tesemnikov-av
commited on
Commit
•
eb585e0
1
Parent(s):
5bf017a
Upload trainer_state.json
Browse files- trainer_state.json +112 -0
trainer_state.json
ADDED
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.5536602700781805,
|
5 |
+
"global_step": 5000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.36,
|
12 |
+
"learning_rate": 1.9289267945984364e-05,
|
13 |
+
"loss": 0.3176,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.71,
|
18 |
+
"learning_rate": 1.857853589196873e-05,
|
19 |
+
"loss": 0.146,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.0,
|
24 |
+
"eval_accuracy": 0.9562212252875999,
|
25 |
+
"eval_f1": 0.7288810260946483,
|
26 |
+
"eval_loss": 0.13318254053592682,
|
27 |
+
"eval_precision": 0.6725223474543335,
|
28 |
+
"eval_recall": 0.7955496299020735,
|
29 |
+
"eval_runtime": 57.743,
|
30 |
+
"eval_samples_per_second": 403.287,
|
31 |
+
"eval_steps_per_second": 3.152,
|
32 |
+
"step": 1407
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.07,
|
36 |
+
"learning_rate": 1.7867803837953093e-05,
|
37 |
+
"loss": 0.1295,
|
38 |
+
"step": 1500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.42,
|
42 |
+
"learning_rate": 1.715707178393746e-05,
|
43 |
+
"loss": 0.1224,
|
44 |
+
"step": 2000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.78,
|
48 |
+
"learning_rate": 1.6446339729921822e-05,
|
49 |
+
"loss": 0.1169,
|
50 |
+
"step": 2500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 2.0,
|
54 |
+
"eval_accuracy": 0.9605602458582077,
|
55 |
+
"eval_f1": 0.7573750499379718,
|
56 |
+
"eval_loss": 0.11934245377779007,
|
57 |
+
"eval_precision": 0.6978456292622442,
|
58 |
+
"eval_recall": 0.8280079076824054,
|
59 |
+
"eval_runtime": 57.4865,
|
60 |
+
"eval_samples_per_second": 405.087,
|
61 |
+
"eval_steps_per_second": 3.166,
|
62 |
+
"step": 2814
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.13,
|
66 |
+
"learning_rate": 1.5735607675906184e-05,
|
67 |
+
"loss": 0.1119,
|
68 |
+
"step": 3000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 2.49,
|
72 |
+
"learning_rate": 1.5024875621890549e-05,
|
73 |
+
"loss": 0.1095,
|
74 |
+
"step": 3500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 2.84,
|
78 |
+
"learning_rate": 1.4314143567874911e-05,
|
79 |
+
"loss": 0.1069,
|
80 |
+
"step": 4000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 3.0,
|
84 |
+
"eval_accuracy": 0.9633008348346227,
|
85 |
+
"eval_f1": 0.7770200720535254,
|
86 |
+
"eval_loss": 0.11019115895032883,
|
87 |
+
"eval_precision": 0.7281459748402396,
|
88 |
+
"eval_recall": 0.8329272217369317,
|
89 |
+
"eval_runtime": 57.3128,
|
90 |
+
"eval_samples_per_second": 406.314,
|
91 |
+
"eval_steps_per_second": 3.176,
|
92 |
+
"step": 4221
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 3.2,
|
96 |
+
"learning_rate": 1.3603411513859277e-05,
|
97 |
+
"loss": 0.1044,
|
98 |
+
"step": 4500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 3.55,
|
102 |
+
"learning_rate": 1.289267945984364e-05,
|
103 |
+
"loss": 0.1038,
|
104 |
+
"step": 5000
|
105 |
+
}
|
106 |
+
],
|
107 |
+
"max_steps": 14070,
|
108 |
+
"num_train_epochs": 10,
|
109 |
+
"total_flos": 3387450405839616.0,
|
110 |
+
"trial_name": null,
|
111 |
+
"trial_params": null
|
112 |
+
}
|