Hugging Face's logo

zjunlp
/
MolGen-7b

Text Generation
Transformers
PyTorch
llama
chemistry
biology
text-generation-inference
Inference Endpoints
Model card Files Community
1
ZJU-Fangyin commited on
Commit
ba8944e
•
1 Parent(s): 0fd57f3

Upload tokenizer

Browse files
Files changed (6) hide show
  1. added_tokens.json +183 -0
  2. merges.txt +1 -0
  3. special_tokens_map.json +33 -0
  4. tokenizer.json +1714 -0
  5. tokenizer_config.json +64 -0
  6. vocab.json +1 -0
added_tokens.json ADDED
@@ -0,0 +1,183 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<mask>": 184,
3
+ "[#Branch1]": 12,
4
+ "[#Branch2]": 178,
5
+ "[#C-1]": 181,
6
+ "[#C]": 165,
7
+ "[#N+1]": 120,
8
+ "[#N]": 23,
9
+ "[#P]": 87,
10
+ "[#Ring1]": 128,
11
+ "[#S]": 40,
12
+ "[-/Ring1]": 129,
13
+ "[-/Ring2]": 156,
14
+ "[-\\Ring1]": 164,
15
+ "[-\\Ring2]": 34,
16
+ "[/123I]": 7,
17
+ "[/B]": 125,
18
+ "[/Br]": 149,
19
+ "[/C-1]": 68,
20
+ "[/C@@H1]": 13,
21
+ "[/C@@]": 14,
22
+ "[/C@H1]": 177,
23
+ "[/C@]": 31,
24
+ "[/CH0]": 126,
25
+ "[/CH1-1]": 88,
26
+ "[/CH1]": 10,
27
+ "[/C]": 118,
28
+ "[/Cl]": 155,
29
+ "[/F]": 6,
30
+ "[/I]": 83,
31
+ "[/N+1]": 152,
32
+ "[/NH0]": 55,
33
+ "[/NH1]": 24,
34
+ "[/N]": 67,
35
+ "[/O-1]": 171,
36
+ "[/OH0]": 174,
37
+ "[/O]": 69,
38
+ "[/P@@]": 16,
39
+ "[/P]": 79,
40
+ "[/S+1]": 47,
41
+ "[/S@@+1]": 21,
42
+ "[/S@@]": 160,
43
+ "[/S@]": 77,
44
+ "[/S]": 112,
45
+ "[/Si]": 38,
46
+ "[11CH3]": 107,
47
+ "[123I]": 124,
48
+ "[124I]": 136,
49
+ "[125I]": 39,
50
+ "[127I]": 36,
51
+ "[17F]": 143,
52
+ "[18F]": 97,
53
+ "[18OH1]": 76,
54
+ "[3H]": 137,
55
+ "[=B]": 25,
56
+ "[=Branch1]": 33,
57
+ "[=Branch2]": 100,
58
+ "[=CH0]": 78,
59
+ "[=C]": 35,
60
+ "[=N+1]": 144,
61
+ "[=N-1]": 161,
62
+ "[=NH0]": 27,
63
+ "[=N]": 105,
64
+ "[=O+1]": 96,
65
+ "[=O]": 30,
66
+ "[=P+1]": 98,
67
+ "[=P@@H1]": 32,
68
+ "[=P@@]": 53,
69
+ "[=P@H1]": 147,
70
+ "[=P@]": 157,
71
+ "[=PH1]": 132,
72
+ "[=P]": 86,
73
+ "[=Ring1]": 163,
74
+ "[=Ring2]": 52,
75
+ "[=S+1]": 133,
76
+ "[=S@+1]": 167,
77
+ "[=S@@+1]": 42,
78
+ "[=S@@H1]": 106,
79
+ "[=S@@]": 5,
80
+ "[=S@]": 44,
81
+ "[=SH1]": 117,
82
+ "[=S]": 73,
83
+ "[B-1]": 122,
84
+ "[B@-1]": 179,
85
+ "[B@@-1]": 17,
86
+ "[B@@H1-1]": 56,
87
+ "[B@H1-1]": 121,
88
+ "[BH1-1]": 54,
89
+ "[BH2-1]": 91,
90
+ "[BH3-1]": 75,
91
+ "[B]": 108,
92
+ "[Br+1]": 166,
93
+ "[Br]": 62,
94
+ "[Branch1]": 104,
95
+ "[Branch2]": 140,
96
+ "[C+1]": 65,
97
+ "[C-1]": 8,
98
+ "[C@@H1]": 15,
99
+ "[C@@]": 173,
100
+ "[C@H1]": 29,
101
+ "[C@]": 9,
102
+ "[CH0]": 28,
103
+ "[CH1+1]": 41,
104
+ "[CH1-1]": 4,
105
+ "[CH1]": 115,
106
+ "[CH2-1]": 110,
107
+ "[CH2]": 146,
108
+ "[C]": 139,
109
+ "[Cl]": 11,
110
+ "[F+1]": 50,
111
+ "[F]": 66,
112
+ "[I]": 63,
113
+ "[N+1]": 150,
114
+ "[N-1]": 84,
115
+ "[N@+1]": 111,
116
+ "[N@@+1]": 131,
117
+ "[N@@H1+1]": 162,
118
+ "[NH0]": 148,
119
+ "[NH1]": 130,
120
+ "[N]": 19,
121
+ "[O+1]": 85,
122
+ "[O-1]": 58,
123
+ "[OH0]": 71,
124
+ "[O]": 20,
125
+ "[P+1]": 82,
126
+ "[P@+1]": 158,
127
+ "[P@@+1]": 94,
128
+ "[P@@H1]": 170,
129
+ "[P@@]": 59,
130
+ "[P@H1]": 81,
131
+ "[P@]": 175,
132
+ "[PH1]": 134,
133
+ "[PH2]": 51,
134
+ "[P]": 80,
135
+ "[Ring1]": 70,
136
+ "[Ring2]": 90,
137
+ "[S+1]": 61,
138
+ "[S@+1]": 72,
139
+ "[S@@+1]": 57,
140
+ "[S@@H1]": 45,
141
+ "[S@@]": 154,
142
+ "[S@]": 93,
143
+ "[SH0]": 182,
144
+ "[SH1]": 123,
145
+ "[SH2]": 101,
146
+ "[SH3]": 102,
147
+ "[S]": 60,
148
+ "[Si]": 119,
149
+ "[Sn+2]": 183,
150
+ "[Sn+3]": 135,
151
+ "[SnH1]": 109,
152
+ "[SnH2]": 99,
153
+ "[Sn]": 114,
154
+ "[\\123I]": 89,
155
+ "[\\B-1]": 64,
156
+ "[\\B]": 159,
157
+ "[\\Br]": 142,
158
+ "[\\C-1]": 127,
159
+ "[\\C@@H1]": 22,
160
+ "[\\C@@]": 46,
161
+ "[\\C@H1]": 49,
162
+ "[\\C@]": 145,
163
+ "[\\CH0]": 176,
164
+ "[\\CH1-1]": 153,
165
+ "[\\C]": 43,
166
+ "[\\Cl]": 37,
167
+ "[\\F]": 26,
168
+ "[\\I]": 172,
169
+ "[\\N+1]": 92,
170
+ "[\\NH1]": 103,
171
+ "[\\N]": 138,
172
+ "[\\O-1]": 141,
173
+ "[\\O]": 18,
174
+ "[\\P@@]": 113,
175
+ "[\\P]": 74,
176
+ "[\\S+1]": 95,
177
+ "[\\S@@+1]": 169,
178
+ "[\\S@@]": 180,
179
+ "[\\S@]": 116,
180
+ "[\\SH1]": 151,
181
+ "[\\S]": 48,
182
+ "[\\Si]": 168
183
+ }
merges.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ #version: 0.2
special_tokens_map.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eos_token": "</s>",
11
+ "mask_token": {
12
+ "content": "<mask>",
13
+ "lstrip": true,
14
+ "normalized": true,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "pad_token": "<pad>",
19
+ "sep_token": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "unk_token": {
27
+ "content": "<unk>",
28
+ "lstrip": false,
29
+ "normalized": true,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
tokenizer.json ADDED
@@ -0,0 +1,1714 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "1.0",
3
+ "truncation": null,
4
+ "padding": null,
5
+ "added_tokens": [
6
+ {
7
+ "id": 0,
8
+ "content": "<s>",
9
+ "single_word": false,
10
+ "lstrip": false,
11
+ "rstrip": false,
12
+ "normalized": true,
13
+ "special": true
14
+ },
15
+ {
16
+ "id": 1,
17
+ "content": "<pad>",
18
+ "single_word": false,
19
+ "lstrip": false,
20
+ "rstrip": false,
21
+ "normalized": false,
22
+ "special": true
23
+ },
24
+ {
25
+ "id": 2,
26
+ "content": "</s>",
27
+ "single_word": false,
28
+ "lstrip": false,
29
+ "rstrip": false,
30
+ "normalized": true,
31
+ "special": true
32
+ },
33
+ {
34
+ "id": 3,
35
+ "content": "<unk>",
36
+ "single_word": false,
37
+ "lstrip": false,
38
+ "rstrip": false,
39
+ "normalized": true,
40
+ "special": true
41
+ },
42
+ {
43
+ "id": 4,
44
+ "content": "[CH1-1]",
45
+ "single_word": false,
46
+ "lstrip": false,
47
+ "rstrip": false,
48
+ "normalized": true,
49
+ "special": false
50
+ },
51
+ {
52
+ "id": 5,
53
+ "content": "[=S@@]",
54
+ "single_word": false,
55
+ "lstrip": false,
56
+ "rstrip": false,
57
+ "normalized": true,
58
+ "special": false
59
+ },
60
+ {
61
+ "id": 6,
62
+ "content": "[/F]",
63
+ "single_word": false,
64
+ "lstrip": false,
65
+ "rstrip": false,
66
+ "normalized": true,
67
+ "special": false
68
+ },
69
+ {
70
+ "id": 7,
71
+ "content": "[/123I]",
72
+ "single_word": false,
73
+ "lstrip": false,
74
+ "rstrip": false,
75
+ "normalized": true,
76
+ "special": false
77
+ },
78
+ {
79
+ "id": 8,
80
+ "content": "[C-1]",
81
+ "single_word": false,
82
+ "lstrip": false,
83
+ "rstrip": false,
84
+ "normalized": true,
85
+ "special": false
86
+ },
87
+ {
88
+ "id": 9,
89
+ "content": "[C@]",
90
+ "single_word": false,
91
+ "lstrip": false,
92
+ "rstrip": false,
93
+ "normalized": true,
94
+ "special": false
95
+ },
96
+ {
97
+ "id": 10,
98
+ "content": "[/CH1]",
99
+ "single_word": false,
100
+ "lstrip": false,
101
+ "rstrip": false,
102
+ "normalized": true,
103
+ "special": false
104
+ },
105
+ {
106
+ "id": 11,
107
+ "content": "[Cl]",
108
+ "single_word": false,
109
+ "lstrip": false,
110
+ "rstrip": false,
111
+ "normalized": true,
112
+ "special": false
113
+ },
114
+ {
115
+ "id": 12,
116
+ "content": "[#Branch1]",
117
+ "single_word": false,
118
+ "lstrip": false,
119
+ "rstrip": false,
120
+ "normalized": true,
121
+ "special": false
122
+ },
123
+ {
124
+ "id": 13,
125
+ "content": "[/C@@H1]",
126
+ "single_word": false,
127
+ "lstrip": false,
128
+ "rstrip": false,
129
+ "normalized": true,
130
+ "special": false
131
+ },
132
+ {
133
+ "id": 14,
134
+ "content": "[/C@@]",
135
+ "single_word": false,
136
+ "lstrip": false,
137
+ "rstrip": false,
138
+ "normalized": true,
139
+ "special": false
140
+ },
141
+ {
142
+ "id": 15,
143
+ "content": "[C@@H1]",
144
+ "single_word": false,
145
+ "lstrip": false,
146
+ "rstrip": false,
147
+ "normalized": true,
148
+ "special": false
149
+ },
150
+ {
151
+ "id": 16,
152
+ "content": "[/P@@]",
153
+ "single_word": false,
154
+ "lstrip": false,
155
+ "rstrip": false,
156
+ "normalized": true,
157
+ "special": false
158
+ },
159
+ {
160
+ "id": 17,
161
+ "content": "[B@@-1]",
162
+ "single_word": false,
163
+ "lstrip": false,
164
+ "rstrip": false,
165
+ "normalized": true,
166
+ "special": false
167
+ },
168
+ {
169
+ "id": 18,
170
+ "content": "[\\O]",
171
+ "single_word": false,
172
+ "lstrip": false,
173
+ "rstrip": false,
174
+ "normalized": true,
175
+ "special": false
176
+ },
177
+ {
178
+ "id": 19,
179
+ "content": "[N]",
180
+ "single_word": false,
181
+ "lstrip": false,
182
+ "rstrip": false,
183
+ "normalized": true,
184
+ "special": false
185
+ },
186
+ {
187
+ "id": 20,
188
+ "content": "[O]",
189
+ "single_word": false,
190
+ "lstrip": false,
191
+ "rstrip": false,
192
+ "normalized": true,
193
+ "special": false
194
+ },
195
+ {
196
+ "id": 21,
197
+ "content": "[/S@@+1]",
198
+ "single_word": false,
199
+ "lstrip": false,
200
+ "rstrip": false,
201
+ "normalized": true,
202
+ "special": false
203
+ },
204
+ {
205
+ "id": 22,
206
+ "content": "[\\C@@H1]",
207
+ "single_word": false,
208
+ "lstrip": false,
209
+ "rstrip": false,
210
+ "normalized": true,
211
+ "special": false
212
+ },
213
+ {
214
+ "id": 23,
215
+ "content": "[#N]",
216
+ "single_word": false,
217
+ "lstrip": false,
218
+ "rstrip": false,
219
+ "normalized": true,
220
+ "special": false
221
+ },
222
+ {
223
+ "id": 24,
224
+ "content": "[/NH1]",
225
+ "single_word": false,
226
+ "lstrip": false,
227
+ "rstrip": false,
228
+ "normalized": true,
229
+ "special": false
230
+ },
231
+ {
232
+ "id": 25,
233
+ "content": "[=B]",
234
+ "single_word": false,
235
+ "lstrip": false,
236
+ "rstrip": false,
237
+ "normalized": true,
238
+ "special": false
239
+ },
240
+ {
241
+ "id": 26,
242
+ "content": "[\\F]",
243
+ "single_word": false,
244
+ "lstrip": false,
245
+ "rstrip": false,
246
+ "normalized": true,
247
+ "special": false
248
+ },
249
+ {
250
+ "id": 27,
251
+ "content": "[=NH0]",
252
+ "single_word": false,
253
+ "lstrip": false,
254
+ "rstrip": false,
255
+ "normalized": true,
256
+ "special": false
257
+ },
258
+ {
259
+ "id": 28,
260
+ "content": "[CH0]",
261
+ "single_word": false,
262
+ "lstrip": false,
263
+ "rstrip": false,
264
+ "normalized": true,
265
+ "special": false
266
+ },
267
+ {
268
+ "id": 29,
269
+ "content": "[C@H1]",
270
+ "single_word": false,
271
+ "lstrip": false,
272
+ "rstrip": false,
273
+ "normalized": true,
274
+ "special": false
275
+ },
276
+ {
277
+ "id": 30,
278
+ "content": "[=O]",
279
+ "single_word": false,
280
+ "lstrip": false,
281
+ "rstrip": false,
282
+ "normalized": true,
283
+ "special": false
284
+ },
285
+ {
286
+ "id": 31,
287
+ "content": "[/C@]",
288
+ "single_word": false,
289
+ "lstrip": false,
290
+ "rstrip": false,
291
+ "normalized": true,
292
+ "special": false
293
+ },
294
+ {
295
+ "id": 32,
296
+ "content": "[=P@@H1]",
297
+ "single_word": false,
298
+ "lstrip": false,
299
+ "rstrip": false,
300
+ "normalized": true,
301
+ "special": false
302
+ },
303
+ {
304
+ "id": 33,
305
+ "content": "[=Branch1]",
306
+ "single_word": false,
307
+ "lstrip": false,
308
+ "rstrip": false,
309
+ "normalized": true,
310
+ "special": false
311
+ },
312
+ {
313
+ "id": 34,
314
+ "content": "[-\\Ring2]",
315
+ "single_word": false,
316
+ "lstrip": false,
317
+ "rstrip": false,
318
+ "normalized": true,
319
+ "special": false
320
+ },
321
+ {
322
+ "id": 35,
323
+ "content": "[=C]",
324
+ "single_word": false,
325
+ "lstrip": false,
326
+ "rstrip": false,
327
+ "normalized": true,
328
+ "special": false
329
+ },
330
+ {
331
+ "id": 36,
332
+ "content": "[127I]",
333
+ "single_word": false,
334
+ "lstrip": false,
335
+ "rstrip": false,
336
+ "normalized": true,
337
+ "special": false
338
+ },
339
+ {
340
+ "id": 37,
341
+ "content": "[\\Cl]",
342
+ "single_word": false,
343
+ "lstrip": false,
344
+ "rstrip": false,
345
+ "normalized": true,
346
+ "special": false
347
+ },
348
+ {
349
+ "id": 38,
350
+ "content": "[/Si]",
351
+ "single_word": false,
352
+ "lstrip": false,
353
+ "rstrip": false,
354
+ "normalized": true,
355
+ "special": false
356
+ },
357
+ {
358
+ "id": 39,
359
+ "content": "[125I]",
360
+ "single_word": false,
361
+ "lstrip": false,
362
+ "rstrip": false,
363
+ "normalized": true,
364
+ "special": false
365
+ },
366
+ {
367
+ "id": 40,
368
+ "content": "[#S]",
369
+ "single_word": false,
370
+ "lstrip": false,
371
+ "rstrip": false,
372
+ "normalized": true,
373
+ "special": false
374
+ },
375
+ {
376
+ "id": 41,
377
+ "content": "[CH1+1]",
378
+ "single_word": false,
379
+ "lstrip": false,
380
+ "rstrip": false,
381
+ "normalized": true,
382
+ "special": false
383
+ },
384
+ {
385
+ "id": 42,
386
+ "content": "[=S@@+1]",
387
+ "single_word": false,
388
+ "lstrip": false,
389
+ "rstrip": false,
390
+ "normalized": true,
391
+ "special": false
392
+ },
393
+ {
394
+ "id": 43,
395
+ "content": "[\\C]",
396
+ "single_word": false,
397
+ "lstrip": false,
398
+ "rstrip": false,
399
+ "normalized": true,
400
+ "special": false
401
+ },
402
+ {
403
+ "id": 44,
404
+ "content": "[=S@]",
405
+ "single_word": false,
406
+ "lstrip": false,
407
+ "rstrip": false,
408
+ "normalized": true,
409
+ "special": false
410
+ },
411
+ {
412
+ "id": 45,
413
+ "content": "[S@@H1]",
414
+ "single_word": false,
415
+ "lstrip": false,
416
+ "rstrip": false,
417
+ "normalized": true,
418
+ "special": false
419
+ },
420
+ {
421
+ "id": 46,
422
+ "content": "[\\C@@]",
423
+ "single_word": false,
424
+ "lstrip": false,
425
+ "rstrip": false,
426
+ "normalized": true,
427
+ "special": false
428
+ },
429
+ {
430
+ "id": 47,
431
+ "content": "[/S+1]",
432
+ "single_word": false,
433
+ "lstrip": false,
434
+ "rstrip": false,
435
+ "normalized": true,
436
+ "special": false
437
+ },
438
+ {
439
+ "id": 48,
440
+ "content": "[\\S]",
441
+ "single_word": false,
442
+ "lstrip": false,
443
+ "rstrip": false,
444
+ "normalized": true,
445
+ "special": false
446
+ },
447
+ {
448
+ "id": 49,
449
+ "content": "[\\C@H1]",
450
+ "single_word": false,
451
+ "lstrip": false,
452
+ "rstrip": false,
453
+ "normalized": true,
454
+ "special": false
455
+ },
456
+ {
457
+ "id": 50,
458
+ "content": "[F+1]",
459
+ "single_word": false,
460
+ "lstrip": false,
461
+ "rstrip": false,
462
+ "normalized": true,
463
+ "special": false
464
+ },
465
+ {
466
+ "id": 51,
467
+ "content": "[PH2]",
468
+ "single_word": false,
469
+ "lstrip": false,
470
+ "rstrip": false,
471
+ "normalized": true,
472
+ "special": false
473
+ },
474
+ {
475
+ "id": 52,
476
+ "content": "[=Ring2]",
477
+ "single_word": false,
478
+ "lstrip": false,
479
+ "rstrip": false,
480
+ "normalized": true,
481
+ "special": false
482
+ },
483
+ {
484
+ "id": 53,
485
+ "content": "[=P@@]",
486
+ "single_word": false,
487
+ "lstrip": false,
488
+ "rstrip": false,
489
+ "normalized": true,
490
+ "special": false
491
+ },
492
+ {
493
+ "id": 54,
494
+ "content": "[BH1-1]",
495
+ "single_word": false,
496
+ "lstrip": false,
497
+ "rstrip": false,
498
+ "normalized": true,
499
+ "special": false
500
+ },
501
+ {
502
+ "id": 55,
503
+ "content": "[/NH0]",
504
+ "single_word": false,
505
+ "lstrip": false,
506
+ "rstrip": false,
507
+ "normalized": true,
508
+ "special": false
509
+ },
510
+ {
511
+ "id": 56,
512
+ "content": "[B@@H1-1]",
513
+ "single_word": false,
514
+ "lstrip": false,
515
+ "rstrip": false,
516
+ "normalized": true,
517
+ "special": false
518
+ },
519
+ {
520
+ "id": 57,
521
+ "content": "[S@@+1]",
522
+ "single_word": false,
523
+ "lstrip": false,
524
+ "rstrip": false,
525
+ "normalized": true,
526
+ "special": false
527
+ },
528
+ {
529
+ "id": 58,
530
+ "content": "[O-1]",
531
+ "single_word": false,
532
+ "lstrip": false,
533
+ "rstrip": false,
534
+ "normalized": true,
535
+ "special": false
536
+ },
537
+ {
538
+ "id": 59,
539
+ "content": "[P@@]",
540
+ "single_word": false,
541
+ "lstrip": false,
542
+ "rstrip": false,
543
+ "normalized": true,
544
+ "special": false
545
+ },
546
+ {
547
+ "id": 60,
548
+ "content": "[S]",
549
+ "single_word": false,
550
+ "lstrip": false,
551
+ "rstrip": false,
552
+ "normalized": true,
553
+ "special": false
554
+ },
555
+ {
556
+ "id": 61,
557
+ "content": "[S+1]",
558
+ "single_word": false,
559
+ "lstrip": false,
560
+ "rstrip": false,
561
+ "normalized": true,
562
+ "special": false
563
+ },
564
+ {
565
+ "id": 62,
566
+ "content": "[Br]",
567
+ "single_word": false,
568
+ "lstrip": false,
569
+ "rstrip": false,
570
+ "normalized": true,
571
+ "special": false
572
+ },
573
+ {
574
+ "id": 63,
575
+ "content": "[I]",
576
+ "single_word": false,
577
+ "lstrip": false,
578
+ "rstrip": false,
579
+ "normalized": true,
580
+ "special": false
581
+ },
582
+ {
583
+ "id": 64,
584
+ "content": "[\\B-1]",
585
+ "single_word": false,
586
+ "lstrip": false,
587
+ "rstrip": false,
588
+ "normalized": true,
589
+ "special": false
590
+ },
591
+ {
592
+ "id": 65,
593
+ "content": "[C+1]",
594
+ "single_word": false,
595
+ "lstrip": false,
596
+ "rstrip": false,
597
+ "normalized": true,
598
+ "special": false
599
+ },
600
+ {
601
+ "id": 66,
602
+ "content": "[F]",
603
+ "single_word": false,
604
+ "lstrip": false,
605
+ "rstrip": false,
606
+ "normalized": true,
607
+ "special": false
608
+ },
609
+ {
610
+ "id": 67,
611
+ "content": "[/N]",
612
+ "single_word": false,
613
+ "lstrip": false,
614
+ "rstrip": false,
615
+ "normalized": true,
616
+ "special": false
617
+ },
618
+ {
619
+ "id": 68,
620
+ "content": "[/C-1]",
621
+ "single_word": false,
622
+ "lstrip": false,
623
+ "rstrip": false,
624
+ "normalized": true,
625
+ "special": false
626
+ },
627
+ {
628
+ "id": 69,
629
+ "content": "[/O]",
630
+ "single_word": false,
631
+ "lstrip": false,
632
+ "rstrip": false,
633
+ "normalized": true,
634
+ "special": false
635
+ },
636
+ {
637
+ "id": 70,
638
+ "content": "[Ring1]",
639
+ "single_word": false,
640
+ "lstrip": false,
641
+ "rstrip": false,
642
+ "normalized": true,
643
+ "special": false
644
+ },
645
+ {
646
+ "id": 71,
647
+ "content": "[OH0]",
648
+ "single_word": false,
649
+ "lstrip": false,
650
+ "rstrip": false,
651
+ "normalized": true,
652
+ "special": false
653
+ },
654
+ {
655
+ "id": 72,
656
+ "content": "[S@+1]",
657
+ "single_word": false,
658
+ "lstrip": false,
659
+ "rstrip": false,
660
+ "normalized": true,
661
+ "special": false
662
+ },
663
+ {
664
+ "id": 73,
665
+ "content": "[=S]",
666
+ "single_word": false,
667
+ "lstrip": false,
668
+ "rstrip": false,
669
+ "normalized": true,
670
+ "special": false
671
+ },
672
+ {
673
+ "id": 74,
674
+ "content": "[\\P]",
675
+ "single_word": false,
676
+ "lstrip": false,
677
+ "rstrip": false,
678
+ "normalized": true,
679
+ "special": false
680
+ },
681
+ {
682
+ "id": 75,
683
+ "content": "[BH3-1]",
684
+ "single_word": false,
685
+ "lstrip": false,
686
+ "rstrip": false,
687
+ "normalized": true,
688
+ "special": false
689
+ },
690
+ {
691
+ "id": 76,
692
+ "content": "[18OH1]",
693
+ "single_word": false,
694
+ "lstrip": false,
695
+ "rstrip": false,
696
+ "normalized": true,
697
+ "special": false
698
+ },
699
+ {
700
+ "id": 77,
701
+ "content": "[/S@]",
702
+ "single_word": false,
703
+ "lstrip": false,
704
+ "rstrip": false,
705
+ "normalized": true,
706
+ "special": false
707
+ },
708
+ {
709
+ "id": 78,
710
+ "content": "[=CH0]",
711
+ "single_word": false,
712
+ "lstrip": false,
713
+ "rstrip": false,
714
+ "normalized": true,
715
+ "special": false
716
+ },
717
+ {
718
+ "id": 79,
719
+ "content": "[/P]",
720
+ "single_word": false,
721
+ "lstrip": false,
722
+ "rstrip": false,
723
+ "normalized": true,
724
+ "special": false
725
+ },
726
+ {
727
+ "id": 80,
728
+ "content": "[P]",
729
+ "single_word": false,
730
+ "lstrip": false,
731
+ "rstrip": false,
732
+ "normalized": true,
733
+ "special": false
734
+ },
735
+ {
736
+ "id": 81,
737
+ "content": "[P@H1]",
738
+ "single_word": false,
739
+ "lstrip": false,
740
+ "rstrip": false,
741
+ "normalized": true,
742
+ "special": false
743
+ },
744
+ {
745
+ "id": 82,
746
+ "content": "[P+1]",
747
+ "single_word": false,
748
+ "lstrip": false,
749
+ "rstrip": false,
750
+ "normalized": true,
751
+ "special": false
752
+ },
753
+ {
754
+ "id": 83,
755
+ "content": "[/I]",
756
+ "single_word": false,
757
+ "lstrip": false,
758
+ "rstrip": false,
759
+ "normalized": true,
760
+ "special": false
761
+ },
762
+ {
763
+ "id": 84,
764
+ "content": "[N-1]",
765
+ "single_word": false,
766
+ "lstrip": false,
767
+ "rstrip": false,
768
+ "normalized": true,
769
+ "special": false
770
+ },
771
+ {
772
+ "id": 85,
773
+ "content": "[O+1]",
774
+ "single_word": false,
775
+ "lstrip": false,
776
+ "rstrip": false,
777
+ "normalized": true,
778
+ "special": false
779
+ },
780
+ {
781
+ "id": 86,
782
+ "content": "[=P]",
783
+ "single_word": false,
784
+ "lstrip": false,
785
+ "rstrip": false,
786
+ "normalized": true,
787
+ "special": false
788
+ },
789
+ {
790
+ "id": 87,
791
+ "content": "[#P]",
792
+ "single_word": false,
793
+ "lstrip": false,
794
+ "rstrip": false,
795
+ "normalized": true,
796
+ "special": false
797
+ },
798
+ {
799
+ "id": 88,
800
+ "content": "[/CH1-1]",
801
+ "single_word": false,
802
+ "lstrip": false,
803
+ "rstrip": false,
804
+ "normalized": true,
805
+ "special": false
806
+ },
807
+ {
808
+ "id": 89,
809
+ "content": "[\\123I]",
810
+ "single_word": false,
811
+ "lstrip": false,
812
+ "rstrip": false,
813
+ "normalized": true,
814
+ "special": false
815
+ },
816
+ {
817
+ "id": 90,
818
+ "content": "[Ring2]",
819
+ "single_word": false,
820
+ "lstrip": false,
821
+ "rstrip": false,
822
+ "normalized": true,
823
+ "special": false
824
+ },
825
+ {
826
+ "id": 91,
827
+ "content": "[BH2-1]",
828
+ "single_word": false,
829
+ "lstrip": false,
830
+ "rstrip": false,
831
+ "normalized": true,
832
+ "special": false
833
+ },
834
+ {
835
+ "id": 92,
836
+ "content": "[\\N+1]",
837
+ "single_word": false,
838
+ "lstrip": false,
839
+ "rstrip": false,
840
+ "normalized": true,
841
+ "special": false
842
+ },
843
+ {
844
+ "id": 93,
845
+ "content": "[S@]",
846
+ "single_word": false,
847
+ "lstrip": false,
848
+ "rstrip": false,
849
+ "normalized": true,
850
+ "special": false
851
+ },
852
+ {
853
+ "id": 94,
854
+ "content": "[P@@+1]",
855
+ "single_word": false,
856
+ "lstrip": false,
857
+ "rstrip": false,
858
+ "normalized": true,
859
+ "special": false
860
+ },
861
+ {
862
+ "id": 95,
863
+ "content": "[\\S+1]",
864
+ "single_word": false,
865
+ "lstrip": false,
866
+ "rstrip": false,
867
+ "normalized": true,
868
+ "special": false
869
+ },
870
+ {
871
+ "id": 96,
872
+ "content": "[=O+1]",
873
+ "single_word": false,
874
+ "lstrip": false,
875
+ "rstrip": false,
876
+ "normalized": true,
877
+ "special": false
878
+ },
879
+ {
880
+ "id": 97,
881
+ "content": "[18F]",
882
+ "single_word": false,
883
+ "lstrip": false,
884
+ "rstrip": false,
885
+ "normalized": true,
886
+ "special": false
887
+ },
888
+ {
889
+ "id": 98,
890
+ "content": "[=P+1]",
891
+ "single_word": false,
892
+ "lstrip": false,
893
+ "rstrip": false,
894
+ "normalized": true,
895
+ "special": false
896
+ },
897
+ {
898
+ "id": 99,
899
+ "content": "[SnH2]",
900
+ "single_word": false,
901
+ "lstrip": false,
902
+ "rstrip": false,
903
+ "normalized": true,
904
+ "special": false
905
+ },
906
+ {
907
+ "id": 100,
908
+ "content": "[=Branch2]",
909
+ "single_word": false,
910
+ "lstrip": false,
911
+ "rstrip": false,
912
+ "normalized": true,
913
+ "special": false
914
+ },
915
+ {
916
+ "id": 101,
917
+ "content": "[SH2]",
918
+ "single_word": false,
919
+ "lstrip": false,
920
+ "rstrip": false,
921
+ "normalized": true,
922
+ "special": false
923
+ },
924
+ {
925
+ "id": 102,
926
+ "content": "[SH3]",
927
+ "single_word": false,
928
+ "lstrip": false,
929
+ "rstrip": false,
930
+ "normalized": true,
931
+ "special": false
932
+ },
933
+ {
934
+ "id": 103,
935
+ "content": "[\\NH1]",
936
+ "single_word": false,
937
+ "lstrip": false,
938
+ "rstrip": false,
939
+ "normalized": true,
940
+ "special": false
941
+ },
942
+ {
943
+ "id": 104,
944
+ "content": "[Branch1]",
945
+ "single_word": false,
946
+ "lstrip": false,
947
+ "rstrip": false,
948
+ "normalized": true,
949
+ "special": false
950
+ },
951
+ {
952
+ "id": 105,
953
+ "content": "[=N]",
954
+ "single_word": false,
955
+ "lstrip": false,
956
+ "rstrip": false,
957
+ "normalized": true,
958
+ "special": false
959
+ },
960
+ {
961
+ "id": 106,
962
+ "content": "[=S@@H1]",
963
+ "single_word": false,
964
+ "lstrip": false,
965
+ "rstrip": false,
966
+ "normalized": true,
967
+ "special": false
968
+ },
969
+ {
970
+ "id": 107,
971
+ "content": "[11CH3]",
972
+ "single_word": false,
973
+ "lstrip": false,
974
+ "rstrip": false,
975
+ "normalized": true,
976
+ "special": false
977
+ },
978
+ {
979
+ "id": 108,
980
+ "content": "[B]",
981
+ "single_word": false,
982
+ "lstrip": false,
983
+ "rstrip": false,
984
+ "normalized": true,
985
+ "special": false
986
+ },
987
+ {
988
+ "id": 109,
989
+ "content": "[SnH1]",
990
+ "single_word": false,
991
+ "lstrip": false,
992
+ "rstrip": false,
993
+ "normalized": true,
994
+ "special": false
995
+ },
996
+ {
997
+ "id": 110,
998
+ "content": "[CH2-1]",
999
+ "single_word": false,
1000
+ "lstrip": false,
1001
+ "rstrip": false,
1002
+ "normalized": true,
1003
+ "special": false
1004
+ },
1005
+ {
1006
+ "id": 111,
1007
+ "content": "[N@+1]",
1008
+ "single_word": false,
1009
+ "lstrip": false,
1010
+ "rstrip": false,
1011
+ "normalized": true,
1012
+ "special": false
1013
+ },
1014
+ {
1015
+ "id": 112,
1016
+ "content": "[/S]",
1017
+ "single_word": false,
1018
+ "lstrip": false,
1019
+ "rstrip": false,
1020
+ "normalized": true,
1021
+ "special": false
1022
+ },
1023
+ {
1024
+ "id": 113,
1025
+ "content": "[\\P@@]",
1026
+ "single_word": false,
1027
+ "lstrip": false,
1028
+ "rstrip": false,
1029
+ "normalized": true,
1030
+ "special": false
1031
+ },
1032
+ {
1033
+ "id": 114,
1034
+ "content": "[Sn]",
1035
+ "single_word": false,
1036
+ "lstrip": false,
1037
+ "rstrip": false,
1038
+ "normalized": true,
1039
+ "special": false
1040
+ },
1041
+ {
1042
+ "id": 115,
1043
+ "content": "[CH1]",
1044
+ "single_word": false,
1045
+ "lstrip": false,
1046
+ "rstrip": false,
1047
+ "normalized": true,
1048
+ "special": false
1049
+ },
1050
+ {
1051
+ "id": 116,
1052
+ "content": "[\\S@]",
1053
+ "single_word": false,
1054
+ "lstrip": false,
1055
+ "rstrip": false,
1056
+ "normalized": true,
1057
+ "special": false
1058
+ },
1059
+ {
1060
+ "id": 117,
1061
+ "content": "[=SH1]",
1062
+ "single_word": false,
1063
+ "lstrip": false,
1064
+ "rstrip": false,
1065
+ "normalized": true,
1066
+ "special": false
1067
+ },
1068
+ {
1069
+ "id": 118,
1070
+ "content": "[/C]",
1071
+ "single_word": false,
1072
+ "lstrip": false,
1073
+ "rstrip": false,
1074
+ "normalized": true,
1075
+ "special": false
1076
+ },
1077
+ {
1078
+ "id": 119,
1079
+ "content": "[Si]",
1080
+ "single_word": false,
1081
+ "lstrip": false,
1082
+ "rstrip": false,
1083
+ "normalized": true,
1084
+ "special": false
1085
+ },
1086
+ {
1087
+ "id": 120,
1088
+ "content": "[#N+1]",
1089
+ "single_word": false,
1090
+ "lstrip": false,
1091
+ "rstrip": false,
1092
+ "normalized": true,
1093
+ "special": false
1094
+ },
1095
+ {
1096
+ "id": 121,
1097
+ "content": "[B@H1-1]",
1098
+ "single_word": false,
1099
+ "lstrip": false,
1100
+ "rstrip": false,
1101
+ "normalized": true,
1102
+ "special": false
1103
+ },
1104
+ {
1105
+ "id": 122,
1106
+ "content": "[B-1]",
1107
+ "single_word": false,
1108
+ "lstrip": false,
1109
+ "rstrip": false,
1110
+ "normalized": true,
1111
+ "special": false
1112
+ },
1113
+ {
1114
+ "id": 123,
1115
+ "content": "[SH1]",
1116
+ "single_word": false,
1117
+ "lstrip": false,
1118
+ "rstrip": false,
1119
+ "normalized": true,
1120
+ "special": false
1121
+ },
1122
+ {
1123
+ "id": 124,
1124
+ "content": "[123I]",
1125
+ "single_word": false,
1126
+ "lstrip": false,
1127
+ "rstrip": false,
1128
+ "normalized": true,
1129
+ "special": false
1130
+ },
1131
+ {
1132
+ "id": 125,
1133
+ "content": "[/B]",
1134
+ "single_word": false,
1135
+ "lstrip": false,
1136
+ "rstrip": false,
1137
+ "normalized": true,
1138
+ "special": false
1139
+ },
1140
+ {
1141
+ "id": 126,
1142
+ "content": "[/CH0]",
1143
+ "single_word": false,
1144
+ "lstrip": false,
1145
+ "rstrip": false,
1146
+ "normalized": true,
1147
+ "special": false
1148
+ },
1149
+ {
1150
+ "id": 127,
1151
+ "content": "[\\C-1]",
1152
+ "single_word": false,
1153
+ "lstrip": false,
1154
+ "rstrip": false,
1155
+ "normalized": true,
1156
+ "special": false
1157
+ },
1158
+ {
1159
+ "id": 128,
1160
+ "content": "[#Ring1]",
1161
+ "single_word": false,
1162
+ "lstrip": false,
1163
+ "rstrip": false,
1164
+ "normalized": true,
1165
+ "special": false
1166
+ },
1167
+ {
1168
+ "id": 129,
1169
+ "content": "[-/Ring1]",
1170
+ "single_word": false,
1171
+ "lstrip": false,
1172
+ "rstrip": false,
1173
+ "normalized": true,
1174
+ "special": false
1175
+ },
1176
+ {
1177
+ "id": 130,
1178
+ "content": "[NH1]",
1179
+ "single_word": false,
1180
+ "lstrip": false,
1181
+ "rstrip": false,
1182
+ "normalized": true,
1183
+ "special": false
1184
+ },
1185
+ {
1186
+ "id": 131,
1187
+ "content": "[N@@+1]",
1188
+ "single_word": false,
1189
+ "lstrip": false,
1190
+ "rstrip": false,
1191
+ "normalized": true,
1192
+ "special": false
1193
+ },
1194
+ {
1195
+ "id": 132,
1196
+ "content": "[=PH1]",
1197
+ "single_word": false,
1198
+ "lstrip": false,
1199
+ "rstrip": false,
1200
+ "normalized": true,
1201
+ "special": false
1202
+ },
1203
+ {
1204
+ "id": 133,
1205
+ "content": "[=S+1]",
1206
+ "single_word": false,
1207
+ "lstrip": false,
1208
+ "rstrip": false,
1209
+ "normalized": true,
1210
+ "special": false
1211
+ },
1212
+ {
1213
+ "id": 134,
1214
+ "content": "[PH1]",
1215
+ "single_word": false,
1216
+ "lstrip": false,
1217
+ "rstrip": false,
1218
+ "normalized": true,
1219
+ "special": false
1220
+ },
1221
+ {
1222
+ "id": 135,
1223
+ "content": "[Sn+3]",
1224
+ "single_word": false,
1225
+ "lstrip": false,
1226
+ "rstrip": false,
1227
+ "normalized": true,
1228
+ "special": false
1229
+ },
1230
+ {
1231
+ "id": 136,
1232
+ "content": "[124I]",
1233
+ "single_word": false,
1234
+ "lstrip": false,
1235
+ "rstrip": false,
1236
+ "normalized": true,
1237
+ "special": false
1238
+ },
1239
+ {
1240
+ "id": 137,
1241
+ "content": "[3H]",
1242
+ "single_word": false,
1243
+ "lstrip": false,
1244
+ "rstrip": false,
1245
+ "normalized": true,
1246
+ "special": false
1247
+ },
1248
+ {
1249
+ "id": 138,
1250
+ "content": "[\\N]",
1251
+ "single_word": false,
1252
+ "lstrip": false,
1253
+ "rstrip": false,
1254
+ "normalized": true,
1255
+ "special": false
1256
+ },
1257
+ {
1258
+ "id": 139,
1259
+ "content": "[C]",
1260
+ "single_word": false,
1261
+ "lstrip": false,
1262
+ "rstrip": false,
1263
+ "normalized": true,
1264
+ "special": false
1265
+ },
1266
+ {
1267
+ "id": 140,
1268
+ "content": "[Branch2]",
1269
+ "single_word": false,
1270
+ "lstrip": false,
1271
+ "rstrip": false,
1272
+ "normalized": true,
1273
+ "special": false
1274
+ },
1275
+ {
1276
+ "id": 141,
1277
+ "content": "[\\O-1]",
1278
+ "single_word": false,
1279
+ "lstrip": false,
1280
+ "rstrip": false,
1281
+ "normalized": true,
1282
+ "special": false
1283
+ },
1284
+ {
1285
+ "id": 142,
1286
+ "content": "[\\Br]",
1287
+ "single_word": false,
1288
+ "lstrip": false,
1289
+ "rstrip": false,
1290
+ "normalized": true,
1291
+ "special": false
1292
+ },
1293
+ {
1294
+ "id": 143,
1295
+ "content": "[17F]",
1296
+ "single_word": false,
1297
+ "lstrip": false,
1298
+ "rstrip": false,
1299
+ "normalized": true,
1300
+ "special": false
1301
+ },
1302
+ {
1303
+ "id": 144,
1304
+ "content": "[=N+1]",
1305
+ "single_word": false,
1306
+ "lstrip": false,
1307
+ "rstrip": false,
1308
+ "normalized": true,
1309
+ "special": false
1310
+ },
1311
+ {
1312
+ "id": 145,
1313
+ "content": "[\\C@]",
1314
+ "single_word": false,
1315
+ "lstrip": false,
1316
+ "rstrip": false,
1317
+ "normalized": true,
1318
+ "special": false
1319
+ },
1320
+ {
1321
+ "id": 146,
1322
+ "content": "[CH2]",
1323
+ "single_word": false,
1324
+ "lstrip": false,
1325
+ "rstrip": false,
1326
+ "normalized": true,
1327
+ "special": false
1328
+ },
1329
+ {
1330
+ "id": 147,
1331
+ "content": "[=P@H1]",
1332
+ "single_word": false,
1333
+ "lstrip": false,
1334
+ "rstrip": false,
1335
+ "normalized": true,
1336
+ "special": false
1337
+ },
1338
+ {
1339
+ "id": 148,
1340
+ "content": "[NH0]",
1341
+ "single_word": false,
1342
+ "lstrip": false,
1343
+ "rstrip": false,
1344
+ "normalized": true,
1345
+ "special": false
1346
+ },
1347
+ {
1348
+ "id": 149,
1349
+ "content": "[/Br]",
1350
+ "single_word": false,
1351
+ "lstrip": false,
1352
+ "rstrip": false,
1353
+ "normalized": true,
1354
+ "special": false
1355
+ },
1356
+ {
1357
+ "id": 150,
1358
+ "content": "[N+1]",
1359
+ "single_word": false,
1360
+ "lstrip": false,
1361
+ "rstrip": false,
1362
+ "normalized": true,
1363
+ "special": false
1364
+ },
1365
+ {
1366
+ "id": 151,
1367
+ "content": "[\\SH1]",
1368
+ "single_word": false,
1369
+ "lstrip": false,
1370
+ "rstrip": false,
1371
+ "normalized": true,
1372
+ "special": false
1373
+ },
1374
+ {
1375
+ "id": 152,
1376
+ "content": "[/N+1]",
1377
+ "single_word": false,
1378
+ "lstrip": false,
1379
+ "rstrip": false,
1380
+ "normalized": true,
1381
+ "special": false
1382
+ },
1383
+ {
1384
+ "id": 153,
1385
+ "content": "[\\CH1-1]",
1386
+ "single_word": false,
1387
+ "lstrip": false,
1388
+ "rstrip": false,
1389
+ "normalized": true,
1390
+ "special": false
1391
+ },
1392
+ {
1393
+ "id": 154,
1394
+ "content": "[S@@]",
1395
+ "single_word": false,
1396
+ "lstrip": false,
1397
+ "rstrip": false,
1398
+ "normalized": true,
1399
+ "special": false
1400
+ },
1401
+ {
1402
+ "id": 155,
1403
+ "content": "[/Cl]",
1404
+ "single_word": false,
1405
+ "lstrip": false,
1406
+ "rstrip": false,
1407
+ "normalized": true,
1408
+ "special": false
1409
+ },
1410
+ {
1411
+ "id": 156,
1412
+ "content": "[-/Ring2]",
1413
+ "single_word": false,
1414
+ "lstrip": false,
1415
+ "rstrip": false,
1416
+ "normalized": true,
1417
+ "special": false
1418
+ },
1419
+ {
1420
+ "id": 157,
1421
+ "content": "[=P@]",
1422
+ "single_word": false,
1423
+ "lstrip": false,
1424
+ "rstrip": false,
1425
+ "normalized": true,
1426
+ "special": false
1427
+ },
1428
+ {
1429
+ "id": 158,
1430
+ "content": "[P@+1]",
1431
+ "single_word": false,
1432
+ "lstrip": false,
1433
+ "rstrip": false,
1434
+ "normalized": true,
1435
+ "special": false
1436
+ },
1437
+ {
1438
+ "id": 159,
1439
+ "content": "[\\B]",
1440
+ "single_word": false,
1441
+ "lstrip": false,
1442
+ "rstrip": false,
1443
+ "normalized": true,
1444
+ "special": false
1445
+ },
1446
+ {
1447
+ "id": 160,
1448
+ "content": "[/S@@]",
1449
+ "single_word": false,
1450
+ "lstrip": false,
1451
+ "rstrip": false,
1452
+ "normalized": true,
1453
+ "special": false
1454
+ },
1455
+ {
1456
+ "id": 161,
1457
+ "content": "[=N-1]",
1458
+ "single_word": false,
1459
+ "lstrip": false,
1460
+ "rstrip": false,
1461
+ "normalized": true,
1462
+ "special": false
1463
+ },
1464
+ {
1465
+ "id": 162,
1466
+ "content": "[N@@H1+1]",
1467
+ "single_word": false,
1468
+ "lstrip": false,
1469
+ "rstrip": false,
1470
+ "normalized": true,
1471
+ "special": false
1472
+ },
1473
+ {
1474
+ "id": 163,
1475
+ "content": "[=Ring1]",
1476
+ "single_word": false,
1477
+ "lstrip": false,
1478
+ "rstrip": false,
1479
+ "normalized": true,
1480
+ "special": false
1481
+ },
1482
+ {
1483
+ "id": 164,
1484
+ "content": "[-\\Ring1]",
1485
+ "single_word": false,
1486
+ "lstrip": false,
1487
+ "rstrip": false,
1488
+ "normalized": true,
1489
+ "special": false
1490
+ },
1491
+ {
1492
+ "id": 165,
1493
+ "content": "[#C]",
1494
+ "single_word": false,
1495
+ "lstrip": false,
1496
+ "rstrip": false,
1497
+ "normalized": true,
1498
+ "special": false
1499
+ },
1500
+ {
1501
+ "id": 166,
1502
+ "content": "[Br+1]",
1503
+ "single_word": false,
1504
+ "lstrip": false,
1505
+ "rstrip": false,
1506
+ "normalized": true,
1507
+ "special": false
1508
+ },
1509
+ {
1510
+ "id": 167,
1511
+ "content": "[=S@+1]",
1512
+ "single_word": false,
1513
+ "lstrip": false,
1514
+ "rstrip": false,
1515
+ "normalized": true,
1516
+ "special": false
1517
+ },
1518
+ {
1519
+ "id": 168,
1520
+ "content": "[\\Si]",
1521
+ "single_word": false,
1522
+ "lstrip": false,
1523
+ "rstrip": false,
1524
+ "normalized": true,
1525
+ "special": false
1526
+ },
1527
+ {
1528
+ "id": 169,
1529
+ "content": "[\\S@@+1]",
1530
+ "single_word": false,
1531
+ "lstrip": false,
1532
+ "rstrip": false,
1533
+ "normalized": true,
1534
+ "special": false
1535
+ },
1536
+ {
1537
+ "id": 170,
1538
+ "content": "[P@@H1]",
1539
+ "single_word": false,
1540
+ "lstrip": false,
1541
+ "rstrip": false,
1542
+ "normalized": true,
1543
+ "special": false
1544
+ },
1545
+ {
1546
+ "id": 171,
1547
+ "content": "[/O-1]",
1548
+ "single_word": false,
1549
+ "lstrip": false,
1550
+ "rstrip": false,
1551
+ "normalized": true,
1552
+ "special": false
1553
+ },
1554
+ {
1555
+ "id": 172,
1556
+ "content": "[\\I]",
1557
+ "single_word": false,
1558
+ "lstrip": false,
1559
+ "rstrip": false,
1560
+ "normalized": true,
1561
+ "special": false
1562
+ },
1563
+ {
1564
+ "id": 173,
1565
+ "content": "[C@@]",
1566
+ "single_word": false,
1567
+ "lstrip": false,
1568
+ "rstrip": false,
1569
+ "normalized": true,
1570
+ "special": false
1571
+ },
1572
+ {
1573
+ "id": 174,
1574
+ "content": "[/OH0]",
1575
+ "single_word": false,
1576
+ "lstrip": false,
1577
+ "rstrip": false,
1578
+ "normalized": true,
1579
+ "special": false
1580
+ },
1581
+ {
1582
+ "id": 175,
1583
+ "content": "[P@]",
1584
+ "single_word": false,
1585
+ "lstrip": false,
1586
+ "rstrip": false,
1587
+ "normalized": true,
1588
+ "special": false
1589
+ },
1590
+ {
1591
+ "id": 176,
1592
+ "content": "[\\CH0]",
1593
+ "single_word": false,
1594
+ "lstrip": false,
1595
+ "rstrip": false,
1596
+ "normalized": true,
1597
+ "special": false
1598
+ },
1599
+ {
1600
+ "id": 177,
1601
+ "content": "[/C@H1]",
1602
+ "single_word": false,
1603
+ "lstrip": false,
1604
+ "rstrip": false,
1605
+ "normalized": true,
1606
+ "special": false
1607
+ },
1608
+ {
1609
+ "id": 178,
1610
+ "content": "[#Branch2]",
1611
+ "single_word": false,
1612
+ "lstrip": false,
1613
+ "rstrip": false,
1614
+ "normalized": true,
1615
+ "special": false
1616
+ },
1617
+ {
1618
+ "id": 179,
1619
+ "content": "[B@-1]",
1620
+ "single_word": false,
1621
+ "lstrip": false,
1622
+ "rstrip": false,
1623
+ "normalized": true,
1624
+ "special": false
1625
+ },
1626
+ {
1627
+ "id": 180,
1628
+ "content": "[\\S@@]",
1629
+ "single_word": false,
1630
+ "lstrip": false,
1631
+ "rstrip": false,
1632
+ "normalized": true,
1633
+ "special": false
1634
+ },
1635
+ {
1636
+ "id": 181,
1637
+ "content": "[#C-1]",
1638
+ "single_word": false,
1639
+ "lstrip": false,
1640
+ "rstrip": false,
1641
+ "normalized": true,
1642
+ "special": false
1643
+ },
1644
+ {
1645
+ "id": 182,
1646
+ "content": "[SH0]",
1647
+ "single_word": false,
1648
+ "lstrip": false,
1649
+ "rstrip": false,
1650
+ "normalized": true,
1651
+ "special": false
1652
+ },
1653
+ {
1654
+ "id": 183,
1655
+ "content": "[Sn+2]",
1656
+ "single_word": false,
1657
+ "lstrip": false,
1658
+ "rstrip": false,
1659
+ "normalized": true,
1660
+ "special": false
1661
+ },
1662
+ {
1663
+ "id": 184,
1664
+ "content": "<mask>",
1665
+ "single_word": false,
1666
+ "lstrip": true,
1667
+ "rstrip": false,
1668
+ "normalized": true,
1669
+ "special": true
1670
+ }
1671
+ ],
1672
+ "normalizer": null,
1673
+ "pre_tokenizer": {
1674
+ "type": "ByteLevel",
1675
+ "add_prefix_space": false,
1676
+ "trim_offsets": true,
1677
+ "use_regex": true
1678
+ },
1679
+ "post_processor": {
1680
+ "type": "RobertaProcessing",
1681
+ "sep": [
1682
+ "</s>",
1683
+ 2
1684
+ ],
1685
+ "cls": [
1686
+ "<s>",
1687
+ 0
1688
+ ],
1689
+ "trim_offsets": true,
1690
+ "add_prefix_space": false
1691
+ },
1692
+ "decoder": {
1693
+ "type": "ByteLevel",
1694
+ "add_prefix_space": true,
1695
+ "trim_offsets": true,
1696
+ "use_regex": true
1697
+ },
1698
+ "model": {
1699
+ "type": "BPE",
1700
+ "dropout": null,
1701
+ "unk_token": null,
1702
+ "continuing_subword_prefix": "",
1703
+ "end_of_word_suffix": "",
1704
+ "fuse_unk": false,
1705
+ "byte_fallback": false,
1706
+ "vocab": {
1707
+ "<s>": 0,
1708
+ "<pad>": 1,
1709
+ "</s>": 2,
1710
+ "<unk>": 3
1711
+ },
1712
+ "merges": []
1713
+ }
1714
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": {
4
+ "__type": "AddedToken",
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "clean_up_tokenization_spaces": true,
12
+ "cls_token": {
13
+ "__type": "AddedToken",
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ },
20
+ "eos_token": {
21
+ "__type": "AddedToken",
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": true,
25
+ "rstrip": false,
26
+ "single_word": false
27
+ },
28
+ "errors": "replace",
29
+ "mask_token": {
30
+ "__type": "AddedToken",
31
+ "content": "<mask>",
32
+ "lstrip": true,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "model_max_length": 1024,
38
+ "pad_token": {
39
+ "__type": "AddedToken",
40
+ "content": "<pad>",
41
+ "lstrip": false,
42
+ "normalized": true,
43
+ "rstrip": false,
44
+ "single_word": false
45
+ },
46
+ "sep_token": {
47
+ "__type": "AddedToken",
48
+ "content": "</s>",
49
+ "lstrip": false,
50
+ "normalized": true,
51
+ "rstrip": false,
52
+ "single_word": false
53
+ },
54
+ "tokenizer_class": "BartTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": {
57
+ "__type": "AddedToken",
58
+ "content": "<unk>",
59
+ "lstrip": false,
60
+ "normalized": true,
61
+ "rstrip": false,
62
+ "single_word": false
63
+ }
64
+ }
vocab.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"<s>":0,"<pad>":1,"</s>":2,"<unk>":3}