leolin12345
commited on
Commit
•
b273d11
1
Parent(s):
b713e4e
Upload trainer_state.json
Browse files- trainer_state.json +76 -0
trainer_state.json
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 4.775549188156638,
|
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": 0.48,
|
12 |
+
"learning_rate": 1.8089780324737347e-05,
|
13 |
+
"loss": 1.2572,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.96,
|
18 |
+
"learning_rate": 1.617956064947469e-05,
|
19 |
+
"loss": 1.0388,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.43,
|
24 |
+
"learning_rate": 1.4269340974212036e-05,
|
25 |
+
"loss": 0.9388,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 1.91,
|
30 |
+
"learning_rate": 1.235912129894938e-05,
|
31 |
+
"loss": 0.8806,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 2.39,
|
36 |
+
"learning_rate": 1.0448901623686724e-05,
|
37 |
+
"loss": 0.8384,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 2.87,
|
42 |
+
"learning_rate": 8.53868194842407e-06,
|
43 |
+
"loss": 0.8012,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 3.34,
|
48 |
+
"learning_rate": 6.628462273161415e-06,
|
49 |
+
"loss": 0.7839,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 3.82,
|
54 |
+
"learning_rate": 4.718242597898759e-06,
|
55 |
+
"loss": 0.7627,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 4.3,
|
60 |
+
"learning_rate": 2.8080229226361035e-06,
|
61 |
+
"loss": 0.7479,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 4.78,
|
66 |
+
"learning_rate": 8.97803247373448e-07,
|
67 |
+
"loss": 0.7422,
|
68 |
+
"step": 5000
|
69 |
+
}
|
70 |
+
],
|
71 |
+
"max_steps": 5235,
|
72 |
+
"num_train_epochs": 5,
|
73 |
+
"total_flos": 1.3155901731827712e+16,
|
74 |
+
"trial_name": null,
|
75 |
+
"trial_params": null
|
76 |
+
}
|