gbert-base-mlx / tokenizer_config.json

deepset/gbert-base prepared for oMLX (float16, mean pooling)

013877f verified 21 days ago

263 Bytes

{
  "do_lower_case": false,
  "init_inputs": [],
  "strip_accents": false,
  "tokenizer_class": "BertTokenizer",
  "model_max_length": 512,
  "cls_token": "[CLS]",
  "sep_token": "[SEP]",
  "pad_token": "[PAD]",
  "unk_token": "[UNK]",
  "mask_token": "[MASK]"
}

	{
	"do_lower_case": false,
	"init_inputs": [],
	"strip_accents": false,
	"tokenizer_class": "BertTokenizer",
	"model_max_length": 512,
	"cls_token": "[CLS]",
	"sep_token": "[SEP]",
	"pad_token": "[PAD]",
	"unk_token": "[UNK]",
	"mask_token": "[MASK]"
	}