End of training
Browse files
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy":
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples":
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples":
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second":
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_accuracy": 0.9067641496658325,
|
| 4 |
+
"eval_loss": 0.6223405599594116,
|
| 5 |
+
"eval_runtime": 3.1215,
|
| 6 |
+
"eval_samples": 1094,
|
| 7 |
+
"eval_samples_per_second": 350.471,
|
| 8 |
+
"eval_steps_per_second": 43.889,
|
| 9 |
+
"train_loss": 0.4635289347472817,
|
| 10 |
+
"train_runtime": 2578.0344,
|
| 11 |
+
"train_samples": 10556,
|
| 12 |
+
"train_samples_per_second": 61.419,
|
| 13 |
+
"train_steps_per_second": 7.68
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy":
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples":
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_accuracy": 0.9067641496658325,
|
| 4 |
+
"eval_loss": 0.6223405599594116,
|
| 5 |
+
"eval_runtime": 3.1215,
|
| 6 |
+
"eval_samples": 1094,
|
| 7 |
+
"eval_samples_per_second": 350.471,
|
| 8 |
+
"eval_steps_per_second": 43.889
|
| 9 |
}
|
runs/Dec07_08-55-12_jbuhmann/events.out.tfevents.1638867383.jbuhmann.18249.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b71c97e16bb144e50a66f88b517b7f36fe7c3df6e9c8a9f4aac04cb2338ecaab
|
| 3 |
+
size 27716
|
runs/Dec07_08-55-12_jbuhmann/events.out.tfevents.1638869966.jbuhmann.18249.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1651e9deba0013bed58761eaad659750223df9783f0487f735b82976eb86c5e
|
| 3 |
+
size 369
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 0.4635289347472817,
|
| 4 |
+
"train_runtime": 2578.0344,
|
| 5 |
+
"train_samples": 10556,
|
| 6 |
+
"train_samples_per_second": 61.419,
|
| 7 |
+
"train_steps_per_second": 7.68
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,100 +1,394 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 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.
|
| 12 |
-
"learning_rate": 1.
|
| 13 |
-
"loss": 4.
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate": 1.
|
| 19 |
-
"loss": 3.
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 1.0,
|
| 24 |
-
"eval_accuracy": 0.
|
| 25 |
-
"eval_loss": 2.
|
| 26 |
-
"eval_runtime":
|
| 27 |
-
"eval_samples_per_second":
|
| 28 |
-
"eval_steps_per_second":
|
| 29 |
-
"step":
|
| 30 |
},
|
| 31 |
{
|
| 32 |
-
"epoch": 1.
|
| 33 |
-
"learning_rate": 1.
|
| 34 |
-
"loss": 2.
|
| 35 |
"step": 1500
|
| 36 |
},
|
| 37 |
{
|
| 38 |
-
"epoch": 1.
|
| 39 |
-
"learning_rate": 1.
|
| 40 |
-
"loss":
|
| 41 |
"step": 2000
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch": 1.
|
| 45 |
-
"learning_rate":
|
| 46 |
-
"loss": 1.
|
| 47 |
"step": 2500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 2.0,
|
| 51 |
-
"eval_accuracy":
|
| 52 |
-
"eval_loss":
|
| 53 |
-
"eval_runtime":
|
| 54 |
-
"eval_samples_per_second":
|
| 55 |
-
"eval_steps_per_second":
|
| 56 |
-
"step":
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 2.
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss": 1.
|
| 62 |
"step": 3000
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 2.
|
| 66 |
-
"learning_rate":
|
| 67 |
-
"loss":
|
| 68 |
"step": 3500
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch":
|
| 72 |
-
"
|
| 73 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
"step": 4000
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 3.
|
| 78 |
-
"
|
| 79 |
-
"
|
| 80 |
-
"
|
| 81 |
-
"eval_samples_per_second": 326.232,
|
| 82 |
-
"eval_steps_per_second": 65.246,
|
| 83 |
-
"step": 4371
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch": 3.
|
| 87 |
-
"
|
| 88 |
-
"
|
| 89 |
-
"
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
}
|
| 94 |
],
|
| 95 |
-
"max_steps":
|
| 96 |
-
"num_train_epochs":
|
| 97 |
-
"total_flos":
|
| 98 |
"trial_name": null,
|
| 99 |
"trial_params": null
|
| 100 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"global_step": 19800,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.38,
|
| 12 |
+
"learning_rate": 1.9494949494949496e-05,
|
| 13 |
+
"loss": 4.4759,
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.76,
|
| 18 |
+
"learning_rate": 1.8989898989898993e-05,
|
| 19 |
+
"loss": 3.4666,
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 1.0,
|
| 24 |
+
"eval_accuracy": 0.5767824649810791,
|
| 25 |
+
"eval_loss": 2.3355202674865723,
|
| 26 |
+
"eval_runtime": 2.6806,
|
| 27 |
+
"eval_samples_per_second": 408.112,
|
| 28 |
+
"eval_steps_per_second": 51.107,
|
| 29 |
+
"step": 1320
|
| 30 |
},
|
| 31 |
{
|
| 32 |
+
"epoch": 1.14,
|
| 33 |
+
"learning_rate": 1.8484848484848487e-05,
|
| 34 |
+
"loss": 2.6641,
|
| 35 |
"step": 1500
|
| 36 |
},
|
| 37 |
{
|
| 38 |
+
"epoch": 1.52,
|
| 39 |
+
"learning_rate": 1.797979797979798e-05,
|
| 40 |
+
"loss": 1.9312,
|
| 41 |
"step": 2000
|
| 42 |
},
|
| 43 |
{
|
| 44 |
+
"epoch": 1.89,
|
| 45 |
+
"learning_rate": 1.7474747474747475e-05,
|
| 46 |
+
"loss": 1.5293,
|
| 47 |
"step": 2500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 2.0,
|
| 51 |
+
"eval_accuracy": 0.8144423961639404,
|
| 52 |
+
"eval_loss": 1.1118409633636475,
|
| 53 |
+
"eval_runtime": 2.7222,
|
| 54 |
+
"eval_samples_per_second": 401.874,
|
| 55 |
+
"eval_steps_per_second": 50.326,
|
| 56 |
+
"step": 2640
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 2.27,
|
| 60 |
+
"learning_rate": 1.6969696969696972e-05,
|
| 61 |
+
"loss": 1.0511,
|
| 62 |
"step": 3000
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 2.65,
|
| 66 |
+
"learning_rate": 1.6464646464646466e-05,
|
| 67 |
+
"loss": 0.8031,
|
| 68 |
"step": 3500
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 3.0,
|
| 72 |
+
"eval_accuracy": 0.880255937576294,
|
| 73 |
+
"eval_loss": 0.6362389922142029,
|
| 74 |
+
"eval_runtime": 2.657,
|
| 75 |
+
"eval_samples_per_second": 411.738,
|
| 76 |
+
"eval_steps_per_second": 51.561,
|
| 77 |
+
"step": 3960
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 3.03,
|
| 81 |
+
"learning_rate": 1.595959595959596e-05,
|
| 82 |
+
"loss": 0.5864,
|
| 83 |
"step": 4000
|
| 84 |
},
|
| 85 |
{
|
| 86 |
+
"epoch": 3.41,
|
| 87 |
+
"learning_rate": 1.5454545454545454e-05,
|
| 88 |
+
"loss": 0.3632,
|
| 89 |
+
"step": 4500
|
|
|
|
|
|
|
|
|
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 3.79,
|
| 93 |
+
"learning_rate": 1.4949494949494952e-05,
|
| 94 |
+
"loss": 0.2985,
|
| 95 |
+
"step": 5000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 4.0,
|
| 99 |
+
"eval_accuracy": 0.8957952260971069,
|
| 100 |
+
"eval_loss": 0.5118863582611084,
|
| 101 |
+
"eval_runtime": 2.6864,
|
| 102 |
+
"eval_samples_per_second": 407.231,
|
| 103 |
+
"eval_steps_per_second": 50.997,
|
| 104 |
+
"step": 5280
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.17,
|
| 108 |
+
"learning_rate": 1.4444444444444446e-05,
|
| 109 |
+
"loss": 0.2188,
|
| 110 |
+
"step": 5500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.55,
|
| 114 |
+
"learning_rate": 1.3939393939393942e-05,
|
| 115 |
+
"loss": 0.1478,
|
| 116 |
+
"step": 6000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.92,
|
| 120 |
+
"learning_rate": 1.3434343434343436e-05,
|
| 121 |
+
"loss": 0.1284,
|
| 122 |
+
"step": 6500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 5.0,
|
| 126 |
+
"eval_accuracy": 0.8930529952049255,
|
| 127 |
+
"eval_loss": 0.5023291707038879,
|
| 128 |
+
"eval_runtime": 2.7918,
|
| 129 |
+
"eval_samples_per_second": 391.858,
|
| 130 |
+
"eval_steps_per_second": 49.072,
|
| 131 |
+
"step": 6600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.3,
|
| 135 |
+
"learning_rate": 1.2929292929292931e-05,
|
| 136 |
+
"loss": 0.0608,
|
| 137 |
+
"step": 7000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 5.68,
|
| 141 |
+
"learning_rate": 1.2424242424242425e-05,
|
| 142 |
+
"loss": 0.0842,
|
| 143 |
+
"step": 7500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.0,
|
| 147 |
+
"eval_accuracy": 0.9021937847137451,
|
| 148 |
+
"eval_loss": 0.5246109962463379,
|
| 149 |
+
"eval_runtime": 2.6464,
|
| 150 |
+
"eval_samples_per_second": 413.391,
|
| 151 |
+
"eval_steps_per_second": 51.768,
|
| 152 |
+
"step": 7920
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 6.06,
|
| 156 |
+
"learning_rate": 1.191919191919192e-05,
|
| 157 |
+
"loss": 0.064,
|
| 158 |
+
"step": 8000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 6.44,
|
| 162 |
+
"learning_rate": 1.1414141414141415e-05,
|
| 163 |
+
"loss": 0.0461,
|
| 164 |
+
"step": 8500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 6.82,
|
| 168 |
+
"learning_rate": 1.0909090909090909e-05,
|
| 169 |
+
"loss": 0.0414,
|
| 170 |
+
"step": 9000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 7.0,
|
| 174 |
+
"eval_accuracy": 0.9012796878814697,
|
| 175 |
+
"eval_loss": 0.5580916404724121,
|
| 176 |
+
"eval_runtime": 2.6062,
|
| 177 |
+
"eval_samples_per_second": 419.767,
|
| 178 |
+
"eval_steps_per_second": 52.567,
|
| 179 |
+
"step": 9240
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 7.2,
|
| 183 |
+
"learning_rate": 1.0404040404040405e-05,
|
| 184 |
+
"loss": 0.0476,
|
| 185 |
+
"step": 9500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.58,
|
| 189 |
+
"learning_rate": 9.8989898989899e-06,
|
| 190 |
+
"loss": 0.0333,
|
| 191 |
+
"step": 10000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.95,
|
| 195 |
+
"learning_rate": 9.393939393939396e-06,
|
| 196 |
+
"loss": 0.0372,
|
| 197 |
+
"step": 10500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 8.0,
|
| 201 |
+
"eval_accuracy": 0.9003656506538391,
|
| 202 |
+
"eval_loss": 0.5721055269241333,
|
| 203 |
+
"eval_runtime": 3.0748,
|
| 204 |
+
"eval_samples_per_second": 355.794,
|
| 205 |
+
"eval_steps_per_second": 44.556,
|
| 206 |
+
"step": 10560
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 8.33,
|
| 210 |
+
"learning_rate": 8.888888888888888e-06,
|
| 211 |
+
"loss": 0.0276,
|
| 212 |
+
"step": 11000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 8.71,
|
| 216 |
+
"learning_rate": 8.383838383838384e-06,
|
| 217 |
+
"loss": 0.0292,
|
| 218 |
+
"step": 11500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.0,
|
| 222 |
+
"eval_accuracy": 0.9140768051147461,
|
| 223 |
+
"eval_loss": 0.5468941926956177,
|
| 224 |
+
"eval_runtime": 3.0391,
|
| 225 |
+
"eval_samples_per_second": 359.977,
|
| 226 |
+
"eval_steps_per_second": 45.079,
|
| 227 |
+
"step": 11880
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 9.09,
|
| 231 |
+
"learning_rate": 7.87878787878788e-06,
|
| 232 |
+
"loss": 0.0298,
|
| 233 |
+
"step": 12000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 9.47,
|
| 237 |
+
"learning_rate": 7.373737373737374e-06,
|
| 238 |
+
"loss": 0.0209,
|
| 239 |
+
"step": 12500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 9.85,
|
| 243 |
+
"learning_rate": 6.868686868686869e-06,
|
| 244 |
+
"loss": 0.0257,
|
| 245 |
+
"step": 13000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 10.0,
|
| 249 |
+
"eval_accuracy": 0.9058501124382019,
|
| 250 |
+
"eval_loss": 0.5871404409408569,
|
| 251 |
+
"eval_runtime": 2.8242,
|
| 252 |
+
"eval_samples_per_second": 387.373,
|
| 253 |
+
"eval_steps_per_second": 48.51,
|
| 254 |
+
"step": 13200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 10.23,
|
| 258 |
+
"learning_rate": 6.363636363636364e-06,
|
| 259 |
+
"loss": 0.0205,
|
| 260 |
+
"step": 13500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 10.61,
|
| 264 |
+
"learning_rate": 5.858585858585859e-06,
|
| 265 |
+
"loss": 0.0145,
|
| 266 |
+
"step": 14000
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 10.98,
|
| 270 |
+
"learning_rate": 5.353535353535354e-06,
|
| 271 |
+
"loss": 0.0189,
|
| 272 |
+
"step": 14500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 11.0,
|
| 276 |
+
"eval_accuracy": 0.9049360156059265,
|
| 277 |
+
"eval_loss": 0.6180957555770874,
|
| 278 |
+
"eval_runtime": 2.9518,
|
| 279 |
+
"eval_samples_per_second": 370.627,
|
| 280 |
+
"eval_steps_per_second": 46.413,
|
| 281 |
+
"step": 14520
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 11.36,
|
| 285 |
+
"learning_rate": 4.848484848484849e-06,
|
| 286 |
+
"loss": 0.0117,
|
| 287 |
+
"step": 15000
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 11.74,
|
| 291 |
+
"learning_rate": 4.343434343434344e-06,
|
| 292 |
+
"loss": 0.0104,
|
| 293 |
+
"step": 15500
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 12.0,
|
| 297 |
+
"eval_accuracy": 0.9067641496658325,
|
| 298 |
+
"eval_loss": 0.618419349193573,
|
| 299 |
+
"eval_runtime": 2.8761,
|
| 300 |
+
"eval_samples_per_second": 380.381,
|
| 301 |
+
"eval_steps_per_second": 47.635,
|
| 302 |
+
"step": 15840
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 12.12,
|
| 306 |
+
"learning_rate": 3.8383838383838385e-06,
|
| 307 |
+
"loss": 0.0155,
|
| 308 |
+
"step": 16000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 12.5,
|
| 312 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 313 |
+
"loss": 0.0087,
|
| 314 |
+
"step": 16500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 12.88,
|
| 318 |
+
"learning_rate": 2.8282828282828286e-06,
|
| 319 |
+
"loss": 0.009,
|
| 320 |
+
"step": 17000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 13.0,
|
| 324 |
+
"eval_accuracy": 0.9049360156059265,
|
| 325 |
+
"eval_loss": 0.6013244986534119,
|
| 326 |
+
"eval_runtime": 2.7727,
|
| 327 |
+
"eval_samples_per_second": 394.554,
|
| 328 |
+
"eval_steps_per_second": 49.409,
|
| 329 |
+
"step": 17160
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.26,
|
| 333 |
+
"learning_rate": 2.3232323232323234e-06,
|
| 334 |
+
"loss": 0.0078,
|
| 335 |
+
"step": 17500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 13.64,
|
| 339 |
+
"learning_rate": 1.8181818181818183e-06,
|
| 340 |
+
"loss": 0.0051,
|
| 341 |
+
"step": 18000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 14.0,
|
| 345 |
+
"eval_accuracy": 0.9058501124382019,
|
| 346 |
+
"eval_loss": 0.620483934879303,
|
| 347 |
+
"eval_runtime": 3.0663,
|
| 348 |
+
"eval_samples_per_second": 356.786,
|
| 349 |
+
"eval_steps_per_second": 44.68,
|
| 350 |
+
"step": 18480
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 14.02,
|
| 354 |
+
"learning_rate": 1.3131313131313134e-06,
|
| 355 |
+
"loss": 0.0091,
|
| 356 |
+
"step": 18500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 14.39,
|
| 360 |
+
"learning_rate": 8.080808080808082e-07,
|
| 361 |
+
"loss": 0.0046,
|
| 362 |
+
"step": 19000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 14.77,
|
| 366 |
+
"learning_rate": 3.0303030303030305e-07,
|
| 367 |
+
"loss": 0.0035,
|
| 368 |
+
"step": 19500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 15.0,
|
| 372 |
+
"eval_accuracy": 0.9067641496658325,
|
| 373 |
+
"eval_loss": 0.6223405599594116,
|
| 374 |
+
"eval_runtime": 3.0619,
|
| 375 |
+
"eval_samples_per_second": 357.298,
|
| 376 |
+
"eval_steps_per_second": 44.744,
|
| 377 |
+
"step": 19800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 15.0,
|
| 381 |
+
"step": 19800,
|
| 382 |
+
"total_flos": 5215995096399360.0,
|
| 383 |
+
"train_loss": 0.4635289347472817,
|
| 384 |
+
"train_runtime": 2578.0344,
|
| 385 |
+
"train_samples_per_second": 61.419,
|
| 386 |
+
"train_steps_per_second": 7.68
|
| 387 |
}
|
| 388 |
],
|
| 389 |
+
"max_steps": 19800,
|
| 390 |
+
"num_train_epochs": 15,
|
| 391 |
+
"total_flos": 5215995096399360.0,
|
| 392 |
"trial_name": null,
|
| 393 |
"trial_params": null
|
| 394 |
}
|