Hugging Face's logo

Davlan
/
bloom-560m_am_continual-pretrain_10000samples

Text Generation
Transformers
PyTorch
bloom
Inference Endpoints
text-generation-inference
Model card Files Community
Davlan commited on
Commit
3596cc3
•
1 Parent(s): 6048b0d

Upload trainer_state.json

Browse files
Files changed (1) hide show
  1. trainer_state.json +125 -0
trainer_state.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.0630253553390503,
3
+ "best_model_checkpoint": "/data/users/didelani/multilingual-modeling/bloom-560m_am_continual-pretrain_10000samples_-1vocab_original/checkpoint-10000",
4
+ "epoch": 6.8417883898334075,
5
+ "global_step": 25000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.68,
12
+ "learning_rate": 9e-05,
13
+ "loss": 1.3776,
14
+ "step": 2500
15
+ },
16
+ {
17
+ "epoch": 1.37,
18
+ "learning_rate": 8e-05,
19
+ "loss": 1.1298,
20
+ "step": 5000
21
+ },
22
+ {
23
+ "epoch": 1.37,
24
+ "eval_loss": 1.1117359399795532,
25
+ "eval_runtime": 419.467,
26
+ "eval_samples_per_second": 6.847,
27
+ "eval_steps_per_second": 3.423,
28
+ "step": 5000
29
+ },
30
+ {
31
+ "epoch": 2.05,
32
+ "learning_rate": 7e-05,
33
+ "loss": 1.0638,
34
+ "step": 7500
35
+ },
36
+ {
37
+ "epoch": 2.74,
38
+ "learning_rate": 6e-05,
39
+ "loss": 0.9777,
40
+ "step": 10000
41
+ },
42
+ {
43
+ "epoch": 2.74,
44
+ "eval_loss": 1.0630253553390503,
45
+ "eval_runtime": 421.4286,
46
+ "eval_samples_per_second": 6.815,
47
+ "eval_steps_per_second": 3.407,
48
+ "step": 10000
49
+ },
50
+ {
51
+ "epoch": 3.42,
52
+ "learning_rate": 5e-05,
53
+ "loss": 0.9059,
54
+ "step": 12500
55
+ },
56
+ {
57
+ "epoch": 4.11,
58
+ "learning_rate": 4e-05,
59
+ "loss": 0.8435,
60
+ "step": 15000
61
+ },
62
+ {
63
+ "epoch": 4.11,
64
+ "eval_loss": 1.1998958587646484,
65
+ "eval_runtime": 422.8876,
66
+ "eval_samples_per_second": 6.791,
67
+ "eval_steps_per_second": 3.396,
68
+ "step": 15000
69
+ },
70
+ {
71
+ "epoch": 4.79,
72
+ "learning_rate": 3e-05,
73
+ "loss": 0.7247,
74
+ "step": 17500
75
+ },
76
+ {
77
+ "epoch": 5.47,
78
+ "learning_rate": 2e-05,
79
+ "loss": 0.6063,
80
+ "step": 20000
81
+ },
82
+ {
83
+ "epoch": 5.47,
84
+ "eval_loss": 1.5086530447006226,
85
+ "eval_runtime": 417.6938,
86
+ "eval_samples_per_second": 6.876,
87
+ "eval_steps_per_second": 3.438,
88
+ "step": 20000
89
+ },
90
+ {
91
+ "epoch": 6.16,
92
+ "learning_rate": 1e-05,
93
+ "loss": 0.5155,
94
+ "step": 22500
95
+ },
96
+ {
97
+ "epoch": 6.84,
98
+ "learning_rate": 0.0,
99
+ "loss": 0.386,
100
+ "step": 25000
101
+ },
102
+ {
103
+ "epoch": 6.84,
104
+ "eval_loss": 2.100792646408081,
105
+ "eval_runtime": 418.1577,
106
+ "eval_samples_per_second": 6.868,
107
+ "eval_steps_per_second": 3.434,
108
+ "step": 25000
109
+ },
110
+ {
111
+ "epoch": 6.84,
112
+ "step": 25000,
113
+ "total_flos": 3.714939388774318e+17,
114
+ "train_loss": 0.8530817114257813,
115
+ "train_runtime": 94563.3332,
116
+ "train_samples_per_second": 2.115,
117
+ "train_steps_per_second": 0.264
118
+ }
119
+ ],
120
+ "max_steps": 25000,
121
+ "num_train_epochs": 7,
122
+ "total_flos": 3.714939388774318e+17,
123
+ "trial_name": null,
124
+ "trial_params": null
125
+ }