Upload tokenizer
Browse files- added_tokens.json +12 -11
- tokenizer.json +2 -2
- tokenizer_config.json +19 -11
added_tokens.json
CHANGED
|
@@ -1,22 +1,23 @@
|
|
| 1 |
{
|
| 2 |
-
"Bioderma":
|
| 3 |
"Cerave": 250002,
|
| 4 |
-
"Drunk Elephant":
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
"Hydrating Cleanser": 250003,
|
| 9 |
"Hydro Boost Water Gel": 250006,
|
| 10 |
-
"L'Oréal":
|
| 11 |
"La Roche-Posay": 250004,
|
| 12 |
-
"Myntra":
|
| 13 |
"Neutrogena": 250005,
|
| 14 |
"Niacinamide": 250007,
|
| 15 |
-
"Nykka":
|
| 16 |
"Retinol": 250008,
|
| 17 |
-
"SPF 50":
|
| 18 |
-
"Tatcha":
|
| 19 |
"Vitamin C": 250009,
|
| 20 |
"Walmart": 250016,
|
| 21 |
-
"
|
|
|
|
| 22 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"Bioderma": 250014,
|
| 3 |
"Cerave": 250002,
|
| 4 |
+
"Drunk Elephant": 250013,
|
| 5 |
+
"Eucerin": 250021,
|
| 6 |
+
"EucerinSPF 50": 250011,
|
| 7 |
+
"Flipkart": 250017,
|
| 8 |
"Hydrating Cleanser": 250003,
|
| 9 |
"Hydro Boost Water Gel": 250006,
|
| 10 |
+
"L'Oréal": 250010,
|
| 11 |
"La Roche-Posay": 250004,
|
| 12 |
+
"Myntra": 250019,
|
| 13 |
"Neutrogena": 250005,
|
| 14 |
"Niacinamide": 250007,
|
| 15 |
+
"Nykka": 250018,
|
| 16 |
"Retinol": 250008,
|
| 17 |
+
"SPF 50": 250022,
|
| 18 |
+
"Tatcha": 250012,
|
| 19 |
"Vitamin C": 250009,
|
| 20 |
"Walmart": 250016,
|
| 21 |
+
"amazon": 250020,
|
| 22 |
+
"squalane": 250015
|
| 23 |
}
|
tokenizer.json
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 sha256:6e713901bac6b865d60cb5e0af6742c24d89c104c5edfcfd3fd8ade1787c1d0c
|
| 3 |
+
size 17086930
|
tokenizer_config.json
CHANGED
|
@@ -113,7 +113,7 @@
|
|
| 113 |
"special": false
|
| 114 |
},
|
| 115 |
"250010": {
|
| 116 |
-
"content": "
|
| 117 |
"lstrip": false,
|
| 118 |
"normalized": true,
|
| 119 |
"rstrip": false,
|
|
@@ -121,7 +121,7 @@
|
|
| 121 |
"special": false
|
| 122 |
},
|
| 123 |
"250011": {
|
| 124 |
-
"content": "
|
| 125 |
"lstrip": false,
|
| 126 |
"normalized": true,
|
| 127 |
"rstrip": false,
|
|
@@ -129,7 +129,7 @@
|
|
| 129 |
"special": false
|
| 130 |
},
|
| 131 |
"250012": {
|
| 132 |
-
"content": "
|
| 133 |
"lstrip": false,
|
| 134 |
"normalized": true,
|
| 135 |
"rstrip": false,
|
|
@@ -137,7 +137,7 @@
|
|
| 137 |
"special": false
|
| 138 |
},
|
| 139 |
"250013": {
|
| 140 |
-
"content": "
|
| 141 |
"lstrip": false,
|
| 142 |
"normalized": true,
|
| 143 |
"rstrip": false,
|
|
@@ -145,7 +145,7 @@
|
|
| 145 |
"special": false
|
| 146 |
},
|
| 147 |
"250014": {
|
| 148 |
-
"content": "
|
| 149 |
"lstrip": false,
|
| 150 |
"normalized": true,
|
| 151 |
"rstrip": false,
|
|
@@ -153,7 +153,7 @@
|
|
| 153 |
"special": false
|
| 154 |
},
|
| 155 |
"250015": {
|
| 156 |
-
"content": "
|
| 157 |
"lstrip": false,
|
| 158 |
"normalized": true,
|
| 159 |
"rstrip": false,
|
|
@@ -169,7 +169,7 @@
|
|
| 169 |
"special": false
|
| 170 |
},
|
| 171 |
"250017": {
|
| 172 |
-
"content": "
|
| 173 |
"lstrip": false,
|
| 174 |
"normalized": true,
|
| 175 |
"rstrip": false,
|
|
@@ -177,7 +177,7 @@
|
|
| 177 |
"special": false
|
| 178 |
},
|
| 179 |
"250018": {
|
| 180 |
-
"content": "
|
| 181 |
"lstrip": false,
|
| 182 |
"normalized": true,
|
| 183 |
"rstrip": false,
|
|
@@ -185,7 +185,7 @@
|
|
| 185 |
"special": false
|
| 186 |
},
|
| 187 |
"250019": {
|
| 188 |
-
"content": "
|
| 189 |
"lstrip": false,
|
| 190 |
"normalized": true,
|
| 191 |
"rstrip": false,
|
|
@@ -193,7 +193,7 @@
|
|
| 193 |
"special": false
|
| 194 |
},
|
| 195 |
"250020": {
|
| 196 |
-
"content": "
|
| 197 |
"lstrip": false,
|
| 198 |
"normalized": true,
|
| 199 |
"rstrip": false,
|
|
@@ -201,7 +201,15 @@
|
|
| 201 |
"special": false
|
| 202 |
},
|
| 203 |
"250021": {
|
| 204 |
-
"content": "
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 205 |
"lstrip": false,
|
| 206 |
"normalized": true,
|
| 207 |
"rstrip": false,
|
|
|
|
| 113 |
"special": false
|
| 114 |
},
|
| 115 |
"250010": {
|
| 116 |
+
"content": "L'Oréal",
|
| 117 |
"lstrip": false,
|
| 118 |
"normalized": true,
|
| 119 |
"rstrip": false,
|
|
|
|
| 121 |
"special": false
|
| 122 |
},
|
| 123 |
"250011": {
|
| 124 |
+
"content": "EucerinSPF 50",
|
| 125 |
"lstrip": false,
|
| 126 |
"normalized": true,
|
| 127 |
"rstrip": false,
|
|
|
|
| 129 |
"special": false
|
| 130 |
},
|
| 131 |
"250012": {
|
| 132 |
+
"content": "Tatcha",
|
| 133 |
"lstrip": false,
|
| 134 |
"normalized": true,
|
| 135 |
"rstrip": false,
|
|
|
|
| 137 |
"special": false
|
| 138 |
},
|
| 139 |
"250013": {
|
| 140 |
+
"content": "Drunk Elephant",
|
| 141 |
"lstrip": false,
|
| 142 |
"normalized": true,
|
| 143 |
"rstrip": false,
|
|
|
|
| 145 |
"special": false
|
| 146 |
},
|
| 147 |
"250014": {
|
| 148 |
+
"content": "Bioderma",
|
| 149 |
"lstrip": false,
|
| 150 |
"normalized": true,
|
| 151 |
"rstrip": false,
|
|
|
|
| 153 |
"special": false
|
| 154 |
},
|
| 155 |
"250015": {
|
| 156 |
+
"content": "squalane",
|
| 157 |
"lstrip": false,
|
| 158 |
"normalized": true,
|
| 159 |
"rstrip": false,
|
|
|
|
| 169 |
"special": false
|
| 170 |
},
|
| 171 |
"250017": {
|
| 172 |
+
"content": "Flipkart",
|
| 173 |
"lstrip": false,
|
| 174 |
"normalized": true,
|
| 175 |
"rstrip": false,
|
|
|
|
| 177 |
"special": false
|
| 178 |
},
|
| 179 |
"250018": {
|
| 180 |
+
"content": "Nykka",
|
| 181 |
"lstrip": false,
|
| 182 |
"normalized": true,
|
| 183 |
"rstrip": false,
|
|
|
|
| 185 |
"special": false
|
| 186 |
},
|
| 187 |
"250019": {
|
| 188 |
+
"content": "Myntra",
|
| 189 |
"lstrip": false,
|
| 190 |
"normalized": true,
|
| 191 |
"rstrip": false,
|
|
|
|
| 193 |
"special": false
|
| 194 |
},
|
| 195 |
"250020": {
|
| 196 |
+
"content": "amazon",
|
| 197 |
"lstrip": false,
|
| 198 |
"normalized": true,
|
| 199 |
"rstrip": false,
|
|
|
|
| 201 |
"special": false
|
| 202 |
},
|
| 203 |
"250021": {
|
| 204 |
+
"content": "Eucerin",
|
| 205 |
+
"lstrip": false,
|
| 206 |
+
"normalized": true,
|
| 207 |
+
"rstrip": false,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": false
|
| 210 |
+
},
|
| 211 |
+
"250022": {
|
| 212 |
+
"content": "SPF 50",
|
| 213 |
"lstrip": false,
|
| 214 |
"normalized": true,
|
| 215 |
"rstrip": false,
|