/resolve/main/tokenizer.json
- Command line
-
hf download hf://subrace/NepBERTa-pytorch/tokenizer.json
-
curl -L -o tokenizer.json https://huggingface.co/subrace/NepBERTa-pytorch/resolve/main/tokenizer.json
1.03 MB
File too large to display, you can check the raw version instead.