File size: 6,249 Bytes
f6751f7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
{
  "best_metric": 0.4608415365219116,
  "best_model_checkpoint": "./Vit-GPT2-COCO2017Flickr-85k-11/checkpoint-4500",
  "epoch": 0.9330969487729776,
  "eval_steps": 500,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09330969487729775,
      "grad_norm": 1.14389169216156,
      "learning_rate": 4.8444693293517484e-05,
      "loss": 0.378,
      "step": 500
    },
    {
      "epoch": 0.09330969487729775,
      "eval_gen_len": 11.7725,
      "eval_loss": 0.469292551279068,
      "eval_rouge1": 40.2274,
      "eval_rouge2": 15.0119,
      "eval_rougeL": 36.4563,
      "eval_rougeLsum": 36.4656,
      "eval_runtime": 450.3853,
      "eval_samples_per_second": 8.881,
      "eval_steps_per_second": 2.22,
      "step": 500
    },
    {
      "epoch": 0.1866193897545955,
      "grad_norm": 0.8436835408210754,
      "learning_rate": 4.6889386587034965e-05,
      "loss": 0.3748,
      "step": 1000
    },
    {
      "epoch": 0.1866193897545955,
      "eval_gen_len": 12.16675,
      "eval_loss": 0.46400758624076843,
      "eval_rouge1": 40.199,
      "eval_rouge2": 15.321,
      "eval_rougeL": 36.4279,
      "eval_rougeLsum": 36.4457,
      "eval_runtime": 431.1423,
      "eval_samples_per_second": 9.278,
      "eval_steps_per_second": 2.319,
      "step": 1000
    },
    {
      "epoch": 0.27992908463189325,
      "grad_norm": 0.8730084896087646,
      "learning_rate": 4.5334079880552446e-05,
      "loss": 0.374,
      "step": 1500
    },
    {
      "epoch": 0.27992908463189325,
      "eval_gen_len": 11.8,
      "eval_loss": 0.4668748080730438,
      "eval_rouge1": 39.9523,
      "eval_rouge2": 15.0587,
      "eval_rougeL": 36.3639,
      "eval_rougeLsum": 36.375,
      "eval_runtime": 424.8986,
      "eval_samples_per_second": 9.414,
      "eval_steps_per_second": 2.354,
      "step": 1500
    },
    {
      "epoch": 0.373238779509191,
      "grad_norm": 0.8020262718200684,
      "learning_rate": 4.3778773174069934e-05,
      "loss": 0.3721,
      "step": 2000
    },
    {
      "epoch": 0.373238779509191,
      "eval_gen_len": 11.2095,
      "eval_loss": 0.46446970105171204,
      "eval_rouge1": 40.3597,
      "eval_rouge2": 15.2173,
      "eval_rougeL": 36.6938,
      "eval_rougeLsum": 36.705,
      "eval_runtime": 410.7504,
      "eval_samples_per_second": 9.738,
      "eval_steps_per_second": 2.435,
      "step": 2000
    },
    {
      "epoch": 0.4665484743864888,
      "grad_norm": 0.8208432793617249,
      "learning_rate": 4.222346646758741e-05,
      "loss": 0.3673,
      "step": 2500
    },
    {
      "epoch": 0.4665484743864888,
      "eval_gen_len": 11.93425,
      "eval_loss": 0.4631750285625458,
      "eval_rouge1": 40.3875,
      "eval_rouge2": 15.2532,
      "eval_rougeL": 36.5923,
      "eval_rougeLsum": 36.6182,
      "eval_runtime": 441.242,
      "eval_samples_per_second": 9.065,
      "eval_steps_per_second": 2.266,
      "step": 2500
    },
    {
      "epoch": 0.5598581692637865,
      "grad_norm": 0.807547926902771,
      "learning_rate": 4.066815976110489e-05,
      "loss": 0.365,
      "step": 3000
    },
    {
      "epoch": 0.5598581692637865,
      "eval_gen_len": 12.26475,
      "eval_loss": 0.4623391330242157,
      "eval_rouge1": 39.9395,
      "eval_rouge2": 15.0315,
      "eval_rougeL": 36.1682,
      "eval_rougeLsum": 36.1781,
      "eval_runtime": 440.5083,
      "eval_samples_per_second": 9.08,
      "eval_steps_per_second": 2.27,
      "step": 3000
    },
    {
      "epoch": 0.6531678641410843,
      "grad_norm": 0.6434993147850037,
      "learning_rate": 3.911285305462237e-05,
      "loss": 0.3652,
      "step": 3500
    },
    {
      "epoch": 0.6531678641410843,
      "eval_gen_len": 11.8965,
      "eval_loss": 0.4611285328865051,
      "eval_rouge1": 39.8792,
      "eval_rouge2": 14.9961,
      "eval_rougeL": 36.2488,
      "eval_rougeLsum": 36.2734,
      "eval_runtime": 429.9024,
      "eval_samples_per_second": 9.304,
      "eval_steps_per_second": 2.326,
      "step": 3500
    },
    {
      "epoch": 0.746477559018382,
      "grad_norm": 0.8740318417549133,
      "learning_rate": 3.755754634813985e-05,
      "loss": 0.3601,
      "step": 4000
    },
    {
      "epoch": 0.746477559018382,
      "eval_gen_len": 12.0545,
      "eval_loss": 0.4625248610973358,
      "eval_rouge1": 40.57,
      "eval_rouge2": 15.2972,
      "eval_rougeL": 36.8012,
      "eval_rougeLsum": 36.8227,
      "eval_runtime": 431.0422,
      "eval_samples_per_second": 9.28,
      "eval_steps_per_second": 2.32,
      "step": 4000
    },
    {
      "epoch": 0.8397872538956798,
      "grad_norm": 0.7394187450408936,
      "learning_rate": 3.600223964165734e-05,
      "loss": 0.3574,
      "step": 4500
    },
    {
      "epoch": 0.8397872538956798,
      "eval_gen_len": 11.72875,
      "eval_loss": 0.4608415365219116,
      "eval_rouge1": 40.3276,
      "eval_rouge2": 15.1742,
      "eval_rougeL": 36.7679,
      "eval_rougeLsum": 36.7575,
      "eval_runtime": 467.6774,
      "eval_samples_per_second": 8.553,
      "eval_steps_per_second": 2.138,
      "step": 4500
    },
    {
      "epoch": 0.9330969487729776,
      "grad_norm": 0.841187059879303,
      "learning_rate": 3.444693293517482e-05,
      "loss": 0.351,
      "step": 5000
    },
    {
      "epoch": 0.9330969487729776,
      "eval_gen_len": 11.76625,
      "eval_loss": 0.4650140404701233,
      "eval_rouge1": 40.7345,
      "eval_rouge2": 15.5295,
      "eval_rougeL": 37.0769,
      "eval_rougeLsum": 37.0911,
      "eval_runtime": 475.4278,
      "eval_samples_per_second": 8.413,
      "eval_steps_per_second": 2.103,
      "step": 5000
    }
  ],
  "logging_steps": 500,
  "max_steps": 16074,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.443711128961024e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}