Create special_tokens_map.json

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "pad_token": "[PAD]",    // Used to make all sequences the same length by padding
+  "unk_token": "[UNK]",    // Used when the tokenizer encounters words it doesn't know
+  "cls_token": "[CLS]",    // Often used at the start of sequences (Classification token)
+  "sep_token": "[SEP]"     // Used to separate different parts of the input (Separator token)
+}