Log update epoch 9
Browse files- training_log.json +113 -0
training_log.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"epochs": 10,
|
| 4 |
+
"lr": 0.0003,
|
| 5 |
+
"warmup_steps": 300,
|
| 6 |
+
"grad_accum_steps": 8,
|
| 7 |
+
"grad_clip": 1.0
|
| 8 |
+
},
|
| 9 |
+
"train": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1,
|
| 12 |
+
"avg_loss": 7.070409,
|
| 13 |
+
"lr": 0.00027,
|
| 14 |
+
"step": 270,
|
| 15 |
+
"time_s": 971.0
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 2,
|
| 19 |
+
"avg_loss": 6.138723,
|
| 20 |
+
"lr": 0.0002986039232296601,
|
| 21 |
+
"step": 540,
|
| 22 |
+
"time_s": 969.5
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 3,
|
| 26 |
+
"avg_loss": 5.511719,
|
| 27 |
+
"lr": 0.00029373201732051774,
|
| 28 |
+
"step": 810,
|
| 29 |
+
"time_s": 970.5
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 4,
|
| 33 |
+
"avg_loss": 5.033876,
|
| 34 |
+
"lr": 0.00028548319801432657,
|
| 35 |
+
"step": 1080,
|
| 36 |
+
"time_s": 966.4
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 5,
|
| 40 |
+
"avg_loss": 4.711938,
|
| 41 |
+
"lr": 0.00027406193668993577,
|
| 42 |
+
"step": 1350,
|
| 43 |
+
"time_s": 977.1
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 6,
|
| 47 |
+
"avg_loss": 4.464801,
|
| 48 |
+
"lr": 0.00027245653264858043,
|
| 49 |
+
"step": 1620,
|
| 50 |
+
"time_s": 973.7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 7,
|
| 54 |
+
"avg_loss": 4.292098,
|
| 55 |
+
"lr": 0.0002676801152109725,
|
| 56 |
+
"step": 1890,
|
| 57 |
+
"time_s": 977.7
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 8,
|
| 61 |
+
"avg_loss": 4.150779,
|
| 62 |
+
"lr": 0.00025985108200750904,
|
| 63 |
+
"step": 2160,
|
| 64 |
+
"time_s": 975.7
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 9,
|
| 68 |
+
"avg_loss": 3.803133,
|
| 69 |
+
"lr": 0.0002587006831839486,
|
| 70 |
+
"step": 2170,
|
| 71 |
+
"time_s": 34.3
|
| 72 |
+
}
|
| 73 |
+
],
|
| 74 |
+
"val": [
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1,
|
| 77 |
+
"avg_loss": 6.459693
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2,
|
| 81 |
+
"avg_loss": 5.804065
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 3,
|
| 85 |
+
"avg_loss": 5.179338
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 4,
|
| 89 |
+
"avg_loss": 4.774045
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5,
|
| 93 |
+
"avg_loss": 4.511194
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 6,
|
| 97 |
+
"avg_loss": 4.305547
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 7,
|
| 101 |
+
"avg_loss": 4.154681
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 8,
|
| 105 |
+
"avg_loss": 4.037178
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 9,
|
| 109 |
+
"avg_loss": 3.911244
|
| 110 |
+
}
|
| 111 |
+
],
|
| 112 |
+
"sanity": []
|
| 113 |
+
}
|