Upload processor

special_tokens_map.json

@@ -1,6 +1,30 @@
 {
-  "bos_token": "<s>",
-  "eos_token": "</s>",
-  "pad_token": "[PAD]",
-  "unk_token": "[UNK]"
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": true,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": true,
+    "single_word": false
+  }
 }

vocab.json

@@ -1 +1,48 @@
-{"g": 0, "k": 1, "\u00ec": 2, "w": 3, "d": 4, "p": 5, "\u00f3": 6, "o": 7, "n": 8, "\u0127": 9, "x": 10, "\u0121": 11, "\u00f9": 12, "b": 13, "f": 14, "\u00e9": 15, "j": 16, "4": 17, "a": 18, "m": 19, "e": 20, "\u0107": 21, "\u017c": 22, "q": 23, "s": 24, "y": 25, "\u010b": 26, "\u00ed": 27, "-": 28, "t": 29, "h": 30, "r": 31, "i": 32, "\u00e1": 33, "u": 34, "v": 35, "c": 37, "\u00f2": 38, "`": 39, "l": 40, "\u00e0": 41, "\u00e8": 42, "z": 43, "|": 36, "[UNK]": 44, "[PAD]": 45}
+{
+  "-": 28,
+  "4": 17,
+  "[PAD]": 45,
+  "[UNK]": 44,
+  "`": 39,
+  "a": 18,
+  "b": 13,
+  "c": 37,
+  "d": 4,
+  "e": 20,
+  "f": 14,
+  "g": 0,
+  "h": 30,
+  "i": 32,
+  "j": 16,
+  "k": 1,
+  "l": 40,
+  "m": 19,
+  "n": 8,
+  "o": 7,
+  "p": 5,
+  "q": 23,
+  "r": 31,
+  "s": 24,
+  "t": 29,
+  "u": 34,
+  "v": 35,
+  "w": 3,
+  "x": 10,
+  "y": 25,
+  "z": 43,
+  "|": 36,
+  "à": 41,
+  "á": 33,
+  "è": 42,
+  "é": 15,
+  "ì": 2,
+  "í": 27,
+  "ò": 38,
+  "ó": 6,
+  "ù": 12,
+  "ć": 21,
+  "ċ": 26,
+  "ġ": 11,
+  "ħ": 9,
+  "ż": 22
+}