andreas122001 commited on
Commit
a4eebb5
1 Parent(s): d3d7101

Upload 10 files

Browse files
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a23afca23cb18bd4ab9093093679f284eefdc69be46267c7861bf5fdea2056ba
3
+ size 133467576
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba606569a67e41a0cdd9f9c6173d986feb13cb3380dc788c5ded71affbf36244
3
+ size 267054330
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38f537364220577dd4e5269dcefc13c34f9b3778f6daf6768a70fafb5a921478
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1c16f6410648ec04afd4538b993744a9235b715a4b6e83844e6981759b4ebeb
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
@@ -0,0 +1,194 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "1.0",
3
+ "truncation": {
4
+ "direction": "Right",
5
+ "max_length": 81,
6
+ "strategy": "LongestFirst",
7
+ "stride": 0
8
+ },
9
+ "padding": {
10
+ "strategy": {
11
+ "Fixed": 81
12
+ },
13
+ "direction": "Right",
14
+ "pad_to_multiple_of": null,
15
+ "pad_id": 0,
16
+ "pad_type_id": 0,
17
+ "pad_token": "[PAD]"
18
+ },
19
+ "added_tokens": [
20
+ {
21
+ "id": 0,
22
+ "content": "[PAD]",
23
+ "single_word": false,
24
+ "lstrip": false,
25
+ "rstrip": false,
26
+ "normalized": false,
27
+ "special": true
28
+ },
29
+ {
30
+ "id": 1,
31
+ "content": "[UNK]",
32
+ "single_word": false,
33
+ "lstrip": false,
34
+ "rstrip": false,
35
+ "normalized": false,
36
+ "special": true
37
+ },
38
+ {
39
+ "id": 2,
40
+ "content": "[CLS]",
41
+ "single_word": false,
42
+ "lstrip": false,
43
+ "rstrip": false,
44
+ "normalized": false,
45
+ "special": true
46
+ },
47
+ {
48
+ "id": 3,
49
+ "content": "[SEP]",
50
+ "single_word": false,
51
+ "lstrip": false,
52
+ "rstrip": false,
53
+ "normalized": false,
54
+ "special": true
55
+ },
56
+ {
57
+ "id": 4,
58
+ "content": "[MASK]",
59
+ "single_word": false,
60
+ "lstrip": false,
61
+ "rstrip": false,
62
+ "normalized": false,
63
+ "special": true
64
+ }
65
+ ],
66
+ "normalizer": {
67
+ "type": "BertNormalizer",
68
+ "clean_text": true,
69
+ "handle_chinese_chars": true,
70
+ "strip_accents": null,
71
+ "lowercase": false
72
+ },
73
+ "pre_tokenizer": {
74
+ "type": "BertPreTokenizer"
75
+ },
76
+ "post_processor": {
77
+ "type": "TemplateProcessing",
78
+ "single": [
79
+ {
80
+ "SpecialToken": {
81
+ "id": "[CLS]",
82
+ "type_id": 0
83
+ }
84
+ },
85
+ {
86
+ "Sequence": {
87
+ "id": "A",
88
+ "type_id": 0
89
+ }
90
+ },
91
+ {
92
+ "SpecialToken": {
93
+ "id": "[SEP]",
94
+ "type_id": 0
95
+ }
96
+ }
97
+ ],
98
+ "pair": [
99
+ {
100
+ "SpecialToken": {
101
+ "id": "[CLS]",
102
+ "type_id": 0
103
+ }
104
+ },
105
+ {
106
+ "Sequence": {
107
+ "id": "A",
108
+ "type_id": 0
109
+ }
110
+ },
111
+ {
112
+ "SpecialToken": {
113
+ "id": "[SEP]",
114
+ "type_id": 0
115
+ }
116
+ },
117
+ {
118
+ "Sequence": {
119
+ "id": "B",
120
+ "type_id": 1
121
+ }
122
+ },
123
+ {
124
+ "SpecialToken": {
125
+ "id": "[SEP]",
126
+ "type_id": 1
127
+ }
128
+ }
129
+ ],
130
+ "special_tokens": {
131
+ "[CLS]": {
132
+ "id": "[CLS]",
133
+ "ids": [
134
+ 2
135
+ ],
136
+ "tokens": [
137
+ "[CLS]"
138
+ ]
139
+ },
140
+ "[SEP]": {
141
+ "id": "[SEP]",
142
+ "ids": [
143
+ 3
144
+ ],
145
+ "tokens": [
146
+ "[SEP]"
147
+ ]
148
+ }
149
+ }
150
+ },
151
+ "decoder": {
152
+ "type": "WordPiece",
153
+ "prefix": "##",
154
+ "cleanup": true
155
+ },
156
+ "model": {
157
+ "type": "WordPiece",
158
+ "unk_token": "[UNK]",
159
+ "continuing_subword_prefix": "##",
160
+ "max_input_chars_per_word": 100,
161
+ "vocab": {
162
+ "[PAD]": 0,
163
+ "[UNK]": 1,
164
+ "[CLS]": 2,
165
+ "[SEP]": 3,
166
+ "[MASK]": 4,
167
+ "L": 5,
168
+ "A": 6,
169
+ "G": 7,
170
+ "V": 8,
171
+ "E": 9,
172
+ "S": 10,
173
+ "I": 11,
174
+ "K": 12,
175
+ "R": 13,
176
+ "D": 14,
177
+ "T": 15,
178
+ "P": 16,
179
+ "N": 17,
180
+ "Q": 18,
181
+ "F": 19,
182
+ "Y": 20,
183
+ "M": 21,
184
+ "H": 22,
185
+ "C": 23,
186
+ "W": 24,
187
+ "X": 25,
188
+ "U": 26,
189
+ "B": 27,
190
+ "Z": 28,
191
+ "O": 29
192
+ }
193
+ }
194
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "device": "cpu",
47
+ "do_basic_tokenize": true,
48
+ "do_lower_case": false,
49
+ "full_tokenizer_file": null,
50
+ "mask_token": "[MASK]",
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "never_split": null,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "strip_accents": null,
56
+ "tokenize_chinese_chars": true,
57
+ "tokenizer_class": "BertTokenizer",
58
+ "unk_token": "[UNK]"
59
+ }
trainer_state.json ADDED
@@ -0,0 +1,927 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 50.0,
5
+ "eval_steps": 10000,
6
+ "global_step": 126850,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 1.9999842333464724e-05,
14
+ "loss": 0.1488,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.39,
19
+ "learning_rate": 1.9842333464722114e-05,
20
+ "loss": 0.1737,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.79,
25
+ "learning_rate": 1.968498226251478e-05,
26
+ "loss": 0.1713,
27
+ "step": 2000
28
+ },
29
+ {
30
+ "epoch": 1.18,
31
+ "learning_rate": 1.9527315727236894e-05,
32
+ "loss": 0.1741,
33
+ "step": 3000
34
+ },
35
+ {
36
+ "epoch": 1.58,
37
+ "learning_rate": 1.9369806858494287e-05,
38
+ "loss": 0.1357,
39
+ "step": 4000
40
+ },
41
+ {
42
+ "epoch": 1.97,
43
+ "learning_rate": 1.92121403232164e-05,
44
+ "loss": 0.0985,
45
+ "step": 5000
46
+ },
47
+ {
48
+ "epoch": 2.36,
49
+ "learning_rate": 1.9054473787938513e-05,
50
+ "loss": 0.0746,
51
+ "step": 6000
52
+ },
53
+ {
54
+ "epoch": 2.76,
55
+ "learning_rate": 1.8897437918801735e-05,
56
+ "loss": 0.077,
57
+ "step": 7000
58
+ },
59
+ {
60
+ "epoch": 3.15,
61
+ "learning_rate": 1.8739771383523847e-05,
62
+ "loss": 0.0747,
63
+ "step": 8000
64
+ },
65
+ {
66
+ "epoch": 3.55,
67
+ "learning_rate": 1.858210484824596e-05,
68
+ "loss": 0.0607,
69
+ "step": 9000
70
+ },
71
+ {
72
+ "epoch": 3.94,
73
+ "learning_rate": 1.8424438312968072e-05,
74
+ "loss": 0.0648,
75
+ "step": 10000
76
+ },
77
+ {
78
+ "epoch": 3.94,
79
+ "eval_accuracy": 0.971966859607309,
80
+ "eval_f1": 0.8975528826213189,
81
+ "eval_loss": 0.047781217843294144,
82
+ "eval_precision": 0.9169491525423729,
83
+ "eval_recall": 0.8789601949634444,
84
+ "eval_runtime": 30.9282,
85
+ "eval_samples_per_second": 284.886,
86
+ "eval_steps_per_second": 35.631,
87
+ "step": 10000
88
+ },
89
+ {
90
+ "epoch": 4.34,
91
+ "learning_rate": 1.8266771777690185e-05,
92
+ "loss": 0.0604,
93
+ "step": 11000
94
+ },
95
+ {
96
+ "epoch": 4.73,
97
+ "learning_rate": 1.8109105242412298e-05,
98
+ "loss": 0.0587,
99
+ "step": 12000
100
+ },
101
+ {
102
+ "epoch": 5.12,
103
+ "learning_rate": 1.795159637366969e-05,
104
+ "loss": 0.0635,
105
+ "step": 13000
106
+ },
107
+ {
108
+ "epoch": 5.52,
109
+ "learning_rate": 1.779408750492708e-05,
110
+ "loss": 0.0558,
111
+ "step": 14000
112
+ },
113
+ {
114
+ "epoch": 5.91,
115
+ "learning_rate": 1.7636420969649194e-05,
116
+ "loss": 0.0525,
117
+ "step": 15000
118
+ },
119
+ {
120
+ "epoch": 6.31,
121
+ "learning_rate": 1.7478754434371306e-05,
122
+ "loss": 0.05,
123
+ "step": 16000
124
+ },
125
+ {
126
+ "epoch": 6.7,
127
+ "learning_rate": 1.7321245565628696e-05,
128
+ "loss": 0.0513,
129
+ "step": 17000
130
+ },
131
+ {
132
+ "epoch": 7.09,
133
+ "learning_rate": 1.716357903035081e-05,
134
+ "loss": 0.0423,
135
+ "step": 18000
136
+ },
137
+ {
138
+ "epoch": 7.49,
139
+ "learning_rate": 1.70060701616082e-05,
140
+ "loss": 0.0417,
141
+ "step": 19000
142
+ },
143
+ {
144
+ "epoch": 7.88,
145
+ "learning_rate": 1.684840362633031e-05,
146
+ "loss": 0.0432,
147
+ "step": 20000
148
+ },
149
+ {
150
+ "epoch": 7.88,
151
+ "eval_accuracy": 0.9800249687890137,
152
+ "eval_f1": 0.9293172690763052,
153
+ "eval_loss": 0.03126443922519684,
154
+ "eval_precision": 0.9189833200953137,
155
+ "eval_recall": 0.9398862713241267,
156
+ "eval_runtime": 30.5459,
157
+ "eval_samples_per_second": 288.451,
158
+ "eval_steps_per_second": 36.077,
159
+ "step": 20000
160
+ },
161
+ {
162
+ "epoch": 8.28,
163
+ "learning_rate": 1.6690737091052424e-05,
164
+ "loss": 0.0378,
165
+ "step": 21000
166
+ },
167
+ {
168
+ "epoch": 8.67,
169
+ "learning_rate": 1.6533070555774536e-05,
170
+ "loss": 0.041,
171
+ "step": 22000
172
+ },
173
+ {
174
+ "epoch": 9.07,
175
+ "learning_rate": 1.637540402049665e-05,
176
+ "loss": 0.0403,
177
+ "step": 23000
178
+ },
179
+ {
180
+ "epoch": 9.46,
181
+ "learning_rate": 1.621805281828932e-05,
182
+ "loss": 0.0338,
183
+ "step": 24000
184
+ },
185
+ {
186
+ "epoch": 9.85,
187
+ "learning_rate": 1.6060386283011432e-05,
188
+ "loss": 0.0352,
189
+ "step": 25000
190
+ },
191
+ {
192
+ "epoch": 10.25,
193
+ "learning_rate": 1.5902877414268822e-05,
194
+ "loss": 0.0356,
195
+ "step": 26000
196
+ },
197
+ {
198
+ "epoch": 10.64,
199
+ "learning_rate": 1.5745210878990935e-05,
200
+ "loss": 0.0343,
201
+ "step": 27000
202
+ },
203
+ {
204
+ "epoch": 11.04,
205
+ "learning_rate": 1.5587702010248328e-05,
206
+ "loss": 0.039,
207
+ "step": 28000
208
+ },
209
+ {
210
+ "epoch": 11.43,
211
+ "learning_rate": 1.543003547497044e-05,
212
+ "loss": 0.0336,
213
+ "step": 29000
214
+ },
215
+ {
216
+ "epoch": 11.82,
217
+ "learning_rate": 1.5272368939692553e-05,
218
+ "loss": 0.0294,
219
+ "step": 30000
220
+ },
221
+ {
222
+ "epoch": 11.82,
223
+ "eval_accuracy": 0.9878560889796845,
224
+ "eval_f1": 0.9563799429270281,
225
+ "eval_loss": 0.023840567097067833,
226
+ "eval_precision": 0.9599018003273322,
227
+ "eval_recall": 0.9528838342810723,
228
+ "eval_runtime": 29.3298,
229
+ "eval_samples_per_second": 300.411,
230
+ "eval_steps_per_second": 37.573,
231
+ "step": 30000
232
+ },
233
+ {
234
+ "epoch": 12.22,
235
+ "learning_rate": 1.5114702404414664e-05,
236
+ "loss": 0.0362,
237
+ "step": 31000
238
+ },
239
+ {
240
+ "epoch": 12.61,
241
+ "learning_rate": 1.4957035869136779e-05,
242
+ "loss": 0.0352,
243
+ "step": 32000
244
+ },
245
+ {
246
+ "epoch": 13.01,
247
+ "learning_rate": 1.4799369333858891e-05,
248
+ "loss": 0.0297,
249
+ "step": 33000
250
+ },
251
+ {
252
+ "epoch": 13.4,
253
+ "learning_rate": 1.464186046511628e-05,
254
+ "loss": 0.0316,
255
+ "step": 34000
256
+ },
257
+ {
258
+ "epoch": 13.8,
259
+ "learning_rate": 1.4484193929838392e-05,
260
+ "loss": 0.0312,
261
+ "step": 35000
262
+ },
263
+ {
264
+ "epoch": 14.19,
265
+ "learning_rate": 1.4326527394560506e-05,
266
+ "loss": 0.0315,
267
+ "step": 36000
268
+ },
269
+ {
270
+ "epoch": 14.58,
271
+ "learning_rate": 1.4168860859282619e-05,
272
+ "loss": 0.0275,
273
+ "step": 37000
274
+ },
275
+ {
276
+ "epoch": 14.98,
277
+ "learning_rate": 1.401135199054001e-05,
278
+ "loss": 0.0431,
279
+ "step": 38000
280
+ },
281
+ {
282
+ "epoch": 15.37,
283
+ "learning_rate": 1.3853843121797399e-05,
284
+ "loss": 0.034,
285
+ "step": 39000
286
+ },
287
+ {
288
+ "epoch": 15.77,
289
+ "learning_rate": 1.3696176586519511e-05,
290
+ "loss": 0.0275,
291
+ "step": 40000
292
+ },
293
+ {
294
+ "epoch": 15.77,
295
+ "eval_accuracy": 0.9868346385200317,
296
+ "eval_f1": 0.9522240527182867,
297
+ "eval_loss": 0.022756921127438545,
298
+ "eval_precision": 0.9657477025898078,
299
+ "eval_recall": 0.9390739236393176,
300
+ "eval_runtime": 29.4347,
301
+ "eval_samples_per_second": 299.34,
302
+ "eval_steps_per_second": 37.439,
303
+ "step": 40000
304
+ },
305
+ {
306
+ "epoch": 16.16,
307
+ "learning_rate": 1.3538510051241624e-05,
308
+ "loss": 0.0285,
309
+ "step": 41000
310
+ },
311
+ {
312
+ "epoch": 16.55,
313
+ "learning_rate": 1.3381001182499016e-05,
314
+ "loss": 0.0286,
315
+ "step": 42000
316
+ },
317
+ {
318
+ "epoch": 16.95,
319
+ "learning_rate": 1.3223334647221128e-05,
320
+ "loss": 0.0279,
321
+ "step": 43000
322
+ },
323
+ {
324
+ "epoch": 17.34,
325
+ "learning_rate": 1.306566811194324e-05,
326
+ "loss": 0.0254,
327
+ "step": 44000
328
+ },
329
+ {
330
+ "epoch": 17.74,
331
+ "learning_rate": 1.2908001576665355e-05,
332
+ "loss": 0.0239,
333
+ "step": 45000
334
+ },
335
+ {
336
+ "epoch": 18.13,
337
+ "learning_rate": 1.2750335041387468e-05,
338
+ "loss": 0.0269,
339
+ "step": 46000
340
+ },
341
+ {
342
+ "epoch": 18.53,
343
+ "learning_rate": 1.259266850610958e-05,
344
+ "loss": 0.0268,
345
+ "step": 47000
346
+ },
347
+ {
348
+ "epoch": 18.92,
349
+ "learning_rate": 1.2435001970831693e-05,
350
+ "loss": 0.0242,
351
+ "step": 48000
352
+ },
353
+ {
354
+ "epoch": 19.31,
355
+ "learning_rate": 1.2277493102089081e-05,
356
+ "loss": 0.0268,
357
+ "step": 49000
358
+ },
359
+ {
360
+ "epoch": 19.71,
361
+ "learning_rate": 1.2119984233346473e-05,
362
+ "loss": 0.0268,
363
+ "step": 50000
364
+ },
365
+ {
366
+ "epoch": 19.71,
367
+ "eval_accuracy": 0.9872886165020996,
368
+ "eval_f1": 0.9541734860883797,
369
+ "eval_loss": 0.017343735322356224,
370
+ "eval_precision": 0.9612530915086562,
371
+ "eval_recall": 0.9471974004874086,
372
+ "eval_runtime": 29.9772,
373
+ "eval_samples_per_second": 293.923,
374
+ "eval_steps_per_second": 36.761,
375
+ "step": 50000
376
+ },
377
+ {
378
+ "epoch": 20.1,
379
+ "learning_rate": 1.1962317698068587e-05,
380
+ "loss": 0.0238,
381
+ "step": 51000
382
+ },
383
+ {
384
+ "epoch": 20.5,
385
+ "learning_rate": 1.1804808829325975e-05,
386
+ "loss": 0.0279,
387
+ "step": 52000
388
+ },
389
+ {
390
+ "epoch": 20.89,
391
+ "learning_rate": 1.1647142294048088e-05,
392
+ "loss": 0.0251,
393
+ "step": 53000
394
+ },
395
+ {
396
+ "epoch": 21.28,
397
+ "learning_rate": 1.148963342530548e-05,
398
+ "loss": 0.0258,
399
+ "step": 54000
400
+ },
401
+ {
402
+ "epoch": 21.68,
403
+ "learning_rate": 1.1331966890027592e-05,
404
+ "loss": 0.0233,
405
+ "step": 55000
406
+ },
407
+ {
408
+ "epoch": 22.07,
409
+ "learning_rate": 1.1174458021284984e-05,
410
+ "loss": 0.0281,
411
+ "step": 56000
412
+ },
413
+ {
414
+ "epoch": 22.47,
415
+ "learning_rate": 1.1016949152542374e-05,
416
+ "loss": 0.0255,
417
+ "step": 57000
418
+ },
419
+ {
420
+ "epoch": 22.86,
421
+ "learning_rate": 1.0859282617264486e-05,
422
+ "loss": 0.0251,
423
+ "step": 58000
424
+ },
425
+ {
426
+ "epoch": 23.26,
427
+ "learning_rate": 1.0701616081986599e-05,
428
+ "loss": 0.0252,
429
+ "step": 59000
430
+ },
431
+ {
432
+ "epoch": 23.65,
433
+ "learning_rate": 1.0543949546708712e-05,
434
+ "loss": 0.0192,
435
+ "step": 60000
436
+ },
437
+ {
438
+ "epoch": 23.65,
439
+ "eval_accuracy": 0.9854727045738282,
440
+ "eval_f1": 0.9498825371965545,
441
+ "eval_loss": 0.01524350605905056,
442
+ "eval_precision": 0.9168556311413454,
443
+ "eval_recall": 0.9853777416734363,
444
+ "eval_runtime": 29.585,
445
+ "eval_samples_per_second": 297.82,
446
+ "eval_steps_per_second": 37.249,
447
+ "step": 60000
448
+ },
449
+ {
450
+ "epoch": 24.04,
451
+ "learning_rate": 1.0386440677966103e-05,
452
+ "loss": 0.0223,
453
+ "step": 61000
454
+ },
455
+ {
456
+ "epoch": 24.44,
457
+ "learning_rate": 1.0228774142688216e-05,
458
+ "loss": 0.0221,
459
+ "step": 62000
460
+ },
461
+ {
462
+ "epoch": 24.83,
463
+ "learning_rate": 1.0071107607410328e-05,
464
+ "loss": 0.0232,
465
+ "step": 63000
466
+ },
467
+ {
468
+ "epoch": 25.23,
469
+ "learning_rate": 9.913441072132441e-06,
470
+ "loss": 0.0215,
471
+ "step": 64000
472
+ },
473
+ {
474
+ "epoch": 25.62,
475
+ "learning_rate": 9.755774536854554e-06,
476
+ "loss": 0.0221,
477
+ "step": 65000
478
+ },
479
+ {
480
+ "epoch": 26.01,
481
+ "learning_rate": 9.598265668111943e-06,
482
+ "loss": 0.0216,
483
+ "step": 66000
484
+ },
485
+ {
486
+ "epoch": 26.41,
487
+ "learning_rate": 9.440599132834056e-06,
488
+ "loss": 0.0215,
489
+ "step": 67000
490
+ },
491
+ {
492
+ "epoch": 26.8,
493
+ "learning_rate": 9.282932597556169e-06,
494
+ "loss": 0.02,
495
+ "step": 68000
496
+ },
497
+ {
498
+ "epoch": 27.2,
499
+ "learning_rate": 9.12542372881356e-06,
500
+ "loss": 0.0227,
501
+ "step": 69000
502
+ },
503
+ {
504
+ "epoch": 27.59,
505
+ "learning_rate": 8.967757193535673e-06,
506
+ "loss": 0.018,
507
+ "step": 70000
508
+ },
509
+ {
510
+ "epoch": 27.59,
511
+ "eval_accuracy": 0.9919418908182953,
512
+ "eval_f1": 0.9716113554578168,
513
+ "eval_loss": 0.011301785707473755,
514
+ "eval_precision": 0.9566929133858267,
515
+ "eval_recall": 0.9870024370430545,
516
+ "eval_runtime": 29.5941,
517
+ "eval_samples_per_second": 297.728,
518
+ "eval_steps_per_second": 37.237,
519
+ "step": 70000
520
+ },
521
+ {
522
+ "epoch": 27.99,
523
+ "learning_rate": 8.810248324793063e-06,
524
+ "loss": 0.0211,
525
+ "step": 71000
526
+ },
527
+ {
528
+ "epoch": 28.38,
529
+ "learning_rate": 8.652581789515177e-06,
530
+ "loss": 0.0199,
531
+ "step": 72000
532
+ },
533
+ {
534
+ "epoch": 28.77,
535
+ "learning_rate": 8.49491525423729e-06,
536
+ "loss": 0.0197,
537
+ "step": 73000
538
+ },
539
+ {
540
+ "epoch": 29.17,
541
+ "learning_rate": 8.337248718959402e-06,
542
+ "loss": 0.0216,
543
+ "step": 74000
544
+ },
545
+ {
546
+ "epoch": 29.56,
547
+ "learning_rate": 8.17989751675207e-06,
548
+ "loss": 0.0177,
549
+ "step": 75000
550
+ },
551
+ {
552
+ "epoch": 29.96,
553
+ "learning_rate": 8.022230981474182e-06,
554
+ "loss": 0.0193,
555
+ "step": 76000
556
+ },
557
+ {
558
+ "epoch": 30.35,
559
+ "learning_rate": 7.864564446196295e-06,
560
+ "loss": 0.0167,
561
+ "step": 77000
562
+ },
563
+ {
564
+ "epoch": 30.74,
565
+ "learning_rate": 7.706897910918409e-06,
566
+ "loss": 0.0181,
567
+ "step": 78000
568
+ },
569
+ {
570
+ "epoch": 31.14,
571
+ "learning_rate": 7.549231375640521e-06,
572
+ "loss": 0.0194,
573
+ "step": 79000
574
+ },
575
+ {
576
+ "epoch": 31.53,
577
+ "learning_rate": 7.3915648403626335e-06,
578
+ "loss": 0.0155,
579
+ "step": 80000
580
+ },
581
+ {
582
+ "epoch": 31.53,
583
+ "eval_accuracy": 0.9923958688003632,
584
+ "eval_f1": 0.9729947601773479,
585
+ "eval_loss": 0.012764379382133484,
586
+ "eval_precision": 0.9656,
587
+ "eval_recall": 0.9805036555645816,
588
+ "eval_runtime": 29.6436,
589
+ "eval_samples_per_second": 297.231,
590
+ "eval_steps_per_second": 37.175,
591
+ "step": 80000
592
+ },
593
+ {
594
+ "epoch": 31.93,
595
+ "learning_rate": 7.233898305084747e-06,
596
+ "loss": 0.018,
597
+ "step": 81000
598
+ },
599
+ {
600
+ "epoch": 32.32,
601
+ "learning_rate": 7.076389436342137e-06,
602
+ "loss": 0.0184,
603
+ "step": 82000
604
+ },
605
+ {
606
+ "epoch": 32.72,
607
+ "learning_rate": 6.9187229010642495e-06,
608
+ "loss": 0.0169,
609
+ "step": 83000
610
+ },
611
+ {
612
+ "epoch": 33.11,
613
+ "learning_rate": 6.76121403232164e-06,
614
+ "loss": 0.0172,
615
+ "step": 84000
616
+ },
617
+ {
618
+ "epoch": 33.5,
619
+ "learning_rate": 6.603547497043753e-06,
620
+ "loss": 0.0191,
621
+ "step": 85000
622
+ },
623
+ {
624
+ "epoch": 33.9,
625
+ "learning_rate": 6.446038628301144e-06,
626
+ "loss": 0.0136,
627
+ "step": 86000
628
+ },
629
+ {
630
+ "epoch": 34.29,
631
+ "learning_rate": 6.288372093023256e-06,
632
+ "loss": 0.0173,
633
+ "step": 87000
634
+ },
635
+ {
636
+ "epoch": 34.69,
637
+ "learning_rate": 6.130705557745369e-06,
638
+ "loss": 0.0161,
639
+ "step": 88000
640
+ },
641
+ {
642
+ "epoch": 35.08,
643
+ "learning_rate": 5.97319668900276e-06,
644
+ "loss": 0.0143,
645
+ "step": 89000
646
+ },
647
+ {
648
+ "epoch": 35.47,
649
+ "learning_rate": 5.815530153724872e-06,
650
+ "loss": 0.0142,
651
+ "step": 90000
652
+ },
653
+ {
654
+ "epoch": 35.47,
655
+ "eval_accuracy": 0.9943252752241516,
656
+ "eval_f1": 0.9798549556809025,
657
+ "eval_loss": 0.00998597126454115,
658
+ "eval_precision": 0.9720223820943246,
659
+ "eval_recall": 0.9878147847278635,
660
+ "eval_runtime": 29.6518,
661
+ "eval_samples_per_second": 297.149,
662
+ "eval_steps_per_second": 37.165,
663
+ "step": 90000
664
+ },
665
+ {
666
+ "epoch": 35.87,
667
+ "learning_rate": 5.657863618446985e-06,
668
+ "loss": 0.0157,
669
+ "step": 91000
670
+ },
671
+ {
672
+ "epoch": 36.26,
673
+ "learning_rate": 5.500354749704376e-06,
674
+ "loss": 0.0164,
675
+ "step": 92000
676
+ },
677
+ {
678
+ "epoch": 36.66,
679
+ "learning_rate": 5.342688214426488e-06,
680
+ "loss": 0.0159,
681
+ "step": 93000
682
+ },
683
+ {
684
+ "epoch": 37.05,
685
+ "learning_rate": 5.18517934568388e-06,
686
+ "loss": 0.0134,
687
+ "step": 94000
688
+ },
689
+ {
690
+ "epoch": 37.45,
691
+ "learning_rate": 5.0275128104059924e-06,
692
+ "loss": 0.0148,
693
+ "step": 95000
694
+ },
695
+ {
696
+ "epoch": 37.84,
697
+ "learning_rate": 4.870003941663382e-06,
698
+ "loss": 0.0119,
699
+ "step": 96000
700
+ },
701
+ {
702
+ "epoch": 38.23,
703
+ "learning_rate": 4.712337406385495e-06,
704
+ "loss": 0.0116,
705
+ "step": 97000
706
+ },
707
+ {
708
+ "epoch": 38.63,
709
+ "learning_rate": 4.554670871107608e-06,
710
+ "loss": 0.0128,
711
+ "step": 98000
712
+ },
713
+ {
714
+ "epoch": 39.02,
715
+ "learning_rate": 4.397162002364998e-06,
716
+ "loss": 0.016,
717
+ "step": 99000
718
+ },
719
+ {
720
+ "epoch": 39.42,
721
+ "learning_rate": 4.239495467087111e-06,
722
+ "loss": 0.016,
723
+ "step": 100000
724
+ },
725
+ {
726
+ "epoch": 39.42,
727
+ "eval_accuracy": 0.9950062421972534,
728
+ "eval_f1": 0.9822437449556094,
729
+ "eval_loss": 0.009744029492139816,
730
+ "eval_precision": 0.9759422614274258,
731
+ "eval_recall": 0.9886271324126726,
732
+ "eval_runtime": 29.5778,
733
+ "eval_samples_per_second": 297.892,
734
+ "eval_steps_per_second": 37.258,
735
+ "step": 100000
736
+ },
737
+ {
738
+ "epoch": 39.81,
739
+ "learning_rate": 4.081986598344502e-06,
740
+ "loss": 0.0126,
741
+ "step": 101000
742
+ },
743
+ {
744
+ "epoch": 40.2,
745
+ "learning_rate": 3.924320063066614e-06,
746
+ "loss": 0.015,
747
+ "step": 102000
748
+ },
749
+ {
750
+ "epoch": 40.6,
751
+ "learning_rate": 3.7666535277887274e-06,
752
+ "loss": 0.0116,
753
+ "step": 103000
754
+ },
755
+ {
756
+ "epoch": 40.99,
757
+ "learning_rate": 3.60898699251084e-06,
758
+ "loss": 0.013,
759
+ "step": 104000
760
+ },
761
+ {
762
+ "epoch": 41.39,
763
+ "learning_rate": 3.4514781237682304e-06,
764
+ "loss": 0.0119,
765
+ "step": 105000
766
+ },
767
+ {
768
+ "epoch": 41.78,
769
+ "learning_rate": 3.2938115884903434e-06,
770
+ "loss": 0.0117,
771
+ "step": 106000
772
+ },
773
+ {
774
+ "epoch": 42.18,
775
+ "learning_rate": 3.136145053212456e-06,
776
+ "loss": 0.0107,
777
+ "step": 107000
778
+ },
779
+ {
780
+ "epoch": 42.57,
781
+ "learning_rate": 2.9786361844698463e-06,
782
+ "loss": 0.0103,
783
+ "step": 108000
784
+ },
785
+ {
786
+ "epoch": 42.96,
787
+ "learning_rate": 2.8209696491919594e-06,
788
+ "loss": 0.0137,
789
+ "step": 109000
790
+ },
791
+ {
792
+ "epoch": 43.36,
793
+ "learning_rate": 2.663303113914072e-06,
794
+ "loss": 0.0122,
795
+ "step": 110000
796
+ },
797
+ {
798
+ "epoch": 43.36,
799
+ "eval_accuracy": 0.9962546816479401,
800
+ "eval_f1": 0.9865689865689866,
801
+ "eval_loss": 0.010858706198632717,
802
+ "eval_precision": 0.9885807504078303,
803
+ "eval_recall": 0.9845653939886271,
804
+ "eval_runtime": 29.5587,
805
+ "eval_samples_per_second": 298.085,
806
+ "eval_steps_per_second": 37.282,
807
+ "step": 110000
808
+ },
809
+ {
810
+ "epoch": 43.75,
811
+ "learning_rate": 2.5056365786361845e-06,
812
+ "loss": 0.0104,
813
+ "step": 111000
814
+ },
815
+ {
816
+ "epoch": 44.15,
817
+ "learning_rate": 2.3481277098935753e-06,
818
+ "loss": 0.013,
819
+ "step": 112000
820
+ },
821
+ {
822
+ "epoch": 44.54,
823
+ "learning_rate": 2.190461174615688e-06,
824
+ "loss": 0.0095,
825
+ "step": 113000
826
+ },
827
+ {
828
+ "epoch": 44.93,
829
+ "learning_rate": 2.0329523058730783e-06,
830
+ "loss": 0.0114,
831
+ "step": 114000
832
+ },
833
+ {
834
+ "epoch": 45.33,
835
+ "learning_rate": 1.8752857705951913e-06,
836
+ "loss": 0.0116,
837
+ "step": 115000
838
+ },
839
+ {
840
+ "epoch": 45.72,
841
+ "learning_rate": 1.7176192353173041e-06,
842
+ "loss": 0.0084,
843
+ "step": 116000
844
+ },
845
+ {
846
+ "epoch": 46.12,
847
+ "learning_rate": 1.559952700039417e-06,
848
+ "loss": 0.0134,
849
+ "step": 117000
850
+ },
851
+ {
852
+ "epoch": 46.51,
853
+ "learning_rate": 1.4024438312968075e-06,
854
+ "loss": 0.0116,
855
+ "step": 118000
856
+ },
857
+ {
858
+ "epoch": 46.91,
859
+ "learning_rate": 1.2449349625541981e-06,
860
+ "loss": 0.0116,
861
+ "step": 119000
862
+ },
863
+ {
864
+ "epoch": 47.3,
865
+ "learning_rate": 1.0872684272763107e-06,
866
+ "loss": 0.0087,
867
+ "step": 120000
868
+ },
869
+ {
870
+ "epoch": 47.3,
871
+ "eval_accuracy": 0.9962546816479401,
872
+ "eval_f1": 0.9865799105327369,
873
+ "eval_loss": 0.010331220924854279,
874
+ "eval_precision": 0.987785016286645,
875
+ "eval_recall": 0.9853777416734363,
876
+ "eval_runtime": 29.6133,
877
+ "eval_samples_per_second": 297.536,
878
+ "eval_steps_per_second": 37.213,
879
+ "step": 120000
880
+ },
881
+ {
882
+ "epoch": 47.69,
883
+ "learning_rate": 9.296018919984234e-07,
884
+ "loss": 0.0127,
885
+ "step": 121000
886
+ },
887
+ {
888
+ "epoch": 48.09,
889
+ "learning_rate": 7.719353567205361e-07,
890
+ "loss": 0.0113,
891
+ "step": 122000
892
+ },
893
+ {
894
+ "epoch": 48.48,
895
+ "learning_rate": 6.144264879779267e-07,
896
+ "loss": 0.0103,
897
+ "step": 123000
898
+ },
899
+ {
900
+ "epoch": 48.88,
901
+ "learning_rate": 4.5675995270003944e-07,
902
+ "loss": 0.0078,
903
+ "step": 124000
904
+ },
905
+ {
906
+ "epoch": 49.27,
907
+ "learning_rate": 2.992510839574301e-07,
908
+ "loss": 0.0095,
909
+ "step": 125000
910
+ },
911
+ {
912
+ "epoch": 49.66,
913
+ "learning_rate": 1.4158454867954278e-07,
914
+ "loss": 0.0097,
915
+ "step": 126000
916
+ }
917
+ ],
918
+ "logging_steps": 1000,
919
+ "max_steps": 126850,
920
+ "num_input_tokens_seen": 0,
921
+ "num_train_epochs": 50,
922
+ "save_steps": 500,
923
+ "total_flos": 0.0,
924
+ "train_batch_size": 8,
925
+ "trial_name": null,
926
+ "trial_params": null
927
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95f7a46484b05353261ca1fa263ffdcabe89248c1ffb008a828acc121717d589
3
+ size 4600
vocab.txt ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [PAD]
2
+ [UNK]
3
+ [CLS]
4
+ [SEP]
5
+ [MASK]
6
+ L
7
+ A
8
+ G
9
+ V
10
+ E
11
+ S
12
+ I
13
+ K
14
+ R
15
+ D
16
+ T
17
+ P
18
+ N
19
+ Q
20
+ F
21
+ Y
22
+ M
23
+ H
24
+ C
25
+ W
26
+ X
27
+ U
28
+ B
29
+ Z
30
+ O