added logs
Browse files- all_results.json +8 -0
- logs/byt5-small-cmudict/log.json +0 -0
- logs/byt5-small-cmudict/metrics.txt +1 -0
- train_results.json +8 -0
- trainer_state.json +185 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.19193305201784394,
|
| 4 |
+
"train_runtime": 3656.7872,
|
| 5 |
+
"train_samples": 108132,
|
| 6 |
+
"train_samples_per_second": 295.702,
|
| 7 |
+
"train_steps_per_second": 2.311
|
| 8 |
+
}
|
logs/byt5-small-cmudict/log.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
logs/byt5-small-cmudict/metrics.txt
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
PER: 0.0665
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.19193305201784394,
|
| 4 |
+
"train_runtime": 3656.7872,
|
| 5 |
+
"train_samples": 108132,
|
| 6 |
+
"train_samples_per_second": 295.702,
|
| 7 |
+
"train_steps_per_second": 2.311
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,185 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 8450,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"learning_rate": 0.0002,
|
| 13 |
+
"loss": 1.1569,
|
| 14 |
+
"step": 845
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_gen_len": 14.5711,
|
| 19 |
+
"eval_loss": 0.14331915974617004,
|
| 20 |
+
"eval_per": 0.1751,
|
| 21 |
+
"eval_runtime": 93.1351,
|
| 22 |
+
"eval_samples_per_second": 58.485,
|
| 23 |
+
"eval_steps_per_second": 1.836,
|
| 24 |
+
"step": 845
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 2.0,
|
| 28 |
+
"learning_rate": 0.00017777777777777779,
|
| 29 |
+
"loss": 0.1409,
|
| 30 |
+
"step": 1690
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 2.0,
|
| 34 |
+
"eval_gen_len": 14.5844,
|
| 35 |
+
"eval_loss": 0.10491351783275604,
|
| 36 |
+
"eval_per": 0.1383,
|
| 37 |
+
"eval_runtime": 92.5515,
|
| 38 |
+
"eval_samples_per_second": 58.854,
|
| 39 |
+
"eval_steps_per_second": 1.848,
|
| 40 |
+
"step": 1690
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 3.0,
|
| 44 |
+
"learning_rate": 0.00015555555555555556,
|
| 45 |
+
"loss": 0.1073,
|
| 46 |
+
"step": 2535
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 3.0,
|
| 50 |
+
"eval_gen_len": 14.6058,
|
| 51 |
+
"eval_loss": 0.09005340933799744,
|
| 52 |
+
"eval_per": 0.1281,
|
| 53 |
+
"eval_runtime": 91.8497,
|
| 54 |
+
"eval_samples_per_second": 59.303,
|
| 55 |
+
"eval_steps_per_second": 1.862,
|
| 56 |
+
"step": 2535
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.0,
|
| 60 |
+
"learning_rate": 0.00013333333333333334,
|
| 61 |
+
"loss": 0.0927,
|
| 62 |
+
"step": 3380
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 4.0,
|
| 66 |
+
"eval_gen_len": 14.5798,
|
| 67 |
+
"eval_loss": 0.08551742881536484,
|
| 68 |
+
"eval_per": 0.1237,
|
| 69 |
+
"eval_runtime": 92.071,
|
| 70 |
+
"eval_samples_per_second": 59.161,
|
| 71 |
+
"eval_steps_per_second": 1.857,
|
| 72 |
+
"step": 3380
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 5.0,
|
| 76 |
+
"learning_rate": 0.00011111111111111112,
|
| 77 |
+
"loss": 0.0833,
|
| 78 |
+
"step": 4225
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 5.0,
|
| 82 |
+
"eval_gen_len": 14.6139,
|
| 83 |
+
"eval_loss": 0.08394311368465424,
|
| 84 |
+
"eval_per": 0.1207,
|
| 85 |
+
"eval_runtime": 92.0384,
|
| 86 |
+
"eval_samples_per_second": 59.182,
|
| 87 |
+
"eval_steps_per_second": 1.858,
|
| 88 |
+
"step": 4225
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 6.0,
|
| 92 |
+
"learning_rate": 8.888888888888889e-05,
|
| 93 |
+
"loss": 0.0762,
|
| 94 |
+
"step": 5070
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 6.0,
|
| 98 |
+
"eval_gen_len": 14.5913,
|
| 99 |
+
"eval_loss": 0.08085508644580841,
|
| 100 |
+
"eval_per": 0.118,
|
| 101 |
+
"eval_runtime": 91.0725,
|
| 102 |
+
"eval_samples_per_second": 59.809,
|
| 103 |
+
"eval_steps_per_second": 1.878,
|
| 104 |
+
"step": 5070
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 7.0,
|
| 108 |
+
"learning_rate": 6.666666666666667e-05,
|
| 109 |
+
"loss": 0.0707,
|
| 110 |
+
"step": 5915
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 7.0,
|
| 114 |
+
"eval_gen_len": 14.6055,
|
| 115 |
+
"eval_loss": 0.08100058883428574,
|
| 116 |
+
"eval_per": 0.1171,
|
| 117 |
+
"eval_runtime": 90.5081,
|
| 118 |
+
"eval_samples_per_second": 60.182,
|
| 119 |
+
"eval_steps_per_second": 1.889,
|
| 120 |
+
"step": 5915
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 8.0,
|
| 124 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 125 |
+
"loss": 0.0667,
|
| 126 |
+
"step": 6760
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 8.0,
|
| 130 |
+
"eval_gen_len": 14.6025,
|
| 131 |
+
"eval_loss": 0.07957806438207626,
|
| 132 |
+
"eval_per": 0.1166,
|
| 133 |
+
"eval_runtime": 91.0929,
|
| 134 |
+
"eval_samples_per_second": 59.796,
|
| 135 |
+
"eval_steps_per_second": 1.877,
|
| 136 |
+
"step": 6760
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 9.0,
|
| 140 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 141 |
+
"loss": 0.0638,
|
| 142 |
+
"step": 7605
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 9.0,
|
| 146 |
+
"eval_gen_len": 14.6115,
|
| 147 |
+
"eval_loss": 0.07928925007581711,
|
| 148 |
+
"eval_per": 0.115,
|
| 149 |
+
"eval_runtime": 90.6785,
|
| 150 |
+
"eval_samples_per_second": 60.069,
|
| 151 |
+
"eval_steps_per_second": 1.886,
|
| 152 |
+
"step": 7605
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 10.0,
|
| 156 |
+
"learning_rate": 0.0,
|
| 157 |
+
"loss": 0.0609,
|
| 158 |
+
"step": 8450
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 10.0,
|
| 162 |
+
"eval_gen_len": 14.6046,
|
| 163 |
+
"eval_loss": 0.0802011638879776,
|
| 164 |
+
"eval_per": 0.1151,
|
| 165 |
+
"eval_runtime": 90.409,
|
| 166 |
+
"eval_samples_per_second": 60.248,
|
| 167 |
+
"eval_steps_per_second": 1.891,
|
| 168 |
+
"step": 8450
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 10.0,
|
| 172 |
+
"step": 8450,
|
| 173 |
+
"total_flos": 1.2418278660440064e+17,
|
| 174 |
+
"train_loss": 0.19193305201784394,
|
| 175 |
+
"train_runtime": 3656.7872,
|
| 176 |
+
"train_samples_per_second": 295.702,
|
| 177 |
+
"train_steps_per_second": 2.311
|
| 178 |
+
}
|
| 179 |
+
],
|
| 180 |
+
"max_steps": 8450,
|
| 181 |
+
"num_train_epochs": 10,
|
| 182 |
+
"total_flos": 1.2418278660440064e+17,
|
| 183 |
+
"trial_name": null,
|
| 184 |
+
"trial_params": null
|
| 185 |
+
}
|