up

config.json

@@ -13,14 +13,14 @@
   "intermediate_size": 16384,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 514,
-  "model_type": "xlm-roberta",
+  "model_type": "xlm-roberta-xl",
   "num_attention_heads": 32,
   "num_hidden_layers": 48,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
-  "transformers_version": "4.11.0.dev0",
   "type_vocab_size": 1,
   "use_cache": true,
-  "vocab_size": 250880
+  "vocab_size": 250880,
+	"tokenizer_class": "XLMRobertaTokenizer"
 }

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}

tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sp_model_kwargs": {}, "special_tokens_map_file": "./special_tokens_map.json", "name_or_path": "../xlm-roberta-xl/", "tokenizer_file": "../xlm-roberta-xl/tokenizer.json", "tokenizer_class": "XLMRobertaTokenizer"}