cxmefzzi / trainer_state.json
tscholak's picture
add T5 model
c9beb6e
raw
history blame
33.1 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 362.58514285714284,
"global_step": 1088,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.29,
"learning_rate": 0.0001,
"loss": 3.7248,
"step": 1
},
{
"epoch": 1.29,
"learning_rate": 0.0001,
"loss": 2.1406,
"step": 4
},
{
"epoch": 2.59,
"learning_rate": 0.0001,
"loss": 1.0705,
"step": 8
},
{
"epoch": 3.88,
"learning_rate": 0.0001,
"loss": 0.6039,
"step": 12
},
{
"epoch": 5.29,
"learning_rate": 0.0001,
"loss": 0.4378,
"step": 16
},
{
"epoch": 6.59,
"learning_rate": 0.0001,
"loss": 0.3218,
"step": 20
},
{
"epoch": 7.88,
"learning_rate": 0.0001,
"loss": 0.2751,
"step": 24
},
{
"epoch": 9.29,
"learning_rate": 0.0001,
"loss": 0.2423,
"step": 28
},
{
"epoch": 10.59,
"learning_rate": 0.0001,
"loss": 0.1763,
"step": 32
},
{
"epoch": 11.88,
"learning_rate": 0.0001,
"loss": 0.1574,
"step": 36
},
{
"epoch": 13.29,
"learning_rate": 0.0001,
"loss": 0.137,
"step": 40
},
{
"epoch": 14.59,
"learning_rate": 0.0001,
"loss": 0.1274,
"step": 44
},
{
"epoch": 15.88,
"learning_rate": 0.0001,
"loss": 0.141,
"step": 48
},
{
"epoch": 17.29,
"learning_rate": 0.0001,
"loss": 0.1345,
"step": 52
},
{
"epoch": 18.59,
"learning_rate": 0.0001,
"loss": 0.0848,
"step": 56
},
{
"epoch": 19.88,
"learning_rate": 0.0001,
"loss": 0.0719,
"step": 60
},
{
"epoch": 21.29,
"learning_rate": 0.0001,
"loss": 0.0814,
"step": 64
},
{
"epoch": 21.29,
"eval_exact_match": 0.5145067698259188,
"eval_loss": 0.15547168254852295,
"eval_runtime": 2276.8146,
"eval_samples_per_second": 0.454,
"step": 64
},
{
"epoch": 22.59,
"learning_rate": 0.0001,
"loss": 0.0665,
"step": 68
},
{
"epoch": 23.88,
"learning_rate": 0.0001,
"loss": 0.0593,
"step": 72
},
{
"epoch": 25.29,
"learning_rate": 0.0001,
"loss": 0.0652,
"step": 76
},
{
"epoch": 26.59,
"learning_rate": 0.0001,
"loss": 0.0896,
"step": 80
},
{
"epoch": 27.88,
"learning_rate": 0.0001,
"loss": 0.0483,
"step": 84
},
{
"epoch": 29.29,
"learning_rate": 0.0001,
"loss": 0.0497,
"step": 88
},
{
"epoch": 30.59,
"learning_rate": 0.0001,
"loss": 0.0479,
"step": 92
},
{
"epoch": 31.88,
"learning_rate": 0.0001,
"loss": 0.0376,
"step": 96
},
{
"epoch": 33.29,
"learning_rate": 0.0001,
"loss": 0.0374,
"step": 100
},
{
"epoch": 34.59,
"learning_rate": 0.0001,
"loss": 0.0342,
"step": 104
},
{
"epoch": 35.88,
"learning_rate": 0.0001,
"loss": 0.0354,
"step": 108
},
{
"epoch": 37.29,
"learning_rate": 0.0001,
"loss": 0.029,
"step": 112
},
{
"epoch": 38.59,
"learning_rate": 0.0001,
"loss": 0.0274,
"step": 116
},
{
"epoch": 39.88,
"learning_rate": 0.0001,
"loss": 0.024,
"step": 120
},
{
"epoch": 41.29,
"learning_rate": 0.0001,
"loss": 0.0331,
"step": 124
},
{
"epoch": 42.59,
"learning_rate": 0.0001,
"loss": 0.0204,
"step": 128
},
{
"epoch": 42.59,
"eval_exact_match": 0.6392649903288201,
"eval_loss": 0.19595815241336823,
"eval_runtime": 1825.3052,
"eval_samples_per_second": 0.566,
"step": 128
},
{
"epoch": 43.88,
"learning_rate": 0.0001,
"loss": 0.0191,
"step": 132
},
{
"epoch": 45.29,
"learning_rate": 0.0001,
"loss": 0.019,
"step": 136
},
{
"epoch": 46.59,
"learning_rate": 0.0001,
"loss": 0.0203,
"step": 140
},
{
"epoch": 47.88,
"learning_rate": 0.0001,
"loss": 0.0182,
"step": 144
},
{
"epoch": 49.29,
"learning_rate": 0.0001,
"loss": 0.0158,
"step": 148
},
{
"epoch": 50.59,
"learning_rate": 0.0001,
"loss": 0.0118,
"step": 152
},
{
"epoch": 51.88,
"learning_rate": 0.0001,
"loss": 0.0139,
"step": 156
},
{
"epoch": 53.29,
"learning_rate": 0.0001,
"loss": 0.016,
"step": 160
},
{
"epoch": 54.59,
"learning_rate": 0.0001,
"loss": 0.022,
"step": 164
},
{
"epoch": 55.88,
"learning_rate": 0.0001,
"loss": 0.0184,
"step": 168
},
{
"epoch": 57.29,
"learning_rate": 0.0001,
"loss": 0.0102,
"step": 172
},
{
"epoch": 58.59,
"learning_rate": 0.0001,
"loss": 0.0096,
"step": 176
},
{
"epoch": 59.88,
"learning_rate": 0.0001,
"loss": 0.0093,
"step": 180
},
{
"epoch": 61.29,
"learning_rate": 0.0001,
"loss": 0.0096,
"step": 184
},
{
"epoch": 62.59,
"learning_rate": 0.0001,
"loss": 0.0075,
"step": 188
},
{
"epoch": 63.88,
"learning_rate": 0.0001,
"loss": 0.0072,
"step": 192
},
{
"epoch": 63.88,
"eval_exact_match": 0.6460348162475822,
"eval_loss": 0.24530905485153198,
"eval_runtime": 1845.9477,
"eval_samples_per_second": 0.56,
"step": 192
},
{
"epoch": 65.29,
"learning_rate": 0.0001,
"loss": 0.007,
"step": 196
},
{
"epoch": 66.59,
"learning_rate": 0.0001,
"loss": 0.0091,
"step": 200
},
{
"epoch": 67.88,
"learning_rate": 0.0001,
"loss": 0.0082,
"step": 204
},
{
"epoch": 69.29,
"learning_rate": 0.0001,
"loss": 0.0063,
"step": 208
},
{
"epoch": 70.59,
"learning_rate": 0.0001,
"loss": 0.0447,
"step": 212
},
{
"epoch": 71.88,
"learning_rate": 0.0001,
"loss": 0.0114,
"step": 216
},
{
"epoch": 73.29,
"learning_rate": 0.0001,
"loss": 0.0065,
"step": 220
},
{
"epoch": 74.59,
"learning_rate": 0.0001,
"loss": 0.0046,
"step": 224
},
{
"epoch": 75.88,
"learning_rate": 0.0001,
"loss": 0.0054,
"step": 228
},
{
"epoch": 77.29,
"learning_rate": 0.0001,
"loss": 0.0048,
"step": 232
},
{
"epoch": 78.59,
"learning_rate": 0.0001,
"loss": 0.0051,
"step": 236
},
{
"epoch": 79.88,
"learning_rate": 0.0001,
"loss": 0.0041,
"step": 240
},
{
"epoch": 81.29,
"learning_rate": 0.0001,
"loss": 0.0044,
"step": 244
},
{
"epoch": 82.59,
"learning_rate": 0.0001,
"loss": 0.0222,
"step": 248
},
{
"epoch": 83.88,
"learning_rate": 0.0001,
"loss": 0.0049,
"step": 252
},
{
"epoch": 85.29,
"learning_rate": 0.0001,
"loss": 0.0632,
"step": 256
},
{
"epoch": 85.29,
"eval_exact_match": 0.648936170212766,
"eval_loss": 0.21414929628372192,
"eval_runtime": 1740.439,
"eval_samples_per_second": 0.594,
"step": 256
},
{
"epoch": 86.59,
"learning_rate": 0.0001,
"loss": 0.0277,
"step": 260
},
{
"epoch": 87.88,
"learning_rate": 0.0001,
"loss": 0.0074,
"step": 264
},
{
"epoch": 89.29,
"learning_rate": 0.0001,
"loss": 0.0041,
"step": 268
},
{
"epoch": 90.59,
"learning_rate": 0.0001,
"loss": 0.0027,
"step": 272
},
{
"epoch": 91.88,
"learning_rate": 0.0001,
"loss": 0.0026,
"step": 276
},
{
"epoch": 93.29,
"learning_rate": 0.0001,
"loss": 0.0024,
"step": 280
},
{
"epoch": 94.59,
"learning_rate": 0.0001,
"loss": 0.0021,
"step": 284
},
{
"epoch": 95.88,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 288
},
{
"epoch": 97.29,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 292
},
{
"epoch": 98.59,
"learning_rate": 0.0001,
"loss": 0.0018,
"step": 296
},
{
"epoch": 99.88,
"learning_rate": 0.0001,
"loss": 0.0017,
"step": 300
},
{
"epoch": 101.29,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 304
},
{
"epoch": 102.59,
"learning_rate": 0.0001,
"loss": 0.0017,
"step": 308
},
{
"epoch": 103.88,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 312
},
{
"epoch": 105.29,
"learning_rate": 0.0001,
"loss": 0.0018,
"step": 316
},
{
"epoch": 106.59,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 320
},
{
"epoch": 106.59,
"eval_exact_match": 0.6808510638297872,
"eval_loss": 0.2944816052913666,
"eval_runtime": 1854.2635,
"eval_samples_per_second": 0.558,
"step": 320
},
{
"epoch": 107.88,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 324
},
{
"epoch": 109.29,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 328
},
{
"epoch": 110.59,
"learning_rate": 0.0001,
"loss": 0.0017,
"step": 332
},
{
"epoch": 111.88,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 336
},
{
"epoch": 113.29,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 340
},
{
"epoch": 114.59,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 344
},
{
"epoch": 115.88,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 348
},
{
"epoch": 117.29,
"learning_rate": 0.0001,
"loss": 0.0021,
"step": 352
},
{
"epoch": 118.59,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 356
},
{
"epoch": 119.88,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 360
},
{
"epoch": 121.29,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 364
},
{
"epoch": 122.59,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 368
},
{
"epoch": 123.88,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 372
},
{
"epoch": 125.29,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 376
},
{
"epoch": 126.59,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 380
},
{
"epoch": 127.88,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 384
},
{
"epoch": 127.88,
"eval_exact_match": 0.6731141199226306,
"eval_loss": 0.31063422560691833,
"eval_runtime": 1905.6919,
"eval_samples_per_second": 0.543,
"step": 384
},
{
"epoch": 129.29,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 388
},
{
"epoch": 130.59,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 392
},
{
"epoch": 131.88,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 396
},
{
"epoch": 133.29,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 400
},
{
"epoch": 134.59,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 404
},
{
"epoch": 135.88,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 408
},
{
"epoch": 137.29,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 412
},
{
"epoch": 138.59,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 416
},
{
"epoch": 139.88,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 420
},
{
"epoch": 141.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 424
},
{
"epoch": 142.59,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 428
},
{
"epoch": 143.88,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 432
},
{
"epoch": 145.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 436
},
{
"epoch": 146.59,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 440
},
{
"epoch": 147.88,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 444
},
{
"epoch": 149.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 448
},
{
"epoch": 149.29,
"eval_exact_match": 0.6847195357833655,
"eval_loss": 0.32885223627090454,
"eval_runtime": 1904.5416,
"eval_samples_per_second": 0.543,
"step": 448
},
{
"epoch": 150.59,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 452
},
{
"epoch": 151.88,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 456
},
{
"epoch": 153.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 460
},
{
"epoch": 154.59,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 464
},
{
"epoch": 155.88,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 468
},
{
"epoch": 157.29,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 472
},
{
"epoch": 158.59,
"learning_rate": 0.0001,
"loss": 0.0096,
"step": 476
},
{
"epoch": 159.88,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 480
},
{
"epoch": 161.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 484
},
{
"epoch": 162.59,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 488
},
{
"epoch": 163.88,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 492
},
{
"epoch": 165.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 496
},
{
"epoch": 166.59,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 500
},
{
"epoch": 167.88,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 504
},
{
"epoch": 169.29,
"learning_rate": 0.0001,
"loss": 0.0148,
"step": 508
},
{
"epoch": 170.59,
"learning_rate": 0.0001,
"loss": 0.0127,
"step": 512
},
{
"epoch": 170.59,
"eval_exact_match": 0.6740812379110251,
"eval_loss": 0.22413159906864166,
"eval_runtime": 1746.478,
"eval_samples_per_second": 0.592,
"step": 512
},
{
"epoch": 171.88,
"learning_rate": 0.0001,
"loss": 0.0023,
"step": 516
},
{
"epoch": 173.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 520
},
{
"epoch": 174.59,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 524
},
{
"epoch": 175.88,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 528
},
{
"epoch": 177.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 532
},
{
"epoch": 178.59,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 536
},
{
"epoch": 179.88,
"learning_rate": 0.0001,
"loss": 0.0044,
"step": 540
},
{
"epoch": 181.29,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 544
},
{
"epoch": 182.59,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 548
},
{
"epoch": 183.88,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 552
},
{
"epoch": 185.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 556
},
{
"epoch": 186.59,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 560
},
{
"epoch": 187.88,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 564
},
{
"epoch": 189.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 568
},
{
"epoch": 190.59,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 572
},
{
"epoch": 191.88,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 576
},
{
"epoch": 191.88,
"eval_exact_match": 0.6876208897485493,
"eval_loss": 0.3054460287094116,
"eval_runtime": 1863.2506,
"eval_samples_per_second": 0.555,
"step": 576
},
{
"epoch": 193.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 580
},
{
"epoch": 194.59,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 584
},
{
"epoch": 195.88,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 588
},
{
"epoch": 197.29,
"learning_rate": 0.0001,
"loss": 0.0069,
"step": 592
},
{
"epoch": 198.59,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 596
},
{
"epoch": 199.88,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 600
},
{
"epoch": 201.29,
"learning_rate": 0.0001,
"loss": 0.0023,
"step": 604
},
{
"epoch": 202.59,
"learning_rate": 0.0001,
"loss": 0.0024,
"step": 608
},
{
"epoch": 203.88,
"learning_rate": 0.0001,
"loss": 0.0253,
"step": 612
},
{
"epoch": 205.29,
"learning_rate": 0.0001,
"loss": 0.0083,
"step": 616
},
{
"epoch": 206.59,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 620
},
{
"epoch": 207.88,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 624
},
{
"epoch": 209.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 628
},
{
"epoch": 210.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 632
},
{
"epoch": 211.88,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 636
},
{
"epoch": 213.29,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 640
},
{
"epoch": 213.29,
"eval_exact_match": 0.6382978723404256,
"eval_loss": 0.3269137442111969,
"eval_runtime": 2353.6231,
"eval_samples_per_second": 0.439,
"step": 640
},
{
"epoch": 214.59,
"learning_rate": 0.0001,
"loss": 0.0095,
"step": 644
},
{
"epoch": 215.88,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 648
},
{
"epoch": 217.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 652
},
{
"epoch": 218.59,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 656
},
{
"epoch": 219.88,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 660
},
{
"epoch": 221.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 664
},
{
"epoch": 222.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 668
},
{
"epoch": 223.88,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 672
},
{
"epoch": 225.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 676
},
{
"epoch": 226.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 680
},
{
"epoch": 227.88,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 684
},
{
"epoch": 229.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 688
},
{
"epoch": 230.59,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 692
},
{
"epoch": 231.88,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 696
},
{
"epoch": 233.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 700
},
{
"epoch": 234.59,
"learning_rate": 0.0001,
"loss": 0.0101,
"step": 704
},
{
"epoch": 234.59,
"eval_exact_match": 0.6769825918762089,
"eval_loss": 0.28256723284721375,
"eval_runtime": 1983.1098,
"eval_samples_per_second": 0.521,
"step": 704
},
{
"epoch": 235.88,
"learning_rate": 0.0001,
"loss": 0.005,
"step": 708
},
{
"epoch": 237.29,
"learning_rate": 0.0001,
"loss": 0.0025,
"step": 712
},
{
"epoch": 238.59,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 716
},
{
"epoch": 239.88,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 720
},
{
"epoch": 241.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 724
},
{
"epoch": 242.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 728
},
{
"epoch": 243.88,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 732
},
{
"epoch": 245.29,
"learning_rate": 0.0001,
"loss": 0.0111,
"step": 736
},
{
"epoch": 246.59,
"learning_rate": 0.0001,
"loss": 0.0112,
"step": 740
},
{
"epoch": 247.88,
"learning_rate": 0.0001,
"loss": 0.0026,
"step": 744
},
{
"epoch": 249.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 748
},
{
"epoch": 250.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 752
},
{
"epoch": 251.88,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 756
},
{
"epoch": 253.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 760
},
{
"epoch": 254.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 764
},
{
"epoch": 255.88,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 768
},
{
"epoch": 255.88,
"eval_exact_match": 0.6982591876208898,
"eval_loss": 0.3340039849281311,
"eval_runtime": 1918.3137,
"eval_samples_per_second": 0.539,
"step": 768
},
{
"epoch": 257.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 772
},
{
"epoch": 258.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 776
},
{
"epoch": 259.88,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 780
},
{
"epoch": 261.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 784
},
{
"epoch": 262.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 788
},
{
"epoch": 263.88,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 792
},
{
"epoch": 265.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 796
},
{
"epoch": 266.59,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 800
},
{
"epoch": 267.88,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 804
},
{
"epoch": 269.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 808
},
{
"epoch": 270.59,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 812
},
{
"epoch": 271.88,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 816
},
{
"epoch": 273.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 820
},
{
"epoch": 274.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 824
},
{
"epoch": 275.88,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 828
},
{
"epoch": 277.29,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 832
},
{
"epoch": 277.29,
"eval_exact_match": 0.6924564796905223,
"eval_loss": 0.3206212818622589,
"eval_runtime": 1930.7912,
"eval_samples_per_second": 0.536,
"step": 832
},
{
"epoch": 278.59,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 836
},
{
"epoch": 279.88,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 840
},
{
"epoch": 281.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 844
},
{
"epoch": 282.59,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 848
},
{
"epoch": 283.88,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 852
},
{
"epoch": 285.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 856
},
{
"epoch": 286.59,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 860
},
{
"epoch": 287.88,
"learning_rate": 0.0001,
"loss": 0.0058,
"step": 864
},
{
"epoch": 289.29,
"learning_rate": 0.0001,
"loss": 0.0153,
"step": 868
},
{
"epoch": 290.59,
"learning_rate": 0.0001,
"loss": 0.004,
"step": 872
},
{
"epoch": 291.88,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 876
},
{
"epoch": 293.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 880
},
{
"epoch": 294.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 884
},
{
"epoch": 295.88,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 888
},
{
"epoch": 297.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 892
},
{
"epoch": 298.59,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 896
},
{
"epoch": 298.59,
"eval_exact_match": 0.7030947775628626,
"eval_loss": 0.3318493366241455,
"eval_runtime": 1959.3529,
"eval_samples_per_second": 0.528,
"step": 896
},
{
"epoch": 299.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 900
},
{
"epoch": 301.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 904
},
{
"epoch": 302.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 908
},
{
"epoch": 303.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 912
},
{
"epoch": 305.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 916
},
{
"epoch": 306.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 920
},
{
"epoch": 307.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 924
},
{
"epoch": 309.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 928
},
{
"epoch": 310.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 932
},
{
"epoch": 311.88,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 936
},
{
"epoch": 313.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 940
},
{
"epoch": 314.59,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 944
},
{
"epoch": 315.88,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 948
},
{
"epoch": 317.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 952
},
{
"epoch": 318.59,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 956
},
{
"epoch": 319.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 960
},
{
"epoch": 319.88,
"eval_exact_match": 0.6972920696324951,
"eval_loss": 0.34125566482543945,
"eval_runtime": 1929.7453,
"eval_samples_per_second": 0.536,
"step": 960
},
{
"epoch": 321.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 964
},
{
"epoch": 322.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 968
},
{
"epoch": 323.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 972
},
{
"epoch": 325.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 976
},
{
"epoch": 326.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 980
},
{
"epoch": 327.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 984
},
{
"epoch": 329.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 988
},
{
"epoch": 330.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 992
},
{
"epoch": 331.88,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 996
},
{
"epoch": 333.29,
"learning_rate": 0.0001,
"loss": 0.015,
"step": 1000
},
{
"epoch": 334.59,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1004
},
{
"epoch": 335.88,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1008
},
{
"epoch": 337.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1012
},
{
"epoch": 338.59,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1016
},
{
"epoch": 339.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1020
},
{
"epoch": 341.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1024
},
{
"epoch": 341.29,
"eval_exact_match": 0.6963249516441006,
"eval_loss": 0.3414294123649597,
"eval_runtime": 1925.945,
"eval_samples_per_second": 0.537,
"step": 1024
},
{
"epoch": 342.59,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1028
},
{
"epoch": 343.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1032
},
{
"epoch": 345.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1036
},
{
"epoch": 346.59,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1040
},
{
"epoch": 347.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1044
},
{
"epoch": 349.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1048
},
{
"epoch": 350.59,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1052
},
{
"epoch": 351.88,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1056
},
{
"epoch": 353.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1060
},
{
"epoch": 354.59,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1064
},
{
"epoch": 355.88,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1068
},
{
"epoch": 357.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1072
},
{
"epoch": 358.59,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1076
},
{
"epoch": 359.88,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1080
},
{
"epoch": 361.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1084
},
{
"epoch": 362.59,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1088
},
{
"epoch": 362.59,
"eval_exact_match": 0.7127659574468085,
"eval_loss": 0.3638148605823517,
"eval_runtime": 1954.8096,
"eval_samples_per_second": 0.529,
"step": 1088
}
],
"max_steps": 9216,
"num_train_epochs": 3072,
"total_flos": 6.244604651315036e+17,
"trial_name": null,
"trial_params": null
}