File size: 5,521 Bytes
70d36c0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 |
{
"best_metric": 0.6879923855830649,
"best_model_checkpoint": "output/pretraining/vihealthbert-w_unsup-SynPD/lr3e-5_wr0.1_wd0.0/checkpoint-55000",
"epoch": 9.477856281233844,
"eval_steps": 5000,
"global_step": 55000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00017232465965879716,
"grad_norm": 41.827266693115234,
"learning_rate": 5.169739789763915e-09,
"loss": 13.7242,
"step": 1
},
{
"epoch": 0.8616232982939859,
"grad_norm": 5.6151957511901855,
"learning_rate": 2.584869894881958e-05,
"loss": 7.0234,
"step": 5000
},
{
"epoch": 0.8616232982939859,
"eval_accuracy": 0.5576033171446898,
"eval_loss": 2.590928077697754,
"eval_runtime": 37.589,
"eval_samples_per_second": 260.023,
"eval_steps_per_second": 16.255,
"step": 5000
},
{
"epoch": 1.7232465965879717,
"grad_norm": 4.410743713378906,
"learning_rate": 2.758917801137343e-05,
"loss": 5.2736,
"step": 10000
},
{
"epoch": 1.7232465965879717,
"eval_accuracy": 0.596153624823046,
"eval_loss": 2.188957452774048,
"eval_runtime": 37.6297,
"eval_samples_per_second": 259.741,
"eval_steps_per_second": 16.237,
"step": 10000
},
{
"epoch": 2.5848698948819577,
"grad_norm": 4.754213809967041,
"learning_rate": 2.4717100350393475e-05,
"loss": 4.9126,
"step": 15000
},
{
"epoch": 2.5848698948819577,
"eval_accuracy": 0.6381037526075694,
"eval_loss": 1.909491777420044,
"eval_runtime": 36.8556,
"eval_samples_per_second": 265.197,
"eval_steps_per_second": 16.578,
"step": 15000
},
{
"epoch": 3.4464931931759435,
"grad_norm": 4.2047200202941895,
"learning_rate": 2.184502268941352e-05,
"loss": 4.791,
"step": 20000
},
{
"epoch": 3.4464931931759435,
"eval_accuracy": 0.6468589272593681,
"eval_loss": 1.8286069631576538,
"eval_runtime": 35.9342,
"eval_samples_per_second": 271.997,
"eval_steps_per_second": 17.003,
"step": 20000
},
{
"epoch": 4.30811649146993,
"grad_norm": 4.629519462585449,
"learning_rate": 1.897294502843357e-05,
"loss": 4.6538,
"step": 25000
},
{
"epoch": 4.30811649146993,
"eval_accuracy": 0.6644416567883901,
"eval_loss": 1.714424967765808,
"eval_runtime": 35.6122,
"eval_samples_per_second": 274.456,
"eval_steps_per_second": 17.157,
"step": 25000
},
{
"epoch": 5.1697397897639155,
"grad_norm": 3.993448257446289,
"learning_rate": 1.6100867367453616e-05,
"loss": 4.5846,
"step": 30000
},
{
"epoch": 5.1697397897639155,
"eval_accuracy": 0.6703549228453575,
"eval_loss": 1.6779303550720215,
"eval_runtime": 36.1572,
"eval_samples_per_second": 270.32,
"eval_steps_per_second": 16.898,
"step": 30000
},
{
"epoch": 6.031363088057901,
"grad_norm": 4.6959357261657715,
"learning_rate": 1.3228789706473663e-05,
"loss": 4.5568,
"step": 35000
},
{
"epoch": 6.031363088057901,
"eval_accuracy": 0.6765665494261385,
"eval_loss": 1.6362268924713135,
"eval_runtime": 36.2906,
"eval_samples_per_second": 269.326,
"eval_steps_per_second": 16.836,
"step": 35000
},
{
"epoch": 6.892986386351887,
"grad_norm": 4.4082441329956055,
"learning_rate": 1.035671204549371e-05,
"loss": 4.5079,
"step": 40000
},
{
"epoch": 6.892986386351887,
"eval_accuracy": 0.681425908452811,
"eval_loss": 1.6008453369140625,
"eval_runtime": 36.1284,
"eval_samples_per_second": 270.535,
"eval_steps_per_second": 16.912,
"step": 40000
},
{
"epoch": 7.754609684645873,
"grad_norm": 5.832913398742676,
"learning_rate": 7.484634384513757e-06,
"loss": 4.469,
"step": 45000
},
{
"epoch": 7.754609684645873,
"eval_accuracy": 0.6805218690729352,
"eval_loss": 1.6063588857650757,
"eval_runtime": 36.0752,
"eval_samples_per_second": 270.934,
"eval_steps_per_second": 16.937,
"step": 45000
},
{
"epoch": 8.61623298293986,
"grad_norm": 3.991995096206665,
"learning_rate": 4.612556723533804e-06,
"loss": 4.4514,
"step": 50000
},
{
"epoch": 8.61623298293986,
"eval_accuracy": 0.6852411542306852,
"eval_loss": 1.5799689292907715,
"eval_runtime": 37.5803,
"eval_samples_per_second": 260.083,
"eval_steps_per_second": 16.259,
"step": 50000
},
{
"epoch": 9.477856281233844,
"grad_norm": 4.424502849578857,
"learning_rate": 1.7404790625538515e-06,
"loss": 4.4317,
"step": 55000
},
{
"epoch": 9.477856281233844,
"eval_accuracy": 0.6879923855830649,
"eval_loss": 1.5539859533309937,
"eval_runtime": 35.7439,
"eval_samples_per_second": 273.445,
"eval_steps_per_second": 17.094,
"step": 55000
}
],
"logging_steps": 5000,
"max_steps": 58030,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 5000,
"total_flos": 0.0,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}
|