MihaiPopa-1 commited on
Commit
ade96c4
·
verified ·
1 Parent(s): f5f1fed

Upload folder using huggingface_hub

Browse files
checkpoint-11655/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5820684ee0afadbefabb638ef5155c4da42f2bcd415ff8dc4b79dffc620da158
3
+ size 70805128
checkpoint-11655/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52b90785d8eaf11e81b2b471428ebb721a807db2f2d6b916c7dac2c7cb45f683
3
+ size 133229854
checkpoint-11655/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01f9a0f7843a37be87edd23f4e88aa93b38b95cc2c07503eeb1cf2e4632453a2
3
+ size 14645
checkpoint-11655/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9c8ce44acdc013cd8111cf838af12f5995d8224dff455d312b5628f999db660
3
+ size 1465
checkpoint-11655/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-11655/tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": true,
7
+ "eos_token": "[SEP]",
8
+ "is_local": false,
9
+ "keep_accents": false,
10
+ "mask_token": "[MASK]",
11
+ "model_max_length": 512,
12
+ "pad_token": "<pad>",
13
+ "sep_token": "[SEP]",
14
+ "tokenizer_class": "AlbertTokenizer",
15
+ "trim_offsets": true,
16
+ "unk_token": "<unk>"
17
+ }
checkpoint-11655/trainer_state.json ADDED
@@ -0,0 +1,228 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 11655,
3
+ "best_metric": 0.06289209425449371,
4
+ "best_model_checkpoint": "./results_albert_punctuation_casing/checkpoint-11655",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 11655,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.1287001287001287,
14
+ "grad_norm": 4.2918195724487305,
15
+ "learning_rate": 1.9143715143715144e-05,
16
+ "loss": 0.2105482635498047,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.2574002574002574,
21
+ "grad_norm": 3.180814504623413,
22
+ "learning_rate": 1.8285714285714288e-05,
23
+ "loss": 0.09811102294921875,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.3861003861003861,
28
+ "grad_norm": 4.069736480712891,
29
+ "learning_rate": 1.742771342771343e-05,
30
+ "loss": 0.08830223083496094,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 0.5148005148005148,
35
+ "grad_norm": 2.3238089084625244,
36
+ "learning_rate": 1.656971256971257e-05,
37
+ "loss": 0.07262681579589844,
38
+ "step": 2000
39
+ },
40
+ {
41
+ "epoch": 0.6435006435006435,
42
+ "grad_norm": 0.725167989730835,
43
+ "learning_rate": 1.5711711711711713e-05,
44
+ "loss": 0.07961682891845703,
45
+ "step": 2500
46
+ },
47
+ {
48
+ "epoch": 0.7722007722007722,
49
+ "grad_norm": 13.671622276306152,
50
+ "learning_rate": 1.4853710853710854e-05,
51
+ "loss": 0.06993846893310547,
52
+ "step": 3000
53
+ },
54
+ {
55
+ "epoch": 0.9009009009009009,
56
+ "grad_norm": 2.6098945140838623,
57
+ "learning_rate": 1.3995709995709996e-05,
58
+ "loss": 0.07217549133300781,
59
+ "step": 3500
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_casing_accuracy": 0.6387912059001499,
64
+ "eval_loss": 0.07048454880714417,
65
+ "eval_overall_accuracy": 0.6404223412931571,
66
+ "eval_punctuation_accuracy": 0.6420534766861643,
67
+ "eval_runtime": 86.2448,
68
+ "eval_samples_per_second": 180.162,
69
+ "eval_steps_per_second": 11.27,
70
+ "step": 3885
71
+ },
72
+ {
73
+ "epoch": 1.0296010296010296,
74
+ "grad_norm": 1.678989052772522,
75
+ "learning_rate": 1.3137709137709139e-05,
76
+ "loss": 0.05899927520751953,
77
+ "step": 4000
78
+ },
79
+ {
80
+ "epoch": 1.1583011583011582,
81
+ "grad_norm": 5.855215549468994,
82
+ "learning_rate": 1.2279708279708281e-05,
83
+ "loss": 0.05248377227783203,
84
+ "step": 4500
85
+ },
86
+ {
87
+ "epoch": 1.287001287001287,
88
+ "grad_norm": 20.59808921813965,
89
+ "learning_rate": 1.1421707421707422e-05,
90
+ "loss": 0.05537939834594727,
91
+ "step": 5000
92
+ },
93
+ {
94
+ "epoch": 1.4157014157014158,
95
+ "grad_norm": 3.922346830368042,
96
+ "learning_rate": 1.0563706563706564e-05,
97
+ "loss": 0.05087580490112305,
98
+ "step": 5500
99
+ },
100
+ {
101
+ "epoch": 1.5444015444015444,
102
+ "grad_norm": 0.129458948969841,
103
+ "learning_rate": 9.705705705705706e-06,
104
+ "loss": 0.0524902229309082,
105
+ "step": 6000
106
+ },
107
+ {
108
+ "epoch": 1.673101673101673,
109
+ "grad_norm": 0.10066387057304382,
110
+ "learning_rate": 8.847704847704849e-06,
111
+ "loss": 0.04880419921875,
112
+ "step": 6500
113
+ },
114
+ {
115
+ "epoch": 1.8018018018018018,
116
+ "grad_norm": 1.1645872592926025,
117
+ "learning_rate": 7.989703989703991e-06,
118
+ "loss": 0.04735799407958984,
119
+ "step": 7000
120
+ },
121
+ {
122
+ "epoch": 1.9305019305019306,
123
+ "grad_norm": 1.6507762670516968,
124
+ "learning_rate": 7.1317031317031325e-06,
125
+ "loss": 0.05284581756591797,
126
+ "step": 7500
127
+ },
128
+ {
129
+ "epoch": 2.0,
130
+ "eval_casing_accuracy": 0.6404749585638992,
131
+ "eval_loss": 0.06381073594093323,
132
+ "eval_overall_accuracy": 0.6414089151195728,
133
+ "eval_punctuation_accuracy": 0.6423428716752462,
134
+ "eval_runtime": 85.5043,
135
+ "eval_samples_per_second": 181.722,
136
+ "eval_steps_per_second": 11.368,
137
+ "step": 7770
138
+ },
139
+ {
140
+ "epoch": 2.0592020592020592,
141
+ "grad_norm": 0.3554779887199402,
142
+ "learning_rate": 6.273702273702275e-06,
143
+ "loss": 0.042236793518066404,
144
+ "step": 8000
145
+ },
146
+ {
147
+ "epoch": 2.187902187902188,
148
+ "grad_norm": 0.3045165240764618,
149
+ "learning_rate": 5.415701415701416e-06,
150
+ "loss": 0.03922730255126953,
151
+ "step": 8500
152
+ },
153
+ {
154
+ "epoch": 2.3166023166023164,
155
+ "grad_norm": 1.6675119400024414,
156
+ "learning_rate": 4.557700557700558e-06,
157
+ "loss": 0.034516990661621096,
158
+ "step": 9000
159
+ },
160
+ {
161
+ "epoch": 2.4453024453024454,
162
+ "grad_norm": 1.5336593389511108,
163
+ "learning_rate": 3.6996996996997e-06,
164
+ "loss": 0.03220732116699219,
165
+ "step": 9500
166
+ },
167
+ {
168
+ "epoch": 2.574002574002574,
169
+ "grad_norm": 2.9998581409454346,
170
+ "learning_rate": 2.8416988416988417e-06,
171
+ "loss": 0.033812404632568356,
172
+ "step": 10000
173
+ },
174
+ {
175
+ "epoch": 2.7027027027027026,
176
+ "grad_norm": 3.3144118785858154,
177
+ "learning_rate": 1.9836979836979837e-06,
178
+ "loss": 0.03330759048461914,
179
+ "step": 10500
180
+ },
181
+ {
182
+ "epoch": 2.8314028314028317,
183
+ "grad_norm": 2.0357117652893066,
184
+ "learning_rate": 1.1256971256971258e-06,
185
+ "loss": 0.03533472442626953,
186
+ "step": 11000
187
+ },
188
+ {
189
+ "epoch": 2.9601029601029603,
190
+ "grad_norm": 3.5177860260009766,
191
+ "learning_rate": 2.676962676962677e-07,
192
+ "loss": 0.0314073543548584,
193
+ "step": 11500
194
+ },
195
+ {
196
+ "epoch": 3.0,
197
+ "eval_casing_accuracy": 0.6400978681235804,
198
+ "eval_loss": 0.06289209425449371,
199
+ "eval_overall_accuracy": 0.6402776437986162,
200
+ "eval_punctuation_accuracy": 0.6404574194736519,
201
+ "eval_runtime": 85.5898,
202
+ "eval_samples_per_second": 181.54,
203
+ "eval_steps_per_second": 11.356,
204
+ "step": 11655
205
+ }
206
+ ],
207
+ "logging_steps": 500,
208
+ "max_steps": 11655,
209
+ "num_input_tokens_seen": 0,
210
+ "num_train_epochs": 3,
211
+ "save_steps": 500,
212
+ "stateful_callbacks": {
213
+ "TrainerControl": {
214
+ "args": {
215
+ "should_epoch_stop": false,
216
+ "should_evaluate": false,
217
+ "should_log": false,
218
+ "should_save": true,
219
+ "should_training_stop": true
220
+ },
221
+ "attributes": {}
222
+ }
223
+ },
224
+ "total_flos": 0.0,
225
+ "train_batch_size": 16,
226
+ "trial_name": null,
227
+ "trial_params": null
228
+ }
checkpoint-11655/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:564d1f2f5aef116f928ab806870d35d30703b30b96f0b6e8ba67019dc262c956
3
+ size 5265
checkpoint-3885/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e8e5e65496ae4e5fee4fe44f3610acc0fec320770fc8da33905cad376e5a0b8
3
+ size 70805128
checkpoint-3885/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce66ebb17456124ac35ea5024805667b5bafcfa0fd8521b60b6f83be5fff1d63
3
+ size 133229854
checkpoint-3885/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
3
+ size 14645
checkpoint-3885/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:576ffb7acdc7f45096a1f62e0c165ae53e15f334d7c72feaf2e0e0834f61b6c9
3
+ size 1465
checkpoint-3885/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3885/tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": true,
7
+ "eos_token": "[SEP]",
8
+ "is_local": false,
9
+ "keep_accents": false,
10
+ "mask_token": "[MASK]",
11
+ "model_max_length": 512,
12
+ "pad_token": "<pad>",
13
+ "sep_token": "[SEP]",
14
+ "tokenizer_class": "AlbertTokenizer",
15
+ "trim_offsets": true,
16
+ "unk_token": "<unk>"
17
+ }
checkpoint-3885/trainer_state.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 3885,
3
+ "best_metric": 0.07048454880714417,
4
+ "best_model_checkpoint": "./results_albert_punctuation_casing/checkpoint-3885",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 3885,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.1287001287001287,
14
+ "grad_norm": 4.2918195724487305,
15
+ "learning_rate": 1.9143715143715144e-05,
16
+ "loss": 0.2105482635498047,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.2574002574002574,
21
+ "grad_norm": 3.180814504623413,
22
+ "learning_rate": 1.8285714285714288e-05,
23
+ "loss": 0.09811102294921875,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.3861003861003861,
28
+ "grad_norm": 4.069736480712891,
29
+ "learning_rate": 1.742771342771343e-05,
30
+ "loss": 0.08830223083496094,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 0.5148005148005148,
35
+ "grad_norm": 2.3238089084625244,
36
+ "learning_rate": 1.656971256971257e-05,
37
+ "loss": 0.07262681579589844,
38
+ "step": 2000
39
+ },
40
+ {
41
+ "epoch": 0.6435006435006435,
42
+ "grad_norm": 0.725167989730835,
43
+ "learning_rate": 1.5711711711711713e-05,
44
+ "loss": 0.07961682891845703,
45
+ "step": 2500
46
+ },
47
+ {
48
+ "epoch": 0.7722007722007722,
49
+ "grad_norm": 13.671622276306152,
50
+ "learning_rate": 1.4853710853710854e-05,
51
+ "loss": 0.06993846893310547,
52
+ "step": 3000
53
+ },
54
+ {
55
+ "epoch": 0.9009009009009009,
56
+ "grad_norm": 2.6098945140838623,
57
+ "learning_rate": 1.3995709995709996e-05,
58
+ "loss": 0.07217549133300781,
59
+ "step": 3500
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_casing_accuracy": 0.6387912059001499,
64
+ "eval_loss": 0.07048454880714417,
65
+ "eval_overall_accuracy": 0.6404223412931571,
66
+ "eval_punctuation_accuracy": 0.6420534766861643,
67
+ "eval_runtime": 86.2448,
68
+ "eval_samples_per_second": 180.162,
69
+ "eval_steps_per_second": 11.27,
70
+ "step": 3885
71
+ }
72
+ ],
73
+ "logging_steps": 500,
74
+ "max_steps": 11655,
75
+ "num_input_tokens_seen": 0,
76
+ "num_train_epochs": 3,
77
+ "save_steps": 500,
78
+ "stateful_callbacks": {
79
+ "TrainerControl": {
80
+ "args": {
81
+ "should_epoch_stop": false,
82
+ "should_evaluate": false,
83
+ "should_log": false,
84
+ "should_save": true,
85
+ "should_training_stop": false
86
+ },
87
+ "attributes": {}
88
+ }
89
+ },
90
+ "total_flos": 0.0,
91
+ "train_batch_size": 16,
92
+ "trial_name": null,
93
+ "trial_params": null
94
+ }
checkpoint-3885/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:564d1f2f5aef116f928ab806870d35d30703b30b96f0b6e8ba67019dc262c956
3
+ size 5265
checkpoint-7770/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9c31ca427b991dd313016a5cec86bd856d365cffdd0962c0297e9e9e28e50b5
3
+ size 70805128
checkpoint-7770/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d10667b9589411bddaef2eb83b702769b132972329cb8a05d216f294ad91489e
3
+ size 133229854
checkpoint-7770/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8e2011629d8bed3ef560fa11175cac55684c4e12a72634bb24abf767b6c7399
3
+ size 14645
checkpoint-7770/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae4867f43b40f4e5c011cbc8c78aa96d2e1218862b1b43e9a91c59a25eec78e9
3
+ size 1465
checkpoint-7770/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-7770/tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": true,
7
+ "eos_token": "[SEP]",
8
+ "is_local": false,
9
+ "keep_accents": false,
10
+ "mask_token": "[MASK]",
11
+ "model_max_length": 512,
12
+ "pad_token": "<pad>",
13
+ "sep_token": "[SEP]",
14
+ "tokenizer_class": "AlbertTokenizer",
15
+ "trim_offsets": true,
16
+ "unk_token": "<unk>"
17
+ }
checkpoint-7770/trainer_state.json ADDED
@@ -0,0 +1,161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 7770,
3
+ "best_metric": 0.06381073594093323,
4
+ "best_model_checkpoint": "./results_albert_punctuation_casing/checkpoint-7770",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 7770,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.1287001287001287,
14
+ "grad_norm": 4.2918195724487305,
15
+ "learning_rate": 1.9143715143715144e-05,
16
+ "loss": 0.2105482635498047,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.2574002574002574,
21
+ "grad_norm": 3.180814504623413,
22
+ "learning_rate": 1.8285714285714288e-05,
23
+ "loss": 0.09811102294921875,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.3861003861003861,
28
+ "grad_norm": 4.069736480712891,
29
+ "learning_rate": 1.742771342771343e-05,
30
+ "loss": 0.08830223083496094,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 0.5148005148005148,
35
+ "grad_norm": 2.3238089084625244,
36
+ "learning_rate": 1.656971256971257e-05,
37
+ "loss": 0.07262681579589844,
38
+ "step": 2000
39
+ },
40
+ {
41
+ "epoch": 0.6435006435006435,
42
+ "grad_norm": 0.725167989730835,
43
+ "learning_rate": 1.5711711711711713e-05,
44
+ "loss": 0.07961682891845703,
45
+ "step": 2500
46
+ },
47
+ {
48
+ "epoch": 0.7722007722007722,
49
+ "grad_norm": 13.671622276306152,
50
+ "learning_rate": 1.4853710853710854e-05,
51
+ "loss": 0.06993846893310547,
52
+ "step": 3000
53
+ },
54
+ {
55
+ "epoch": 0.9009009009009009,
56
+ "grad_norm": 2.6098945140838623,
57
+ "learning_rate": 1.3995709995709996e-05,
58
+ "loss": 0.07217549133300781,
59
+ "step": 3500
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_casing_accuracy": 0.6387912059001499,
64
+ "eval_loss": 0.07048454880714417,
65
+ "eval_overall_accuracy": 0.6404223412931571,
66
+ "eval_punctuation_accuracy": 0.6420534766861643,
67
+ "eval_runtime": 86.2448,
68
+ "eval_samples_per_second": 180.162,
69
+ "eval_steps_per_second": 11.27,
70
+ "step": 3885
71
+ },
72
+ {
73
+ "epoch": 1.0296010296010296,
74
+ "grad_norm": 1.678989052772522,
75
+ "learning_rate": 1.3137709137709139e-05,
76
+ "loss": 0.05899927520751953,
77
+ "step": 4000
78
+ },
79
+ {
80
+ "epoch": 1.1583011583011582,
81
+ "grad_norm": 5.855215549468994,
82
+ "learning_rate": 1.2279708279708281e-05,
83
+ "loss": 0.05248377227783203,
84
+ "step": 4500
85
+ },
86
+ {
87
+ "epoch": 1.287001287001287,
88
+ "grad_norm": 20.59808921813965,
89
+ "learning_rate": 1.1421707421707422e-05,
90
+ "loss": 0.05537939834594727,
91
+ "step": 5000
92
+ },
93
+ {
94
+ "epoch": 1.4157014157014158,
95
+ "grad_norm": 3.922346830368042,
96
+ "learning_rate": 1.0563706563706564e-05,
97
+ "loss": 0.05087580490112305,
98
+ "step": 5500
99
+ },
100
+ {
101
+ "epoch": 1.5444015444015444,
102
+ "grad_norm": 0.129458948969841,
103
+ "learning_rate": 9.705705705705706e-06,
104
+ "loss": 0.0524902229309082,
105
+ "step": 6000
106
+ },
107
+ {
108
+ "epoch": 1.673101673101673,
109
+ "grad_norm": 0.10066387057304382,
110
+ "learning_rate": 8.847704847704849e-06,
111
+ "loss": 0.04880419921875,
112
+ "step": 6500
113
+ },
114
+ {
115
+ "epoch": 1.8018018018018018,
116
+ "grad_norm": 1.1645872592926025,
117
+ "learning_rate": 7.989703989703991e-06,
118
+ "loss": 0.04735799407958984,
119
+ "step": 7000
120
+ },
121
+ {
122
+ "epoch": 1.9305019305019306,
123
+ "grad_norm": 1.6507762670516968,
124
+ "learning_rate": 7.1317031317031325e-06,
125
+ "loss": 0.05284581756591797,
126
+ "step": 7500
127
+ },
128
+ {
129
+ "epoch": 2.0,
130
+ "eval_casing_accuracy": 0.6404749585638992,
131
+ "eval_loss": 0.06381073594093323,
132
+ "eval_overall_accuracy": 0.6414089151195728,
133
+ "eval_punctuation_accuracy": 0.6423428716752462,
134
+ "eval_runtime": 85.5043,
135
+ "eval_samples_per_second": 181.722,
136
+ "eval_steps_per_second": 11.368,
137
+ "step": 7770
138
+ }
139
+ ],
140
+ "logging_steps": 500,
141
+ "max_steps": 11655,
142
+ "num_input_tokens_seen": 0,
143
+ "num_train_epochs": 3,
144
+ "save_steps": 500,
145
+ "stateful_callbacks": {
146
+ "TrainerControl": {
147
+ "args": {
148
+ "should_epoch_stop": false,
149
+ "should_evaluate": false,
150
+ "should_log": false,
151
+ "should_save": true,
152
+ "should_training_stop": false
153
+ },
154
+ "attributes": {}
155
+ }
156
+ },
157
+ "total_flos": 0.0,
158
+ "train_batch_size": 16,
159
+ "trial_name": null,
160
+ "trial_params": null
161
+ }
checkpoint-7770/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:564d1f2f5aef116f928ab806870d35d30703b30b96f0b6e8ba67019dc262c956
3
+ size 5265
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5820684ee0afadbefabb638ef5155c4da42f2bcd415ff8dc4b79dffc620da158
3
+ size 70805128
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52b90785d8eaf11e81b2b471428ebb721a807db2f2d6b916c7dac2c7cb45f683
3
+ size 133229854
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01f9a0f7843a37be87edd23f4e88aa93b38b95cc2c07503eeb1cf2e4632453a2
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9c8ce44acdc013cd8111cf838af12f5995d8224dff455d312b5628f999db660
3
+ size 1465
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": true,
7
+ "eos_token": "[SEP]",
8
+ "is_local": false,
9
+ "keep_accents": false,
10
+ "mask_token": "[MASK]",
11
+ "model_max_length": 512,
12
+ "pad_token": "<pad>",
13
+ "sep_token": "[SEP]",
14
+ "tokenizer_class": "AlbertTokenizer",
15
+ "trim_offsets": true,
16
+ "unk_token": "<unk>"
17
+ }
trainer_state.json ADDED
@@ -0,0 +1,228 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 11655,
3
+ "best_metric": 0.06289209425449371,
4
+ "best_model_checkpoint": "./results_albert_punctuation_casing/checkpoint-11655",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 11655,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.1287001287001287,
14
+ "grad_norm": 4.2918195724487305,
15
+ "learning_rate": 1.9143715143715144e-05,
16
+ "loss": 0.2105482635498047,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.2574002574002574,
21
+ "grad_norm": 3.180814504623413,
22
+ "learning_rate": 1.8285714285714288e-05,
23
+ "loss": 0.09811102294921875,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.3861003861003861,
28
+ "grad_norm": 4.069736480712891,
29
+ "learning_rate": 1.742771342771343e-05,
30
+ "loss": 0.08830223083496094,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 0.5148005148005148,
35
+ "grad_norm": 2.3238089084625244,
36
+ "learning_rate": 1.656971256971257e-05,
37
+ "loss": 0.07262681579589844,
38
+ "step": 2000
39
+ },
40
+ {
41
+ "epoch": 0.6435006435006435,
42
+ "grad_norm": 0.725167989730835,
43
+ "learning_rate": 1.5711711711711713e-05,
44
+ "loss": 0.07961682891845703,
45
+ "step": 2500
46
+ },
47
+ {
48
+ "epoch": 0.7722007722007722,
49
+ "grad_norm": 13.671622276306152,
50
+ "learning_rate": 1.4853710853710854e-05,
51
+ "loss": 0.06993846893310547,
52
+ "step": 3000
53
+ },
54
+ {
55
+ "epoch": 0.9009009009009009,
56
+ "grad_norm": 2.6098945140838623,
57
+ "learning_rate": 1.3995709995709996e-05,
58
+ "loss": 0.07217549133300781,
59
+ "step": 3500
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "eval_casing_accuracy": 0.6387912059001499,
64
+ "eval_loss": 0.07048454880714417,
65
+ "eval_overall_accuracy": 0.6404223412931571,
66
+ "eval_punctuation_accuracy": 0.6420534766861643,
67
+ "eval_runtime": 86.2448,
68
+ "eval_samples_per_second": 180.162,
69
+ "eval_steps_per_second": 11.27,
70
+ "step": 3885
71
+ },
72
+ {
73
+ "epoch": 1.0296010296010296,
74
+ "grad_norm": 1.678989052772522,
75
+ "learning_rate": 1.3137709137709139e-05,
76
+ "loss": 0.05899927520751953,
77
+ "step": 4000
78
+ },
79
+ {
80
+ "epoch": 1.1583011583011582,
81
+ "grad_norm": 5.855215549468994,
82
+ "learning_rate": 1.2279708279708281e-05,
83
+ "loss": 0.05248377227783203,
84
+ "step": 4500
85
+ },
86
+ {
87
+ "epoch": 1.287001287001287,
88
+ "grad_norm": 20.59808921813965,
89
+ "learning_rate": 1.1421707421707422e-05,
90
+ "loss": 0.05537939834594727,
91
+ "step": 5000
92
+ },
93
+ {
94
+ "epoch": 1.4157014157014158,
95
+ "grad_norm": 3.922346830368042,
96
+ "learning_rate": 1.0563706563706564e-05,
97
+ "loss": 0.05087580490112305,
98
+ "step": 5500
99
+ },
100
+ {
101
+ "epoch": 1.5444015444015444,
102
+ "grad_norm": 0.129458948969841,
103
+ "learning_rate": 9.705705705705706e-06,
104
+ "loss": 0.0524902229309082,
105
+ "step": 6000
106
+ },
107
+ {
108
+ "epoch": 1.673101673101673,
109
+ "grad_norm": 0.10066387057304382,
110
+ "learning_rate": 8.847704847704849e-06,
111
+ "loss": 0.04880419921875,
112
+ "step": 6500
113
+ },
114
+ {
115
+ "epoch": 1.8018018018018018,
116
+ "grad_norm": 1.1645872592926025,
117
+ "learning_rate": 7.989703989703991e-06,
118
+ "loss": 0.04735799407958984,
119
+ "step": 7000
120
+ },
121
+ {
122
+ "epoch": 1.9305019305019306,
123
+ "grad_norm": 1.6507762670516968,
124
+ "learning_rate": 7.1317031317031325e-06,
125
+ "loss": 0.05284581756591797,
126
+ "step": 7500
127
+ },
128
+ {
129
+ "epoch": 2.0,
130
+ "eval_casing_accuracy": 0.6404749585638992,
131
+ "eval_loss": 0.06381073594093323,
132
+ "eval_overall_accuracy": 0.6414089151195728,
133
+ "eval_punctuation_accuracy": 0.6423428716752462,
134
+ "eval_runtime": 85.5043,
135
+ "eval_samples_per_second": 181.722,
136
+ "eval_steps_per_second": 11.368,
137
+ "step": 7770
138
+ },
139
+ {
140
+ "epoch": 2.0592020592020592,
141
+ "grad_norm": 0.3554779887199402,
142
+ "learning_rate": 6.273702273702275e-06,
143
+ "loss": 0.042236793518066404,
144
+ "step": 8000
145
+ },
146
+ {
147
+ "epoch": 2.187902187902188,
148
+ "grad_norm": 0.3045165240764618,
149
+ "learning_rate": 5.415701415701416e-06,
150
+ "loss": 0.03922730255126953,
151
+ "step": 8500
152
+ },
153
+ {
154
+ "epoch": 2.3166023166023164,
155
+ "grad_norm": 1.6675119400024414,
156
+ "learning_rate": 4.557700557700558e-06,
157
+ "loss": 0.034516990661621096,
158
+ "step": 9000
159
+ },
160
+ {
161
+ "epoch": 2.4453024453024454,
162
+ "grad_norm": 1.5336593389511108,
163
+ "learning_rate": 3.6996996996997e-06,
164
+ "loss": 0.03220732116699219,
165
+ "step": 9500
166
+ },
167
+ {
168
+ "epoch": 2.574002574002574,
169
+ "grad_norm": 2.9998581409454346,
170
+ "learning_rate": 2.8416988416988417e-06,
171
+ "loss": 0.033812404632568356,
172
+ "step": 10000
173
+ },
174
+ {
175
+ "epoch": 2.7027027027027026,
176
+ "grad_norm": 3.3144118785858154,
177
+ "learning_rate": 1.9836979836979837e-06,
178
+ "loss": 0.03330759048461914,
179
+ "step": 10500
180
+ },
181
+ {
182
+ "epoch": 2.8314028314028317,
183
+ "grad_norm": 2.0357117652893066,
184
+ "learning_rate": 1.1256971256971258e-06,
185
+ "loss": 0.03533472442626953,
186
+ "step": 11000
187
+ },
188
+ {
189
+ "epoch": 2.9601029601029603,
190
+ "grad_norm": 3.5177860260009766,
191
+ "learning_rate": 2.676962676962677e-07,
192
+ "loss": 0.0314073543548584,
193
+ "step": 11500
194
+ },
195
+ {
196
+ "epoch": 3.0,
197
+ "eval_casing_accuracy": 0.6400978681235804,
198
+ "eval_loss": 0.06289209425449371,
199
+ "eval_overall_accuracy": 0.6402776437986162,
200
+ "eval_punctuation_accuracy": 0.6404574194736519,
201
+ "eval_runtime": 85.5898,
202
+ "eval_samples_per_second": 181.54,
203
+ "eval_steps_per_second": 11.356,
204
+ "step": 11655
205
+ }
206
+ ],
207
+ "logging_steps": 500,
208
+ "max_steps": 11655,
209
+ "num_input_tokens_seen": 0,
210
+ "num_train_epochs": 3,
211
+ "save_steps": 500,
212
+ "stateful_callbacks": {
213
+ "TrainerControl": {
214
+ "args": {
215
+ "should_epoch_stop": false,
216
+ "should_evaluate": false,
217
+ "should_log": false,
218
+ "should_save": true,
219
+ "should_training_stop": true
220
+ },
221
+ "attributes": {}
222
+ }
223
+ },
224
+ "total_flos": 0.0,
225
+ "train_batch_size": 16,
226
+ "trial_name": null,
227
+ "trial_params": null
228
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:564d1f2f5aef116f928ab806870d35d30703b30b96f0b6e8ba67019dc262c956
3
+ size 5265