End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_cer": 0.2821031799577205,
|
| 4 |
+
"eval_loss": 0.9065462946891785,
|
| 5 |
+
"eval_runtime": 81.1677,
|
| 6 |
+
"eval_samples": 1647,
|
| 7 |
+
"eval_samples_per_second": 20.291,
|
| 8 |
+
"eval_steps_per_second": 2.538,
|
| 9 |
+
"eval_wer": 0.8799918292309263,
|
| 10 |
+
"train_loss": 4.048810622560868,
|
| 11 |
+
"train_runtime": 250.3759,
|
| 12 |
+
"train_samples": 3478,
|
| 13 |
+
"train_samples_per_second": 13.891,
|
| 14 |
+
"train_steps_per_second": 0.435
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_cer": 0.2821031799577205,
|
| 4 |
+
"eval_loss": 0.9065462946891785,
|
| 5 |
+
"eval_runtime": 81.1677,
|
| 6 |
+
"eval_samples": 1647,
|
| 7 |
+
"eval_samples_per_second": 20.291,
|
| 8 |
+
"eval_steps_per_second": 2.538,
|
| 9 |
+
"eval_wer": 0.8799918292309263
|
| 10 |
+
}
|
runs/Feb01_20-37-49_job-98a7eb1d-c6a6-4e4b-936b-3898557b5e73/events.out.tfevents.1643748701.job-98a7eb1d-c6a6-4e4b-936b-3898557b5e73.156945.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:817933007bf6d9d5a04289b8d731ef5d9910c34dd8dd8fc41742950163d448b9
|
| 3 |
+
size 398
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 4.048810622560868,
|
| 4 |
+
"train_runtime": 250.3759,
|
| 5 |
+
"train_samples": 3478,
|
| 6 |
+
"train_samples_per_second": 13.891,
|
| 7 |
+
"train_steps_per_second": 0.435
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,679 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"global_step": 109,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 0.0,
|
| 13 |
+
"loss": 14.822,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 0.0,
|
| 19 |
+
"loss": 15.2906,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 1e-06,
|
| 25 |
+
"loss": 15.9083,
|
| 26 |
+
"step": 3
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 2e-06,
|
| 31 |
+
"loss": 15.062,
|
| 32 |
+
"step": 4
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.05,
|
| 36 |
+
"learning_rate": 3e-06,
|
| 37 |
+
"loss": 14.1132,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.06,
|
| 42 |
+
"learning_rate": 4e-06,
|
| 43 |
+
"loss": 15.2949,
|
| 44 |
+
"step": 6
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06,
|
| 48 |
+
"learning_rate": 5e-06,
|
| 49 |
+
"loss": 13.8691,
|
| 50 |
+
"step": 7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.07,
|
| 54 |
+
"learning_rate": 6e-06,
|
| 55 |
+
"loss": 13.307,
|
| 56 |
+
"step": 8
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.08,
|
| 60 |
+
"learning_rate": 6e-06,
|
| 61 |
+
"loss": 12.5476,
|
| 62 |
+
"step": 9
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.09,
|
| 66 |
+
"learning_rate": 7e-06,
|
| 67 |
+
"loss": 13.2011,
|
| 68 |
+
"step": 10
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.1,
|
| 72 |
+
"learning_rate": 8e-06,
|
| 73 |
+
"loss": 12.3008,
|
| 74 |
+
"step": 11
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.11,
|
| 78 |
+
"learning_rate": 9e-06,
|
| 79 |
+
"loss": 11.5046,
|
| 80 |
+
"step": 12
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.12,
|
| 84 |
+
"learning_rate": 1e-05,
|
| 85 |
+
"loss": 10.6207,
|
| 86 |
+
"step": 13
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.13,
|
| 90 |
+
"learning_rate": 1.1e-05,
|
| 91 |
+
"loss": 10.1038,
|
| 92 |
+
"step": 14
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.14,
|
| 96 |
+
"learning_rate": 1.2e-05,
|
| 97 |
+
"loss": 8.8277,
|
| 98 |
+
"step": 15
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.15,
|
| 102 |
+
"learning_rate": 1.3e-05,
|
| 103 |
+
"loss": 7.8204,
|
| 104 |
+
"step": 16
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.16,
|
| 108 |
+
"learning_rate": 1.4e-05,
|
| 109 |
+
"loss": 6.3586,
|
| 110 |
+
"step": 17
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.17,
|
| 114 |
+
"learning_rate": 1.5e-05,
|
| 115 |
+
"loss": 5.6274,
|
| 116 |
+
"step": 18
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.17,
|
| 120 |
+
"learning_rate": 1.6e-05,
|
| 121 |
+
"loss": 5.6395,
|
| 122 |
+
"step": 19
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.18,
|
| 126 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 127 |
+
"loss": 5.0595,
|
| 128 |
+
"step": 20
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.19,
|
| 132 |
+
"learning_rate": 1.8e-05,
|
| 133 |
+
"loss": 5.2143,
|
| 134 |
+
"step": 21
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.2,
|
| 138 |
+
"learning_rate": 1.9e-05,
|
| 139 |
+
"loss": 5.5995,
|
| 140 |
+
"step": 22
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.21,
|
| 144 |
+
"learning_rate": 2e-05,
|
| 145 |
+
"loss": 4.9733,
|
| 146 |
+
"step": 23
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.22,
|
| 150 |
+
"learning_rate": 2.1000000000000002e-05,
|
| 151 |
+
"loss": 4.9453,
|
| 152 |
+
"step": 24
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.23,
|
| 156 |
+
"learning_rate": 2.2e-05,
|
| 157 |
+
"loss": 4.8826,
|
| 158 |
+
"step": 25
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.24,
|
| 162 |
+
"learning_rate": 2.3e-05,
|
| 163 |
+
"loss": 4.4724,
|
| 164 |
+
"step": 26
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.25,
|
| 168 |
+
"learning_rate": 2.4e-05,
|
| 169 |
+
"loss": 3.7938,
|
| 170 |
+
"step": 27
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.26,
|
| 174 |
+
"learning_rate": 2.5e-05,
|
| 175 |
+
"loss": 3.2653,
|
| 176 |
+
"step": 28
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.27,
|
| 180 |
+
"learning_rate": 2.6e-05,
|
| 181 |
+
"loss": 3.3843,
|
| 182 |
+
"step": 29
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.28,
|
| 186 |
+
"learning_rate": 2.7e-05,
|
| 187 |
+
"loss": 3.5122,
|
| 188 |
+
"step": 30
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.28,
|
| 192 |
+
"learning_rate": 2.8e-05,
|
| 193 |
+
"loss": 3.4207,
|
| 194 |
+
"step": 31
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.29,
|
| 198 |
+
"learning_rate": 2.9000000000000004e-05,
|
| 199 |
+
"loss": 3.4373,
|
| 200 |
+
"step": 32
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.3,
|
| 204 |
+
"learning_rate": 3e-05,
|
| 205 |
+
"loss": 3.2939,
|
| 206 |
+
"step": 33
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.31,
|
| 210 |
+
"learning_rate": 3.1e-05,
|
| 211 |
+
"loss": 3.2784,
|
| 212 |
+
"step": 34
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.32,
|
| 216 |
+
"learning_rate": 3.2e-05,
|
| 217 |
+
"loss": 3.2101,
|
| 218 |
+
"step": 35
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.33,
|
| 222 |
+
"learning_rate": 3.3e-05,
|
| 223 |
+
"loss": 3.1498,
|
| 224 |
+
"step": 36
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.34,
|
| 228 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 229 |
+
"loss": 3.2037,
|
| 230 |
+
"step": 37
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.35,
|
| 234 |
+
"learning_rate": 3.5000000000000004e-05,
|
| 235 |
+
"loss": 3.1997,
|
| 236 |
+
"step": 38
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.36,
|
| 240 |
+
"learning_rate": 3.6e-05,
|
| 241 |
+
"loss": 3.1802,
|
| 242 |
+
"step": 39
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.37,
|
| 246 |
+
"learning_rate": 3.7e-05,
|
| 247 |
+
"loss": 3.1123,
|
| 248 |
+
"step": 40
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.38,
|
| 252 |
+
"learning_rate": 3.8e-05,
|
| 253 |
+
"loss": 3.1104,
|
| 254 |
+
"step": 41
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.39,
|
| 258 |
+
"learning_rate": 3.9e-05,
|
| 259 |
+
"loss": 3.1473,
|
| 260 |
+
"step": 42
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.39,
|
| 264 |
+
"learning_rate": 4e-05,
|
| 265 |
+
"loss": 3.1615,
|
| 266 |
+
"step": 43
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.4,
|
| 270 |
+
"learning_rate": 4.1e-05,
|
| 271 |
+
"loss": 3.1428,
|
| 272 |
+
"step": 44
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.41,
|
| 276 |
+
"learning_rate": 4.2000000000000004e-05,
|
| 277 |
+
"loss": 3.0819,
|
| 278 |
+
"step": 45
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.42,
|
| 282 |
+
"learning_rate": 4.2999999999999995e-05,
|
| 283 |
+
"loss": 3.1113,
|
| 284 |
+
"step": 46
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.43,
|
| 288 |
+
"learning_rate": 4.4e-05,
|
| 289 |
+
"loss": 3.0946,
|
| 290 |
+
"step": 47
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.44,
|
| 294 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 295 |
+
"loss": 3.0418,
|
| 296 |
+
"step": 48
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.45,
|
| 300 |
+
"learning_rate": 4.6e-05,
|
| 301 |
+
"loss": 3.0446,
|
| 302 |
+
"step": 49
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.46,
|
| 306 |
+
"learning_rate": 4.7000000000000004e-05,
|
| 307 |
+
"loss": 3.0657,
|
| 308 |
+
"step": 50
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.47,
|
| 312 |
+
"learning_rate": 4.8e-05,
|
| 313 |
+
"loss": 3.079,
|
| 314 |
+
"step": 51
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.48,
|
| 318 |
+
"learning_rate": 4.9000000000000005e-05,
|
| 319 |
+
"loss": 3.0337,
|
| 320 |
+
"step": 52
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.49,
|
| 324 |
+
"learning_rate": 5e-05,
|
| 325 |
+
"loss": 2.9961,
|
| 326 |
+
"step": 53
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.5,
|
| 330 |
+
"learning_rate": 5.1e-05,
|
| 331 |
+
"loss": 3.0675,
|
| 332 |
+
"step": 54
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5,
|
| 336 |
+
"learning_rate": 5.2e-05,
|
| 337 |
+
"loss": 3.2045,
|
| 338 |
+
"step": 55
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.51,
|
| 342 |
+
"learning_rate": 5.3e-05,
|
| 343 |
+
"loss": 3.1336,
|
| 344 |
+
"step": 56
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.52,
|
| 348 |
+
"learning_rate": 5.4e-05,
|
| 349 |
+
"loss": 2.8979,
|
| 350 |
+
"step": 57
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.53,
|
| 354 |
+
"learning_rate": 5.5e-05,
|
| 355 |
+
"loss": 3.0432,
|
| 356 |
+
"step": 58
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.54,
|
| 360 |
+
"learning_rate": 5.6e-05,
|
| 361 |
+
"loss": 3.1173,
|
| 362 |
+
"step": 59
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.55,
|
| 366 |
+
"learning_rate": 5.7e-05,
|
| 367 |
+
"loss": 2.8984,
|
| 368 |
+
"step": 60
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.56,
|
| 372 |
+
"learning_rate": 5.800000000000001e-05,
|
| 373 |
+
"loss": 3.0313,
|
| 374 |
+
"step": 61
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.57,
|
| 378 |
+
"learning_rate": 5.9e-05,
|
| 379 |
+
"loss": 3.0274,
|
| 380 |
+
"step": 62
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.58,
|
| 384 |
+
"learning_rate": 6e-05,
|
| 385 |
+
"loss": 2.8286,
|
| 386 |
+
"step": 63
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.59,
|
| 390 |
+
"learning_rate": 6.1e-05,
|
| 391 |
+
"loss": 2.7637,
|
| 392 |
+
"step": 64
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.6,
|
| 396 |
+
"learning_rate": 6.2e-05,
|
| 397 |
+
"loss": 2.7111,
|
| 398 |
+
"step": 65
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.61,
|
| 402 |
+
"learning_rate": 6.3e-05,
|
| 403 |
+
"loss": 2.6568,
|
| 404 |
+
"step": 66
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.61,
|
| 408 |
+
"learning_rate": 6.4e-05,
|
| 409 |
+
"loss": 2.5761,
|
| 410 |
+
"step": 67
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.62,
|
| 414 |
+
"learning_rate": 6.500000000000001e-05,
|
| 415 |
+
"loss": 2.594,
|
| 416 |
+
"step": 68
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.63,
|
| 420 |
+
"learning_rate": 6.6e-05,
|
| 421 |
+
"loss": 2.5337,
|
| 422 |
+
"step": 69
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.64,
|
| 426 |
+
"learning_rate": 6.7e-05,
|
| 427 |
+
"loss": 2.328,
|
| 428 |
+
"step": 70
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.65,
|
| 432 |
+
"learning_rate": 6.800000000000001e-05,
|
| 433 |
+
"loss": 2.2416,
|
| 434 |
+
"step": 71
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.66,
|
| 438 |
+
"learning_rate": 6.900000000000001e-05,
|
| 439 |
+
"loss": 2.1625,
|
| 440 |
+
"step": 72
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.67,
|
| 444 |
+
"learning_rate": 7.000000000000001e-05,
|
| 445 |
+
"loss": 2.0447,
|
| 446 |
+
"step": 73
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.68,
|
| 450 |
+
"learning_rate": 7.099999999999999e-05,
|
| 451 |
+
"loss": 1.8691,
|
| 452 |
+
"step": 74
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.69,
|
| 456 |
+
"learning_rate": 7.2e-05,
|
| 457 |
+
"loss": 1.8868,
|
| 458 |
+
"step": 75
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.7,
|
| 462 |
+
"learning_rate": 7.3e-05,
|
| 463 |
+
"loss": 1.6314,
|
| 464 |
+
"step": 76
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.71,
|
| 468 |
+
"learning_rate": 7.4e-05,
|
| 469 |
+
"loss": 1.6226,
|
| 470 |
+
"step": 77
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.72,
|
| 474 |
+
"learning_rate": 7.5e-05,
|
| 475 |
+
"loss": 1.7013,
|
| 476 |
+
"step": 78
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.72,
|
| 480 |
+
"learning_rate": 7.6e-05,
|
| 481 |
+
"loss": 1.5557,
|
| 482 |
+
"step": 79
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.73,
|
| 486 |
+
"learning_rate": 7.7e-05,
|
| 487 |
+
"loss": 1.4843,
|
| 488 |
+
"step": 80
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.74,
|
| 492 |
+
"learning_rate": 7.8e-05,
|
| 493 |
+
"loss": 1.563,
|
| 494 |
+
"step": 81
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.75,
|
| 498 |
+
"learning_rate": 7.9e-05,
|
| 499 |
+
"loss": 1.3356,
|
| 500 |
+
"step": 82
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.76,
|
| 504 |
+
"learning_rate": 8e-05,
|
| 505 |
+
"loss": 1.3477,
|
| 506 |
+
"step": 83
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.77,
|
| 510 |
+
"learning_rate": 8.1e-05,
|
| 511 |
+
"loss": 1.2732,
|
| 512 |
+
"step": 84
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.78,
|
| 516 |
+
"learning_rate": 8.2e-05,
|
| 517 |
+
"loss": 1.1838,
|
| 518 |
+
"step": 85
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.79,
|
| 522 |
+
"learning_rate": 8.300000000000001e-05,
|
| 523 |
+
"loss": 1.1471,
|
| 524 |
+
"step": 86
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.8,
|
| 528 |
+
"learning_rate": 8.400000000000001e-05,
|
| 529 |
+
"loss": 1.0627,
|
| 530 |
+
"step": 87
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.81,
|
| 534 |
+
"learning_rate": 8.5e-05,
|
| 535 |
+
"loss": 0.9847,
|
| 536 |
+
"step": 88
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.82,
|
| 540 |
+
"learning_rate": 8.599999999999999e-05,
|
| 541 |
+
"loss": 0.9781,
|
| 542 |
+
"step": 89
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.83,
|
| 546 |
+
"learning_rate": 8.7e-05,
|
| 547 |
+
"loss": 0.96,
|
| 548 |
+
"step": 90
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.83,
|
| 552 |
+
"learning_rate": 8.8e-05,
|
| 553 |
+
"loss": 0.9061,
|
| 554 |
+
"step": 91
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.84,
|
| 558 |
+
"learning_rate": 8.9e-05,
|
| 559 |
+
"loss": 0.9697,
|
| 560 |
+
"step": 92
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.85,
|
| 564 |
+
"learning_rate": 8.999999999999999e-05,
|
| 565 |
+
"loss": 1.0049,
|
| 566 |
+
"step": 93
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.86,
|
| 570 |
+
"learning_rate": 9.1e-05,
|
| 571 |
+
"loss": 1.1034,
|
| 572 |
+
"step": 94
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.87,
|
| 576 |
+
"learning_rate": 9.2e-05,
|
| 577 |
+
"loss": 0.9069,
|
| 578 |
+
"step": 95
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.88,
|
| 582 |
+
"learning_rate": 9.3e-05,
|
| 583 |
+
"loss": 1.026,
|
| 584 |
+
"step": 96
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.89,
|
| 588 |
+
"learning_rate": 9.400000000000001e-05,
|
| 589 |
+
"loss": 1.0734,
|
| 590 |
+
"step": 97
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.9,
|
| 594 |
+
"learning_rate": 9.5e-05,
|
| 595 |
+
"loss": 0.9626,
|
| 596 |
+
"step": 98
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.91,
|
| 600 |
+
"learning_rate": 9.6e-05,
|
| 601 |
+
"loss": 0.8731,
|
| 602 |
+
"step": 99
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.92,
|
| 606 |
+
"learning_rate": 9.7e-05,
|
| 607 |
+
"loss": 1.0878,
|
| 608 |
+
"step": 100
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.93,
|
| 612 |
+
"learning_rate": 9.800000000000001e-05,
|
| 613 |
+
"loss": 0.9423,
|
| 614 |
+
"step": 101
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.94,
|
| 618 |
+
"learning_rate": 9.900000000000001e-05,
|
| 619 |
+
"loss": 0.9755,
|
| 620 |
+
"step": 102
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.94,
|
| 624 |
+
"learning_rate": 0.0001,
|
| 625 |
+
"loss": 0.9086,
|
| 626 |
+
"step": 103
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.95,
|
| 630 |
+
"learning_rate": 0.000101,
|
| 631 |
+
"loss": 0.8551,
|
| 632 |
+
"step": 104
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.96,
|
| 636 |
+
"learning_rate": 0.000102,
|
| 637 |
+
"loss": 1.0771,
|
| 638 |
+
"step": 105
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.97,
|
| 642 |
+
"learning_rate": 0.000103,
|
| 643 |
+
"loss": 0.9983,
|
| 644 |
+
"step": 106
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.98,
|
| 648 |
+
"learning_rate": 0.000104,
|
| 649 |
+
"loss": 0.9869,
|
| 650 |
+
"step": 107
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.99,
|
| 654 |
+
"learning_rate": 0.000105,
|
| 655 |
+
"loss": 1.1931,
|
| 656 |
+
"step": 108
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 1.0,
|
| 660 |
+
"learning_rate": 0.000106,
|
| 661 |
+
"loss": 1.1017,
|
| 662 |
+
"step": 109
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.0,
|
| 666 |
+
"step": 109,
|
| 667 |
+
"total_flos": 1.3366257233080689e+18,
|
| 668 |
+
"train_loss": 4.048810622560868,
|
| 669 |
+
"train_runtime": 250.3759,
|
| 670 |
+
"train_samples_per_second": 13.891,
|
| 671 |
+
"train_steps_per_second": 0.435
|
| 672 |
+
}
|
| 673 |
+
],
|
| 674 |
+
"max_steps": 109,
|
| 675 |
+
"num_train_epochs": 1,
|
| 676 |
+
"total_flos": 1.3366257233080689e+18,
|
| 677 |
+
"trial_name": null,
|
| 678 |
+
"trial_params": null
|
| 679 |
+
}
|