Training in progress, epoch 0

added_tokens.json

@@ -1,3 +1,4 @@
 {
-  "<s_500k>": 35054
+  "<s_500k>": 35054,
+  "<s_donut_rus>": 35055
 }

tokenizer.json

@@ -56,6 +56,15 @@
       "rstrip": false,
       "normalized": true,
       "special": false
+    },
+    {
+      "id": 35055,
+      "content": "<s_donut_rus>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": true,
+      "special": false
     }
   ],
   "normalizer": {

tokenizer_config.json

@@ -1,5 +1,6 @@
 {
   "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
   "cls_token": "<s>",
   "eos_token": "</s>",
   "mask_token": {
@@ -15,7 +16,6 @@
   "processor_class": "DonutProcessor",
   "sep_token": "</s>",
   "sp_model_kwargs": {},
-  "special_tokens_map_file": "/root/.cache/huggingface/transformers/6ea33bfad55f516eeb1e52bb6e3f5c198c976c6ebfaa7e1fb9f56b33b2ddc8a9.f6ac38e94db1fb749ad0f52af2f7cd5b7863b9b30f1aa680a1b0d22ec09b717d",
   "tokenizer_class": "XLMRobertaTokenizer",
   "unk_token": "<unk>"
 }