Training in progress, step 500
Browse files- added_tokens.json +33 -0
- config.json +2 -2
- model.safetensors +2 -2
- tokenizer.json +279 -0
- tokenizer_config.json +248 -0
- training_args.bin +1 -1
added_tokens.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<": 32103,
|
3 |
+
"{": 32120,
|
4 |
+
"±": 32104,
|
5 |
+
"À": 32123,
|
6 |
+
"Â": 32122,
|
7 |
+
"Ä": 32129,
|
8 |
+
"È": 32124,
|
9 |
+
"Ë": 32106,
|
10 |
+
"Ó": 32116,
|
11 |
+
"Ô": 32119,
|
12 |
+
"Ö": 32128,
|
13 |
+
"×": 32108,
|
14 |
+
"Ø": 32109,
|
15 |
+
"ë": 32101,
|
16 |
+
"í": 32107,
|
17 |
+
"ï": 32125,
|
18 |
+
"ñ": 32130,
|
19 |
+
"ú": 32126,
|
20 |
+
"úč": 32117,
|
21 |
+
"ý": 32118,
|
22 |
+
"ć": 32112,
|
23 |
+
"č": 32102,
|
24 |
+
"ę": 32114,
|
25 |
+
"İ": 32100,
|
26 |
+
"œ": 32121,
|
27 |
+
"ś": 32115,
|
28 |
+
"Š": 32127,
|
29 |
+
"š": 32113,
|
30 |
+
"Ω": 32111,
|
31 |
+
"μ": 32105,
|
32 |
+
"‐": 32110
|
33 |
+
}
|
config.json
CHANGED
@@ -55,7 +55,7 @@
|
|
55 |
}
|
56 |
},
|
57 |
"torch_dtype": "float32",
|
58 |
-
"transformers_version": "4.45.
|
59 |
"use_cache": true,
|
60 |
-
"vocab_size":
|
61 |
}
|
|
|
55 |
}
|
56 |
},
|
57 |
"torch_dtype": "float32",
|
58 |
+
"transformers_version": "4.45.2",
|
59 |
"use_cache": true,
|
60 |
+
"vocab_size": 32131
|
61 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2aabb6fdc61c1feedc034e435ca3b09dca7f738850e4847e03c2b5cfee086f95
|
3 |
+
size 242048040
|
tokenizer.json
CHANGED
@@ -934,6 +934,285 @@
|
|
934 |
"rstrip": false,
|
935 |
"normalized": false,
|
936 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
937 |
}
|
938 |
],
|
939 |
"normalizer": {
|
|
|
934 |
"rstrip": false,
|
935 |
"normalized": false,
|
936 |
"special": true
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"id": 32100,
|
940 |
+
"content": "İ",
|
941 |
+
"single_word": false,
|
942 |
+
"lstrip": false,
|
943 |
+
"rstrip": false,
|
944 |
+
"normalized": true,
|
945 |
+
"special": false
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"id": 32101,
|
949 |
+
"content": "ë",
|
950 |
+
"single_word": false,
|
951 |
+
"lstrip": false,
|
952 |
+
"rstrip": false,
|
953 |
+
"normalized": true,
|
954 |
+
"special": false
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"id": 32102,
|
958 |
+
"content": "č",
|
959 |
+
"single_word": false,
|
960 |
+
"lstrip": false,
|
961 |
+
"rstrip": false,
|
962 |
+
"normalized": true,
|
963 |
+
"special": false
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"id": 32103,
|
967 |
+
"content": "<",
|
968 |
+
"single_word": false,
|
969 |
+
"lstrip": false,
|
970 |
+
"rstrip": false,
|
971 |
+
"normalized": true,
|
972 |
+
"special": false
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"id": 32104,
|
976 |
+
"content": "±",
|
977 |
+
"single_word": false,
|
978 |
+
"lstrip": false,
|
979 |
+
"rstrip": false,
|
980 |
+
"normalized": true,
|
981 |
+
"special": false
|
982 |
+
},
|
983 |
+
{
|
984 |
+
"id": 32105,
|
985 |
+
"content": "μ",
|
986 |
+
"single_word": false,
|
987 |
+
"lstrip": false,
|
988 |
+
"rstrip": false,
|
989 |
+
"normalized": true,
|
990 |
+
"special": false
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"id": 32106,
|
994 |
+
"content": "Ë",
|
995 |
+
"single_word": false,
|
996 |
+
"lstrip": false,
|
997 |
+
"rstrip": false,
|
998 |
+
"normalized": true,
|
999 |
+
"special": false
|
1000 |
+
},
|
1001 |
+
{
|
1002 |
+
"id": 32107,
|
1003 |
+
"content": "í",
|
1004 |
+
"single_word": false,
|
1005 |
+
"lstrip": false,
|
1006 |
+
"rstrip": false,
|
1007 |
+
"normalized": true,
|
1008 |
+
"special": false
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"id": 32108,
|
1012 |
+
"content": "×",
|
1013 |
+
"single_word": false,
|
1014 |
+
"lstrip": false,
|
1015 |
+
"rstrip": false,
|
1016 |
+
"normalized": true,
|
1017 |
+
"special": false
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"id": 32109,
|
1021 |
+
"content": "Ø",
|
1022 |
+
"single_word": false,
|
1023 |
+
"lstrip": false,
|
1024 |
+
"rstrip": false,
|
1025 |
+
"normalized": true,
|
1026 |
+
"special": false
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"id": 32110,
|
1030 |
+
"content": "‐",
|
1031 |
+
"single_word": false,
|
1032 |
+
"lstrip": false,
|
1033 |
+
"rstrip": false,
|
1034 |
+
"normalized": true,
|
1035 |
+
"special": false
|
1036 |
+
},
|
1037 |
+
{
|
1038 |
+
"id": 32111,
|
1039 |
+
"content": "Ω",
|
1040 |
+
"single_word": false,
|
1041 |
+
"lstrip": false,
|
1042 |
+
"rstrip": false,
|
1043 |
+
"normalized": true,
|
1044 |
+
"special": false
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"id": 32112,
|
1048 |
+
"content": "ć",
|
1049 |
+
"single_word": false,
|
1050 |
+
"lstrip": false,
|
1051 |
+
"rstrip": false,
|
1052 |
+
"normalized": true,
|
1053 |
+
"special": false
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"id": 32113,
|
1057 |
+
"content": "š",
|
1058 |
+
"single_word": false,
|
1059 |
+
"lstrip": false,
|
1060 |
+
"rstrip": false,
|
1061 |
+
"normalized": true,
|
1062 |
+
"special": false
|
1063 |
+
},
|
1064 |
+
{
|
1065 |
+
"id": 32114,
|
1066 |
+
"content": "ę",
|
1067 |
+
"single_word": false,
|
1068 |
+
"lstrip": false,
|
1069 |
+
"rstrip": false,
|
1070 |
+
"normalized": true,
|
1071 |
+
"special": false
|
1072 |
+
},
|
1073 |
+
{
|
1074 |
+
"id": 32115,
|
1075 |
+
"content": "ś",
|
1076 |
+
"single_word": false,
|
1077 |
+
"lstrip": false,
|
1078 |
+
"rstrip": false,
|
1079 |
+
"normalized": true,
|
1080 |
+
"special": false
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"id": 32116,
|
1084 |
+
"content": "Ó",
|
1085 |
+
"single_word": false,
|
1086 |
+
"lstrip": false,
|
1087 |
+
"rstrip": false,
|
1088 |
+
"normalized": true,
|
1089 |
+
"special": false
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"id": 32117,
|
1093 |
+
"content": "úč",
|
1094 |
+
"single_word": false,
|
1095 |
+
"lstrip": false,
|
1096 |
+
"rstrip": false,
|
1097 |
+
"normalized": true,
|
1098 |
+
"special": false
|
1099 |
+
},
|
1100 |
+
{
|
1101 |
+
"id": 32118,
|
1102 |
+
"content": "ý",
|
1103 |
+
"single_word": false,
|
1104 |
+
"lstrip": false,
|
1105 |
+
"rstrip": false,
|
1106 |
+
"normalized": true,
|
1107 |
+
"special": false
|
1108 |
+
},
|
1109 |
+
{
|
1110 |
+
"id": 32119,
|
1111 |
+
"content": "Ô",
|
1112 |
+
"single_word": false,
|
1113 |
+
"lstrip": false,
|
1114 |
+
"rstrip": false,
|
1115 |
+
"normalized": true,
|
1116 |
+
"special": false
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"id": 32120,
|
1120 |
+
"content": "{",
|
1121 |
+
"single_word": false,
|
1122 |
+
"lstrip": false,
|
1123 |
+
"rstrip": false,
|
1124 |
+
"normalized": true,
|
1125 |
+
"special": false
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"id": 32121,
|
1129 |
+
"content": "œ",
|
1130 |
+
"single_word": false,
|
1131 |
+
"lstrip": false,
|
1132 |
+
"rstrip": false,
|
1133 |
+
"normalized": true,
|
1134 |
+
"special": false
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"id": 32122,
|
1138 |
+
"content": "Â",
|
1139 |
+
"single_word": false,
|
1140 |
+
"lstrip": false,
|
1141 |
+
"rstrip": false,
|
1142 |
+
"normalized": true,
|
1143 |
+
"special": false
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"id": 32123,
|
1147 |
+
"content": "À",
|
1148 |
+
"single_word": false,
|
1149 |
+
"lstrip": false,
|
1150 |
+
"rstrip": false,
|
1151 |
+
"normalized": true,
|
1152 |
+
"special": false
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"id": 32124,
|
1156 |
+
"content": "È",
|
1157 |
+
"single_word": false,
|
1158 |
+
"lstrip": false,
|
1159 |
+
"rstrip": false,
|
1160 |
+
"normalized": true,
|
1161 |
+
"special": false
|
1162 |
+
},
|
1163 |
+
{
|
1164 |
+
"id": 32125,
|
1165 |
+
"content": "ï",
|
1166 |
+
"single_word": false,
|
1167 |
+
"lstrip": false,
|
1168 |
+
"rstrip": false,
|
1169 |
+
"normalized": true,
|
1170 |
+
"special": false
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"id": 32126,
|
1174 |
+
"content": "ú",
|
1175 |
+
"single_word": false,
|
1176 |
+
"lstrip": false,
|
1177 |
+
"rstrip": false,
|
1178 |
+
"normalized": true,
|
1179 |
+
"special": false
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"id": 32127,
|
1183 |
+
"content": "Š",
|
1184 |
+
"single_word": false,
|
1185 |
+
"lstrip": false,
|
1186 |
+
"rstrip": false,
|
1187 |
+
"normalized": true,
|
1188 |
+
"special": false
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"id": 32128,
|
1192 |
+
"content": "Ö",
|
1193 |
+
"single_word": false,
|
1194 |
+
"lstrip": false,
|
1195 |
+
"rstrip": false,
|
1196 |
+
"normalized": true,
|
1197 |
+
"special": false
|
1198 |
+
},
|
1199 |
+
{
|
1200 |
+
"id": 32129,
|
1201 |
+
"content": "Ä",
|
1202 |
+
"single_word": false,
|
1203 |
+
"lstrip": false,
|
1204 |
+
"rstrip": false,
|
1205 |
+
"normalized": true,
|
1206 |
+
"special": false
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"id": 32130,
|
1210 |
+
"content": "ñ",
|
1211 |
+
"single_word": false,
|
1212 |
+
"lstrip": false,
|
1213 |
+
"rstrip": false,
|
1214 |
+
"normalized": true,
|
1215 |
+
"special": false
|
1216 |
}
|
1217 |
],
|
1218 |
"normalizer": {
|
tokenizer_config.json
CHANGED
@@ -823,6 +823,254 @@
|
|
823 |
"rstrip": false,
|
824 |
"single_word": false,
|
825 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
826 |
}
|
827 |
},
|
828 |
"additional_special_tokens": [
|
|
|
823 |
"rstrip": false,
|
824 |
"single_word": false,
|
825 |
"special": true
|
826 |
+
},
|
827 |
+
"32100": {
|
828 |
+
"content": "İ",
|
829 |
+
"lstrip": false,
|
830 |
+
"normalized": true,
|
831 |
+
"rstrip": false,
|
832 |
+
"single_word": false,
|
833 |
+
"special": false
|
834 |
+
},
|
835 |
+
"32101": {
|
836 |
+
"content": "ë",
|
837 |
+
"lstrip": false,
|
838 |
+
"normalized": true,
|
839 |
+
"rstrip": false,
|
840 |
+
"single_word": false,
|
841 |
+
"special": false
|
842 |
+
},
|
843 |
+
"32102": {
|
844 |
+
"content": "č",
|
845 |
+
"lstrip": false,
|
846 |
+
"normalized": true,
|
847 |
+
"rstrip": false,
|
848 |
+
"single_word": false,
|
849 |
+
"special": false
|
850 |
+
},
|
851 |
+
"32103": {
|
852 |
+
"content": "<",
|
853 |
+
"lstrip": false,
|
854 |
+
"normalized": true,
|
855 |
+
"rstrip": false,
|
856 |
+
"single_word": false,
|
857 |
+
"special": false
|
858 |
+
},
|
859 |
+
"32104": {
|
860 |
+
"content": "±",
|
861 |
+
"lstrip": false,
|
862 |
+
"normalized": true,
|
863 |
+
"rstrip": false,
|
864 |
+
"single_word": false,
|
865 |
+
"special": false
|
866 |
+
},
|
867 |
+
"32105": {
|
868 |
+
"content": "μ",
|
869 |
+
"lstrip": false,
|
870 |
+
"normalized": true,
|
871 |
+
"rstrip": false,
|
872 |
+
"single_word": false,
|
873 |
+
"special": false
|
874 |
+
},
|
875 |
+
"32106": {
|
876 |
+
"content": "Ë",
|
877 |
+
"lstrip": false,
|
878 |
+
"normalized": true,
|
879 |
+
"rstrip": false,
|
880 |
+
"single_word": false,
|
881 |
+
"special": false
|
882 |
+
},
|
883 |
+
"32107": {
|
884 |
+
"content": "í",
|
885 |
+
"lstrip": false,
|
886 |
+
"normalized": true,
|
887 |
+
"rstrip": false,
|
888 |
+
"single_word": false,
|
889 |
+
"special": false
|
890 |
+
},
|
891 |
+
"32108": {
|
892 |
+
"content": "×",
|
893 |
+
"lstrip": false,
|
894 |
+
"normalized": true,
|
895 |
+
"rstrip": false,
|
896 |
+
"single_word": false,
|
897 |
+
"special": false
|
898 |
+
},
|
899 |
+
"32109": {
|
900 |
+
"content": "Ø",
|
901 |
+
"lstrip": false,
|
902 |
+
"normalized": true,
|
903 |
+
"rstrip": false,
|
904 |
+
"single_word": false,
|
905 |
+
"special": false
|
906 |
+
},
|
907 |
+
"32110": {
|
908 |
+
"content": "‐",
|
909 |
+
"lstrip": false,
|
910 |
+
"normalized": true,
|
911 |
+
"rstrip": false,
|
912 |
+
"single_word": false,
|
913 |
+
"special": false
|
914 |
+
},
|
915 |
+
"32111": {
|
916 |
+
"content": "Ω",
|
917 |
+
"lstrip": false,
|
918 |
+
"normalized": true,
|
919 |
+
"rstrip": false,
|
920 |
+
"single_word": false,
|
921 |
+
"special": false
|
922 |
+
},
|
923 |
+
"32112": {
|
924 |
+
"content": "ć",
|
925 |
+
"lstrip": false,
|
926 |
+
"normalized": true,
|
927 |
+
"rstrip": false,
|
928 |
+
"single_word": false,
|
929 |
+
"special": false
|
930 |
+
},
|
931 |
+
"32113": {
|
932 |
+
"content": "š",
|
933 |
+
"lstrip": false,
|
934 |
+
"normalized": true,
|
935 |
+
"rstrip": false,
|
936 |
+
"single_word": false,
|
937 |
+
"special": false
|
938 |
+
},
|
939 |
+
"32114": {
|
940 |
+
"content": "ę",
|
941 |
+
"lstrip": false,
|
942 |
+
"normalized": true,
|
943 |
+
"rstrip": false,
|
944 |
+
"single_word": false,
|
945 |
+
"special": false
|
946 |
+
},
|
947 |
+
"32115": {
|
948 |
+
"content": "ś",
|
949 |
+
"lstrip": false,
|
950 |
+
"normalized": true,
|
951 |
+
"rstrip": false,
|
952 |
+
"single_word": false,
|
953 |
+
"special": false
|
954 |
+
},
|
955 |
+
"32116": {
|
956 |
+
"content": "Ó",
|
957 |
+
"lstrip": false,
|
958 |
+
"normalized": true,
|
959 |
+
"rstrip": false,
|
960 |
+
"single_word": false,
|
961 |
+
"special": false
|
962 |
+
},
|
963 |
+
"32117": {
|
964 |
+
"content": "úč",
|
965 |
+
"lstrip": false,
|
966 |
+
"normalized": true,
|
967 |
+
"rstrip": false,
|
968 |
+
"single_word": false,
|
969 |
+
"special": false
|
970 |
+
},
|
971 |
+
"32118": {
|
972 |
+
"content": "ý",
|
973 |
+
"lstrip": false,
|
974 |
+
"normalized": true,
|
975 |
+
"rstrip": false,
|
976 |
+
"single_word": false,
|
977 |
+
"special": false
|
978 |
+
},
|
979 |
+
"32119": {
|
980 |
+
"content": "Ô",
|
981 |
+
"lstrip": false,
|
982 |
+
"normalized": true,
|
983 |
+
"rstrip": false,
|
984 |
+
"single_word": false,
|
985 |
+
"special": false
|
986 |
+
},
|
987 |
+
"32120": {
|
988 |
+
"content": "{",
|
989 |
+
"lstrip": false,
|
990 |
+
"normalized": true,
|
991 |
+
"rstrip": false,
|
992 |
+
"single_word": false,
|
993 |
+
"special": false
|
994 |
+
},
|
995 |
+
"32121": {
|
996 |
+
"content": "œ",
|
997 |
+
"lstrip": false,
|
998 |
+
"normalized": true,
|
999 |
+
"rstrip": false,
|
1000 |
+
"single_word": false,
|
1001 |
+
"special": false
|
1002 |
+
},
|
1003 |
+
"32122": {
|
1004 |
+
"content": "Â",
|
1005 |
+
"lstrip": false,
|
1006 |
+
"normalized": true,
|
1007 |
+
"rstrip": false,
|
1008 |
+
"single_word": false,
|
1009 |
+
"special": false
|
1010 |
+
},
|
1011 |
+
"32123": {
|
1012 |
+
"content": "À",
|
1013 |
+
"lstrip": false,
|
1014 |
+
"normalized": true,
|
1015 |
+
"rstrip": false,
|
1016 |
+
"single_word": false,
|
1017 |
+
"special": false
|
1018 |
+
},
|
1019 |
+
"32124": {
|
1020 |
+
"content": "È",
|
1021 |
+
"lstrip": false,
|
1022 |
+
"normalized": true,
|
1023 |
+
"rstrip": false,
|
1024 |
+
"single_word": false,
|
1025 |
+
"special": false
|
1026 |
+
},
|
1027 |
+
"32125": {
|
1028 |
+
"content": "ï",
|
1029 |
+
"lstrip": false,
|
1030 |
+
"normalized": true,
|
1031 |
+
"rstrip": false,
|
1032 |
+
"single_word": false,
|
1033 |
+
"special": false
|
1034 |
+
},
|
1035 |
+
"32126": {
|
1036 |
+
"content": "ú",
|
1037 |
+
"lstrip": false,
|
1038 |
+
"normalized": true,
|
1039 |
+
"rstrip": false,
|
1040 |
+
"single_word": false,
|
1041 |
+
"special": false
|
1042 |
+
},
|
1043 |
+
"32127": {
|
1044 |
+
"content": "Š",
|
1045 |
+
"lstrip": false,
|
1046 |
+
"normalized": true,
|
1047 |
+
"rstrip": false,
|
1048 |
+
"single_word": false,
|
1049 |
+
"special": false
|
1050 |
+
},
|
1051 |
+
"32128": {
|
1052 |
+
"content": "Ö",
|
1053 |
+
"lstrip": false,
|
1054 |
+
"normalized": true,
|
1055 |
+
"rstrip": false,
|
1056 |
+
"single_word": false,
|
1057 |
+
"special": false
|
1058 |
+
},
|
1059 |
+
"32129": {
|
1060 |
+
"content": "Ä",
|
1061 |
+
"lstrip": false,
|
1062 |
+
"normalized": true,
|
1063 |
+
"rstrip": false,
|
1064 |
+
"single_word": false,
|
1065 |
+
"special": false
|
1066 |
+
},
|
1067 |
+
"32130": {
|
1068 |
+
"content": "ñ",
|
1069 |
+
"lstrip": false,
|
1070 |
+
"normalized": true,
|
1071 |
+
"rstrip": false,
|
1072 |
+
"single_word": false,
|
1073 |
+
"special": false
|
1074 |
}
|
1075 |
},
|
1076 |
"additional_special_tokens": [
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5432
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9957b54933144fb6de9c5132afe8fc93198d9ac6531287b47ee574079ab69fe4
|
3 |
size 5432
|