Upload trainer_state.json
Browse files- trainer_state.json +186 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,186 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.907652942544778,
|
| 5 |
+
"global_step": 12500,
|
| 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,
|
| 12 |
+
"learning_rate": 1.9224625881988063e-05,
|
| 13 |
+
"loss": 0.9542,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.23,
|
| 18 |
+
"learning_rate": 1.844925176397612e-05,
|
| 19 |
+
"loss": 0.9606,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.35,
|
| 24 |
+
"learning_rate": 1.767387764596418e-05,
|
| 25 |
+
"loss": 0.9752,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.47,
|
| 30 |
+
"learning_rate": 1.689850352795224e-05,
|
| 31 |
+
"loss": 0.9758,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.58,
|
| 36 |
+
"learning_rate": 1.6123129409940298e-05,
|
| 37 |
+
"loss": 0.9911,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.7,
|
| 42 |
+
"learning_rate": 1.5347755291928356e-05,
|
| 43 |
+
"loss": 1.0023,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.81,
|
| 48 |
+
"learning_rate": 1.4572381173916417e-05,
|
| 49 |
+
"loss": 1.0072,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.93,
|
| 54 |
+
"learning_rate": 1.3797007055904474e-05,
|
| 55 |
+
"loss": 1.0212,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.0,
|
| 60 |
+
"eval_bleu": 38.6916,
|
| 61 |
+
"eval_gen_len": 28.7748,
|
| 62 |
+
"eval_loss": 1.050680160522461,
|
| 63 |
+
"eval_runtime": 181.503,
|
| 64 |
+
"eval_samples_per_second": 7.779,
|
| 65 |
+
"eval_steps_per_second": 0.127,
|
| 66 |
+
"step": 4299
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.05,
|
| 70 |
+
"learning_rate": 1.3021632937892535e-05,
|
| 71 |
+
"loss": 1.0013,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.16,
|
| 76 |
+
"learning_rate": 1.2246258819880595e-05,
|
| 77 |
+
"loss": 0.9579,
|
| 78 |
+
"step": 5000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.28,
|
| 82 |
+
"learning_rate": 1.1470884701868653e-05,
|
| 83 |
+
"loss": 0.9527,
|
| 84 |
+
"step": 5500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.4,
|
| 88 |
+
"learning_rate": 1.0695510583856712e-05,
|
| 89 |
+
"loss": 0.959,
|
| 90 |
+
"step": 6000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.51,
|
| 94 |
+
"learning_rate": 9.92013646584477e-06,
|
| 95 |
+
"loss": 0.9606,
|
| 96 |
+
"step": 6500
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.63,
|
| 100 |
+
"learning_rate": 9.14476234783283e-06,
|
| 101 |
+
"loss": 0.9579,
|
| 102 |
+
"step": 7000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.74,
|
| 106 |
+
"learning_rate": 8.36938822982089e-06,
|
| 107 |
+
"loss": 0.9563,
|
| 108 |
+
"step": 7500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.86,
|
| 112 |
+
"learning_rate": 7.594014111808948e-06,
|
| 113 |
+
"loss": 0.957,
|
| 114 |
+
"step": 8000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.98,
|
| 118 |
+
"learning_rate": 6.818639993797007e-06,
|
| 119 |
+
"loss": 0.9558,
|
| 120 |
+
"step": 8500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 2.0,
|
| 124 |
+
"eval_bleu": 38.8229,
|
| 125 |
+
"eval_gen_len": 28.7649,
|
| 126 |
+
"eval_loss": 1.0432215929031372,
|
| 127 |
+
"eval_runtime": 185.3749,
|
| 128 |
+
"eval_samples_per_second": 7.617,
|
| 129 |
+
"eval_steps_per_second": 0.124,
|
| 130 |
+
"step": 8598
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.09,
|
| 134 |
+
"learning_rate": 6.043265875785067e-06,
|
| 135 |
+
"loss": 0.9358,
|
| 136 |
+
"step": 9000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 2.21,
|
| 140 |
+
"learning_rate": 5.2678917577731256e-06,
|
| 141 |
+
"loss": 0.9268,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.33,
|
| 146 |
+
"learning_rate": 4.492517639761185e-06,
|
| 147 |
+
"loss": 0.9238,
|
| 148 |
+
"step": 10000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 2.44,
|
| 152 |
+
"learning_rate": 3.717143521749244e-06,
|
| 153 |
+
"loss": 0.9267,
|
| 154 |
+
"step": 10500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 2.56,
|
| 158 |
+
"learning_rate": 2.9417694037373037e-06,
|
| 159 |
+
"loss": 0.9219,
|
| 160 |
+
"step": 11000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.68,
|
| 164 |
+
"learning_rate": 2.1663952857253625e-06,
|
| 165 |
+
"loss": 0.9285,
|
| 166 |
+
"step": 11500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 2.79,
|
| 170 |
+
"learning_rate": 1.3910211677134219e-06,
|
| 171 |
+
"loss": 0.925,
|
| 172 |
+
"step": 12000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.91,
|
| 176 |
+
"learning_rate": 6.15647049701481e-07,
|
| 177 |
+
"loss": 0.9215,
|
| 178 |
+
"step": 12500
|
| 179 |
+
}
|
| 180 |
+
],
|
| 181 |
+
"max_steps": 12897,
|
| 182 |
+
"num_train_epochs": 3,
|
| 183 |
+
"total_flos": 1.868489262563328e+16,
|
| 184 |
+
"trial_name": null,
|
| 185 |
+
"trial_params": null
|
| 186 |
+
}
|