Commit
•
9b814a3
1
Parent(s):
cb57d23
End of training
Browse files- README.md +2 -2
- all_results.json +8 -0
- runs/Apr04_19-08-19_4b69906a4c29/events.out.tfevents.1712259622.4b69906a4c29.2785.1 +3 -0
- test_results.json +8 -0
- trainer_state.json +663 -0
README.md
CHANGED
|
@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model was trained from scratch on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
-
- Accuracy: 0.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 15 |
|
| 16 |
This model was trained from scratch on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.0202
|
| 19 |
+
- Accuracy: 0.9887
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.25,
|
| 3 |
+
"eval_accuracy": 0.9887429643527205,
|
| 4 |
+
"eval_loss": 0.020199885591864586,
|
| 5 |
+
"eval_runtime": 328.4712,
|
| 6 |
+
"eval_samples_per_second": 1.623,
|
| 7 |
+
"eval_steps_per_second": 0.813
|
| 8 |
+
}
|
runs/Apr04_19-08-19_4b69906a4c29/events.out.tfevents.1712259622.4b69906a4c29.2785.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6cea177e1c3aac62fb4fee584c5960c2b6d75186fdb9e80cb17101e3b91096a8
|
| 3 |
+
size 411
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.25,
|
| 3 |
+
"eval_accuracy": 0.9887429643527205,
|
| 4 |
+
"eval_loss": 0.020199885591864586,
|
| 5 |
+
"eval_runtime": 328.4712,
|
| 6 |
+
"eval_samples_per_second": 1.623,
|
| 7 |
+
"eval_steps_per_second": 0.813
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,663 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9964912280701754,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-bekhoaxe/checkpoint-636",
|
| 4 |
+
"epoch": 3.25,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 848,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 10.807732582092285,
|
| 14 |
+
"learning_rate": 5.882352941176471e-06,
|
| 15 |
+
"loss": 0.6707,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"grad_norm": 6.269227504730225,
|
| 21 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 22 |
+
"loss": 0.6435,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04,
|
| 27 |
+
"grad_norm": 10.753657341003418,
|
| 28 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 29 |
+
"loss": 0.571,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05,
|
| 34 |
+
"grad_norm": 9.196399688720703,
|
| 35 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 36 |
+
"loss": 0.4578,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06,
|
| 41 |
+
"grad_norm": 3.8521430492401123,
|
| 42 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 43 |
+
"loss": 0.2468,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07,
|
| 48 |
+
"grad_norm": 4.0225911140441895,
|
| 49 |
+
"learning_rate": 3.529411764705883e-05,
|
| 50 |
+
"loss": 0.1737,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.08,
|
| 55 |
+
"grad_norm": 0.2649253010749817,
|
| 56 |
+
"learning_rate": 4.11764705882353e-05,
|
| 57 |
+
"loss": 0.0356,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.09,
|
| 62 |
+
"grad_norm": 106.07608795166016,
|
| 63 |
+
"learning_rate": 4.705882352941177e-05,
|
| 64 |
+
"loss": 0.3572,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11,
|
| 69 |
+
"grad_norm": 0.02139919251203537,
|
| 70 |
+
"learning_rate": 4.9672346002621236e-05,
|
| 71 |
+
"loss": 0.2095,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.12,
|
| 76 |
+
"grad_norm": 87.85816955566406,
|
| 77 |
+
"learning_rate": 4.90170380078637e-05,
|
| 78 |
+
"loss": 0.389,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.13,
|
| 83 |
+
"grad_norm": 0.4829784035682678,
|
| 84 |
+
"learning_rate": 4.836173001310616e-05,
|
| 85 |
+
"loss": 0.1773,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.14,
|
| 90 |
+
"grad_norm": 0.020957980304956436,
|
| 91 |
+
"learning_rate": 4.7706422018348626e-05,
|
| 92 |
+
"loss": 0.0059,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.15,
|
| 97 |
+
"grad_norm": 0.018635602667927742,
|
| 98 |
+
"learning_rate": 4.705111402359109e-05,
|
| 99 |
+
"loss": 1.1257,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.17,
|
| 104 |
+
"grad_norm": 0.2680661976337433,
|
| 105 |
+
"learning_rate": 4.6395806028833554e-05,
|
| 106 |
+
"loss": 0.4074,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.18,
|
| 111 |
+
"grad_norm": 0.034663423895835876,
|
| 112 |
+
"learning_rate": 4.5740498034076015e-05,
|
| 113 |
+
"loss": 0.0125,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.19,
|
| 118 |
+
"grad_norm": 0.027404017746448517,
|
| 119 |
+
"learning_rate": 4.508519003931848e-05,
|
| 120 |
+
"loss": 0.1217,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2,
|
| 125 |
+
"grad_norm": 25.80691146850586,
|
| 126 |
+
"learning_rate": 4.4429882044560943e-05,
|
| 127 |
+
"loss": 0.3167,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.21,
|
| 132 |
+
"grad_norm": 0.13836145401000977,
|
| 133 |
+
"learning_rate": 4.3774574049803404e-05,
|
| 134 |
+
"loss": 0.5605,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.22,
|
| 139 |
+
"grad_norm": 0.010564002208411694,
|
| 140 |
+
"learning_rate": 4.311926605504588e-05,
|
| 141 |
+
"loss": 0.002,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.24,
|
| 146 |
+
"grad_norm": 0.02207314595580101,
|
| 147 |
+
"learning_rate": 4.246395806028834e-05,
|
| 148 |
+
"loss": 0.1598,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.25,
|
| 153 |
+
"grad_norm": 0.039745017886161804,
|
| 154 |
+
"learning_rate": 4.18086500655308e-05,
|
| 155 |
+
"loss": 0.337,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.25,
|
| 160 |
+
"eval_accuracy": 0.9508771929824561,
|
| 161 |
+
"eval_loss": 0.13453243672847748,
|
| 162 |
+
"eval_runtime": 161.5975,
|
| 163 |
+
"eval_samples_per_second": 1.764,
|
| 164 |
+
"eval_steps_per_second": 0.885,
|
| 165 |
+
"step": 212
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.01,
|
| 169 |
+
"grad_norm": 1.155490756034851,
|
| 170 |
+
"learning_rate": 4.115334207077327e-05,
|
| 171 |
+
"loss": 0.3642,
|
| 172 |
+
"step": 220
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.02,
|
| 176 |
+
"grad_norm": 0.03908234462141991,
|
| 177 |
+
"learning_rate": 4.049803407601573e-05,
|
| 178 |
+
"loss": 0.0017,
|
| 179 |
+
"step": 230
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.03,
|
| 183 |
+
"grad_norm": 0.07498504221439362,
|
| 184 |
+
"learning_rate": 3.984272608125819e-05,
|
| 185 |
+
"loss": 0.0012,
|
| 186 |
+
"step": 240
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.04,
|
| 190 |
+
"grad_norm": 0.011099644936621189,
|
| 191 |
+
"learning_rate": 3.918741808650066e-05,
|
| 192 |
+
"loss": 0.6515,
|
| 193 |
+
"step": 250
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.06,
|
| 197 |
+
"grad_norm": 0.13020159304141998,
|
| 198 |
+
"learning_rate": 3.8532110091743125e-05,
|
| 199 |
+
"loss": 0.0009,
|
| 200 |
+
"step": 260
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.07,
|
| 204 |
+
"grad_norm": 0.03461850434541702,
|
| 205 |
+
"learning_rate": 3.7876802096985586e-05,
|
| 206 |
+
"loss": 0.3921,
|
| 207 |
+
"step": 270
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.08,
|
| 211 |
+
"grad_norm": 0.013056197203695774,
|
| 212 |
+
"learning_rate": 3.722149410222805e-05,
|
| 213 |
+
"loss": 0.2843,
|
| 214 |
+
"step": 280
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.09,
|
| 218 |
+
"grad_norm": 0.00816601887345314,
|
| 219 |
+
"learning_rate": 3.6566186107470514e-05,
|
| 220 |
+
"loss": 0.2704,
|
| 221 |
+
"step": 290
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.1,
|
| 225 |
+
"grad_norm": 0.4981946647167206,
|
| 226 |
+
"learning_rate": 3.5910878112712975e-05,
|
| 227 |
+
"loss": 0.0053,
|
| 228 |
+
"step": 300
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.12,
|
| 232 |
+
"grad_norm": 0.014170478098094463,
|
| 233 |
+
"learning_rate": 3.5255570117955436e-05,
|
| 234 |
+
"loss": 0.2984,
|
| 235 |
+
"step": 310
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.13,
|
| 239 |
+
"grad_norm": 0.03322271630167961,
|
| 240 |
+
"learning_rate": 3.460026212319791e-05,
|
| 241 |
+
"loss": 0.0012,
|
| 242 |
+
"step": 320
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.14,
|
| 246 |
+
"grad_norm": 0.019952110946178436,
|
| 247 |
+
"learning_rate": 3.394495412844037e-05,
|
| 248 |
+
"loss": 0.0167,
|
| 249 |
+
"step": 330
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.15,
|
| 253 |
+
"grad_norm": 2.8213207721710205,
|
| 254 |
+
"learning_rate": 3.328964613368283e-05,
|
| 255 |
+
"loss": 0.4437,
|
| 256 |
+
"step": 340
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.16,
|
| 260 |
+
"grad_norm": 0.026097455993294716,
|
| 261 |
+
"learning_rate": 3.26343381389253e-05,
|
| 262 |
+
"loss": 0.0015,
|
| 263 |
+
"step": 350
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.17,
|
| 267 |
+
"grad_norm": 0.32980480790138245,
|
| 268 |
+
"learning_rate": 3.197903014416776e-05,
|
| 269 |
+
"loss": 0.2044,
|
| 270 |
+
"step": 360
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.19,
|
| 274 |
+
"grad_norm": 0.3815198540687561,
|
| 275 |
+
"learning_rate": 3.132372214941022e-05,
|
| 276 |
+
"loss": 0.1395,
|
| 277 |
+
"step": 370
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 1.2,
|
| 281 |
+
"grad_norm": 0.014452760107815266,
|
| 282 |
+
"learning_rate": 3.066841415465269e-05,
|
| 283 |
+
"loss": 0.0014,
|
| 284 |
+
"step": 380
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.21,
|
| 288 |
+
"grad_norm": 7.192620754241943,
|
| 289 |
+
"learning_rate": 3.0013106159895154e-05,
|
| 290 |
+
"loss": 0.0059,
|
| 291 |
+
"step": 390
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.22,
|
| 295 |
+
"grad_norm": 0.009683230891823769,
|
| 296 |
+
"learning_rate": 2.9357798165137618e-05,
|
| 297 |
+
"loss": 0.0054,
|
| 298 |
+
"step": 400
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.23,
|
| 302 |
+
"grad_norm": 0.0035556950606405735,
|
| 303 |
+
"learning_rate": 2.8702490170380082e-05,
|
| 304 |
+
"loss": 0.0122,
|
| 305 |
+
"step": 410
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.25,
|
| 309 |
+
"grad_norm": 0.010539459995925426,
|
| 310 |
+
"learning_rate": 2.8047182175622543e-05,
|
| 311 |
+
"loss": 0.5753,
|
| 312 |
+
"step": 420
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.25,
|
| 316 |
+
"eval_accuracy": 0.9789473684210527,
|
| 317 |
+
"eval_loss": 0.05574525147676468,
|
| 318 |
+
"eval_runtime": 89.1683,
|
| 319 |
+
"eval_samples_per_second": 3.196,
|
| 320 |
+
"eval_steps_per_second": 1.604,
|
| 321 |
+
"step": 424
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.01,
|
| 325 |
+
"grad_norm": 0.017384065315127373,
|
| 326 |
+
"learning_rate": 2.7391874180865007e-05,
|
| 327 |
+
"loss": 0.0032,
|
| 328 |
+
"step": 430
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.02,
|
| 332 |
+
"grad_norm": 0.041157085448503494,
|
| 333 |
+
"learning_rate": 2.673656618610747e-05,
|
| 334 |
+
"loss": 0.0388,
|
| 335 |
+
"step": 440
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.03,
|
| 339 |
+
"grad_norm": 0.014596754685044289,
|
| 340 |
+
"learning_rate": 2.6081258191349932e-05,
|
| 341 |
+
"loss": 0.002,
|
| 342 |
+
"step": 450
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 2.04,
|
| 346 |
+
"grad_norm": 0.004826271440833807,
|
| 347 |
+
"learning_rate": 2.5425950196592403e-05,
|
| 348 |
+
"loss": 0.1837,
|
| 349 |
+
"step": 460
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 2.05,
|
| 353 |
+
"grad_norm": 0.01182369515299797,
|
| 354 |
+
"learning_rate": 2.4770642201834864e-05,
|
| 355 |
+
"loss": 0.2806,
|
| 356 |
+
"step": 470
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 2.07,
|
| 360 |
+
"grad_norm": 0.012596765533089638,
|
| 361 |
+
"learning_rate": 2.411533420707733e-05,
|
| 362 |
+
"loss": 0.001,
|
| 363 |
+
"step": 480
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.08,
|
| 367 |
+
"grad_norm": 0.014254101552069187,
|
| 368 |
+
"learning_rate": 2.3460026212319793e-05,
|
| 369 |
+
"loss": 0.0149,
|
| 370 |
+
"step": 490
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 2.09,
|
| 374 |
+
"grad_norm": 0.0014000836526975036,
|
| 375 |
+
"learning_rate": 2.2804718217562254e-05,
|
| 376 |
+
"loss": 0.3089,
|
| 377 |
+
"step": 500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.1,
|
| 381 |
+
"grad_norm": 0.0013152866158634424,
|
| 382 |
+
"learning_rate": 2.214941022280472e-05,
|
| 383 |
+
"loss": 0.0019,
|
| 384 |
+
"step": 510
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.11,
|
| 388 |
+
"grad_norm": 0.0013162486720830202,
|
| 389 |
+
"learning_rate": 2.1494102228047182e-05,
|
| 390 |
+
"loss": 0.0012,
|
| 391 |
+
"step": 520
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.12,
|
| 395 |
+
"grad_norm": 0.01350224670022726,
|
| 396 |
+
"learning_rate": 2.0838794233289646e-05,
|
| 397 |
+
"loss": 0.0004,
|
| 398 |
+
"step": 530
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.14,
|
| 402 |
+
"grad_norm": 0.0011886212741956115,
|
| 403 |
+
"learning_rate": 2.018348623853211e-05,
|
| 404 |
+
"loss": 0.0002,
|
| 405 |
+
"step": 540
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.15,
|
| 409 |
+
"grad_norm": 0.013104724697768688,
|
| 410 |
+
"learning_rate": 1.9528178243774575e-05,
|
| 411 |
+
"loss": 0.0005,
|
| 412 |
+
"step": 550
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 2.16,
|
| 416 |
+
"grad_norm": 141.24769592285156,
|
| 417 |
+
"learning_rate": 1.887287024901704e-05,
|
| 418 |
+
"loss": 0.0858,
|
| 419 |
+
"step": 560
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 2.17,
|
| 423 |
+
"grad_norm": 89.11790466308594,
|
| 424 |
+
"learning_rate": 1.82175622542595e-05,
|
| 425 |
+
"loss": 0.6766,
|
| 426 |
+
"step": 570
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 2.18,
|
| 430 |
+
"grad_norm": 0.03498009964823723,
|
| 431 |
+
"learning_rate": 1.7562254259501968e-05,
|
| 432 |
+
"loss": 0.0006,
|
| 433 |
+
"step": 580
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 2.2,
|
| 437 |
+
"grad_norm": 0.03515917435288429,
|
| 438 |
+
"learning_rate": 1.6906946264744432e-05,
|
| 439 |
+
"loss": 0.0003,
|
| 440 |
+
"step": 590
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 2.21,
|
| 444 |
+
"grad_norm": 0.003264626022428274,
|
| 445 |
+
"learning_rate": 1.6251638269986893e-05,
|
| 446 |
+
"loss": 0.0004,
|
| 447 |
+
"step": 600
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.22,
|
| 451 |
+
"grad_norm": 0.011713879182934761,
|
| 452 |
+
"learning_rate": 1.559633027522936e-05,
|
| 453 |
+
"loss": 0.5946,
|
| 454 |
+
"step": 610
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 2.23,
|
| 458 |
+
"grad_norm": 0.06233768165111542,
|
| 459 |
+
"learning_rate": 1.4941022280471823e-05,
|
| 460 |
+
"loss": 0.1258,
|
| 461 |
+
"step": 620
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 2.24,
|
| 465 |
+
"grad_norm": 0.0129940714687109,
|
| 466 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 467 |
+
"loss": 0.1747,
|
| 468 |
+
"step": 630
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 2.25,
|
| 472 |
+
"eval_accuracy": 0.9964912280701754,
|
| 473 |
+
"eval_loss": 0.009211267344653606,
|
| 474 |
+
"eval_runtime": 75.987,
|
| 475 |
+
"eval_samples_per_second": 3.751,
|
| 476 |
+
"eval_steps_per_second": 1.882,
|
| 477 |
+
"step": 636
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 3.0,
|
| 481 |
+
"grad_norm": 0.0608808733522892,
|
| 482 |
+
"learning_rate": 1.3630406290956751e-05,
|
| 483 |
+
"loss": 0.0226,
|
| 484 |
+
"step": 640
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 3.02,
|
| 488 |
+
"grad_norm": 0.08016235381364822,
|
| 489 |
+
"learning_rate": 1.2975098296199214e-05,
|
| 490 |
+
"loss": 0.0644,
|
| 491 |
+
"step": 650
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.03,
|
| 495 |
+
"grad_norm": 0.6168311834335327,
|
| 496 |
+
"learning_rate": 1.2319790301441678e-05,
|
| 497 |
+
"loss": 0.0011,
|
| 498 |
+
"step": 660
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 3.04,
|
| 502 |
+
"grad_norm": 0.0013453299179673195,
|
| 503 |
+
"learning_rate": 1.1664482306684142e-05,
|
| 504 |
+
"loss": 0.1834,
|
| 505 |
+
"step": 670
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 3.05,
|
| 509 |
+
"grad_norm": 0.0041399141773581505,
|
| 510 |
+
"learning_rate": 1.1009174311926607e-05,
|
| 511 |
+
"loss": 0.2115,
|
| 512 |
+
"step": 680
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 3.06,
|
| 516 |
+
"grad_norm": 0.10097178816795349,
|
| 517 |
+
"learning_rate": 1.035386631716907e-05,
|
| 518 |
+
"loss": 0.2718,
|
| 519 |
+
"step": 690
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.08,
|
| 523 |
+
"grad_norm": 0.03554755076766014,
|
| 524 |
+
"learning_rate": 9.698558322411533e-06,
|
| 525 |
+
"loss": 0.0316,
|
| 526 |
+
"step": 700
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 3.09,
|
| 530 |
+
"grad_norm": 0.3528023958206177,
|
| 531 |
+
"learning_rate": 9.043250327653998e-06,
|
| 532 |
+
"loss": 0.001,
|
| 533 |
+
"step": 710
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 3.1,
|
| 537 |
+
"grad_norm": 0.016770780086517334,
|
| 538 |
+
"learning_rate": 8.387942332896462e-06,
|
| 539 |
+
"loss": 0.0004,
|
| 540 |
+
"step": 720
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 3.11,
|
| 544 |
+
"grad_norm": 0.6642473936080933,
|
| 545 |
+
"learning_rate": 7.732634338138926e-06,
|
| 546 |
+
"loss": 0.0012,
|
| 547 |
+
"step": 730
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 3.12,
|
| 551 |
+
"grad_norm": 0.0030567694921046495,
|
| 552 |
+
"learning_rate": 7.07732634338139e-06,
|
| 553 |
+
"loss": 0.2103,
|
| 554 |
+
"step": 740
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 3.13,
|
| 558 |
+
"grad_norm": 0.003925285767763853,
|
| 559 |
+
"learning_rate": 6.422018348623854e-06,
|
| 560 |
+
"loss": 0.0006,
|
| 561 |
+
"step": 750
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.15,
|
| 565 |
+
"grad_norm": 0.02660321444272995,
|
| 566 |
+
"learning_rate": 5.766710353866317e-06,
|
| 567 |
+
"loss": 0.0006,
|
| 568 |
+
"step": 760
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 3.16,
|
| 572 |
+
"grad_norm": 0.1314881145954132,
|
| 573 |
+
"learning_rate": 5.1114023591087816e-06,
|
| 574 |
+
"loss": 0.0008,
|
| 575 |
+
"step": 770
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 3.17,
|
| 579 |
+
"grad_norm": 0.008924220688641071,
|
| 580 |
+
"learning_rate": 4.456094364351245e-06,
|
| 581 |
+
"loss": 0.0021,
|
| 582 |
+
"step": 780
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 3.18,
|
| 586 |
+
"grad_norm": 0.02253740094602108,
|
| 587 |
+
"learning_rate": 3.800786369593709e-06,
|
| 588 |
+
"loss": 0.0005,
|
| 589 |
+
"step": 790
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 3.19,
|
| 593 |
+
"grad_norm": 0.006063047330826521,
|
| 594 |
+
"learning_rate": 3.145478374836173e-06,
|
| 595 |
+
"loss": 0.0005,
|
| 596 |
+
"step": 800
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 3.21,
|
| 600 |
+
"grad_norm": 0.01928607001900673,
|
| 601 |
+
"learning_rate": 2.490170380078637e-06,
|
| 602 |
+
"loss": 0.0002,
|
| 603 |
+
"step": 810
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 3.22,
|
| 607 |
+
"grad_norm": 0.2391887903213501,
|
| 608 |
+
"learning_rate": 1.8348623853211011e-06,
|
| 609 |
+
"loss": 0.0005,
|
| 610 |
+
"step": 820
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 3.23,
|
| 614 |
+
"grad_norm": 0.012002573348581791,
|
| 615 |
+
"learning_rate": 1.179554390563565e-06,
|
| 616 |
+
"loss": 0.0003,
|
| 617 |
+
"step": 830
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 3.24,
|
| 621 |
+
"grad_norm": 0.009785789065063,
|
| 622 |
+
"learning_rate": 5.242463958060289e-07,
|
| 623 |
+
"loss": 0.001,
|
| 624 |
+
"step": 840
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 3.25,
|
| 628 |
+
"eval_accuracy": 0.9964912280701754,
|
| 629 |
+
"eval_loss": 0.005051769781857729,
|
| 630 |
+
"eval_runtime": 84.2043,
|
| 631 |
+
"eval_samples_per_second": 3.385,
|
| 632 |
+
"eval_steps_per_second": 1.698,
|
| 633 |
+
"step": 848
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 3.25,
|
| 637 |
+
"step": 848,
|
| 638 |
+
"total_flos": 2.1133255437440778e+18,
|
| 639 |
+
"train_loss": 0.16699378463049214,
|
| 640 |
+
"train_runtime": 1543.4976,
|
| 641 |
+
"train_samples_per_second": 1.099,
|
| 642 |
+
"train_steps_per_second": 0.549
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 3.25,
|
| 646 |
+
"eval_accuracy": 0.9887429643527205,
|
| 647 |
+
"eval_loss": 0.020199885591864586,
|
| 648 |
+
"eval_runtime": 328.4712,
|
| 649 |
+
"eval_samples_per_second": 1.623,
|
| 650 |
+
"eval_steps_per_second": 0.813,
|
| 651 |
+
"step": 848
|
| 652 |
+
}
|
| 653 |
+
],
|
| 654 |
+
"logging_steps": 10,
|
| 655 |
+
"max_steps": 848,
|
| 656 |
+
"num_input_tokens_seen": 0,
|
| 657 |
+
"num_train_epochs": 9223372036854775807,
|
| 658 |
+
"save_steps": 500,
|
| 659 |
+
"total_flos": 2.1133255437440778e+18,
|
| 660 |
+
"train_batch_size": 2,
|
| 661 |
+
"trial_name": null,
|
| 662 |
+
"trial_params": null
|
| 663 |
+
}
|