SF-Foundation commited on
Commit
7f98715
·
verified ·
1 Parent(s): 06882ff

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +134 -0
trainer_state.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.05115089514066496,
6
+ "eval_steps": 500,
7
+ "global_step": 100,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "entropy": 3.2780354261398315,
14
+ "epoch": 0.005115089514066497,
15
+ "grad_norm": 25.268398857866387,
16
+ "learning_rate": 7.666098807495741e-07,
17
+ "loss": 2.5805,
18
+ "mean_token_accuracy": 0.44439449459314345,
19
+ "num_tokens": 201201.0,
20
+ "step": 10
21
+ },
22
+ {
23
+ "entropy": 3.270029139518738,
24
+ "epoch": 0.010230179028132993,
25
+ "grad_norm": 61.66870761901112,
26
+ "learning_rate": 1.6183986371379898e-06,
27
+ "loss": 2.6136,
28
+ "mean_token_accuracy": 0.4849040985107422,
29
+ "num_tokens": 404577.0,
30
+ "step": 20
31
+ },
32
+ {
33
+ "entropy": 3.303619647026062,
34
+ "epoch": 0.015345268542199489,
35
+ "grad_norm": 85.93666778656973,
36
+ "learning_rate": 2.4701873935264056e-06,
37
+ "loss": 2.6838,
38
+ "mean_token_accuracy": 0.4358963340520859,
39
+ "num_tokens": 609983.0,
40
+ "step": 30
41
+ },
42
+ {
43
+ "entropy": 3.291827130317688,
44
+ "epoch": 0.020460358056265986,
45
+ "grad_norm": 15.511161480666924,
46
+ "learning_rate": 3.321976149914821e-06,
47
+ "loss": 1.8307,
48
+ "mean_token_accuracy": 0.5674405485391617,
49
+ "num_tokens": 802203.0,
50
+ "step": 40
51
+ },
52
+ {
53
+ "entropy": 3.367189645767212,
54
+ "epoch": 0.02557544757033248,
55
+ "grad_norm": 14.365831800288149,
56
+ "learning_rate": 4.173764906303237e-06,
57
+ "loss": 1.2772,
58
+ "mean_token_accuracy": 0.6300624251365662,
59
+ "num_tokens": 1024749.0,
60
+ "step": 50
61
+ },
62
+ {
63
+ "entropy": 3.36285605430603,
64
+ "epoch": 0.030690537084398978,
65
+ "grad_norm": 33.84168925997731,
66
+ "learning_rate": 5.025553662691653e-06,
67
+ "loss": 0.9931,
68
+ "mean_token_accuracy": 0.7703894019126892,
69
+ "num_tokens": 1242772.0,
70
+ "step": 60
71
+ },
72
+ {
73
+ "entropy": 3.3700726509094237,
74
+ "epoch": 0.03580562659846547,
75
+ "grad_norm": 62.73563012853256,
76
+ "learning_rate": 5.877342419080068e-06,
77
+ "loss": 1.1674,
78
+ "mean_token_accuracy": 0.74255530834198,
79
+ "num_tokens": 1458515.0,
80
+ "step": 70
81
+ },
82
+ {
83
+ "entropy": 3.516653847694397,
84
+ "epoch": 0.04092071611253197,
85
+ "grad_norm": 3.5686450164550396,
86
+ "learning_rate": 6.7291311754684835e-06,
87
+ "loss": 1.0982,
88
+ "mean_token_accuracy": 0.7218687653541564,
89
+ "num_tokens": 1655834.0,
90
+ "step": 80
91
+ },
92
+ {
93
+ "entropy": 3.517510437965393,
94
+ "epoch": 0.04603580562659847,
95
+ "grad_norm": 22.108671955044745,
96
+ "learning_rate": 7.5809199318569e-06,
97
+ "loss": 0.8307,
98
+ "mean_token_accuracy": 0.771921706199646,
99
+ "num_tokens": 1871549.0,
100
+ "step": 90
101
+ },
102
+ {
103
+ "entropy": 3.653357815742493,
104
+ "epoch": 0.05115089514066496,
105
+ "grad_norm": 21.61728498405485,
106
+ "learning_rate": 8.432708688245315e-06,
107
+ "loss": 0.9384,
108
+ "mean_token_accuracy": 0.7474159002304077,
109
+ "num_tokens": 2079472.0,
110
+ "step": 100
111
+ }
112
+ ],
113
+ "logging_steps": 10,
114
+ "max_steps": 5865,
115
+ "num_input_tokens_seen": 0,
116
+ "num_train_epochs": 3,
117
+ "save_steps": 100,
118
+ "stateful_callbacks": {
119
+ "TrainerControl": {
120
+ "args": {
121
+ "should_epoch_stop": false,
122
+ "should_evaluate": false,
123
+ "should_log": false,
124
+ "should_save": true,
125
+ "should_training_stop": false
126
+ },
127
+ "attributes": {}
128
+ }
129
+ },
130
+ "total_flos": 213113167151104.0,
131
+ "train_batch_size": 2,
132
+ "trial_name": null,
133
+ "trial_params": null
134
+ }