1wnr382e / trainer_state.json
tscholak's picture
add T5 model
3325e82 unverified
raw
history blame
188 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2069.2914285714287,
"global_step": 6208,
"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": 4.5844,
"step": 1
},
{
"epoch": 1.29,
"learning_rate": 0.0001,
"loss": 3.1856,
"step": 4
},
{
"epoch": 2.58,
"learning_rate": 0.0001,
"loss": 1.5583,
"step": 8
},
{
"epoch": 3.87,
"learning_rate": 0.0001,
"loss": 0.9704,
"step": 12
},
{
"epoch": 5.29,
"learning_rate": 0.0001,
"loss": 0.8027,
"step": 16
},
{
"epoch": 6.58,
"learning_rate": 0.0001,
"loss": 0.5799,
"step": 20
},
{
"epoch": 7.87,
"learning_rate": 0.0001,
"loss": 0.4733,
"step": 24
},
{
"epoch": 9.29,
"learning_rate": 0.0001,
"loss": 0.4734,
"step": 28
},
{
"epoch": 10.58,
"learning_rate": 0.0001,
"loss": 0.3928,
"step": 32
},
{
"epoch": 11.87,
"learning_rate": 0.0001,
"loss": 0.33,
"step": 36
},
{
"epoch": 13.29,
"learning_rate": 0.0001,
"loss": 0.3141,
"step": 40
},
{
"epoch": 14.58,
"learning_rate": 0.0001,
"loss": 0.3032,
"step": 44
},
{
"epoch": 15.87,
"learning_rate": 0.0001,
"loss": 0.2666,
"step": 48
},
{
"epoch": 17.29,
"learning_rate": 0.0001,
"loss": 0.2819,
"step": 52
},
{
"epoch": 18.58,
"learning_rate": 0.0001,
"loss": 0.2308,
"step": 56
},
{
"epoch": 19.87,
"learning_rate": 0.0001,
"loss": 0.2034,
"step": 60
},
{
"epoch": 21.29,
"learning_rate": 0.0001,
"loss": 0.2175,
"step": 64
},
{
"epoch": 21.29,
"eval_exact_match": 0.3742746615087041,
"eval_loss": 0.2133236974477768,
"eval_runtime": 392.3685,
"eval_samples_per_second": 2.635,
"step": 64
},
{
"epoch": 22.58,
"learning_rate": 0.0001,
"loss": 0.2026,
"step": 68
},
{
"epoch": 23.87,
"learning_rate": 0.0001,
"loss": 0.1963,
"step": 72
},
{
"epoch": 25.29,
"learning_rate": 0.0001,
"loss": 0.187,
"step": 76
},
{
"epoch": 26.58,
"learning_rate": 0.0001,
"loss": 0.1723,
"step": 80
},
{
"epoch": 27.87,
"learning_rate": 0.0001,
"loss": 0.1549,
"step": 84
},
{
"epoch": 29.29,
"learning_rate": 0.0001,
"loss": 0.1544,
"step": 88
},
{
"epoch": 30.58,
"learning_rate": 0.0001,
"loss": 0.1283,
"step": 92
},
{
"epoch": 31.87,
"learning_rate": 0.0001,
"loss": 0.1298,
"step": 96
},
{
"epoch": 33.29,
"learning_rate": 0.0001,
"loss": 0.1174,
"step": 100
},
{
"epoch": 34.58,
"learning_rate": 0.0001,
"loss": 0.1021,
"step": 104
},
{
"epoch": 35.87,
"learning_rate": 0.0001,
"loss": 0.105,
"step": 108
},
{
"epoch": 37.29,
"learning_rate": 0.0001,
"loss": 0.1146,
"step": 112
},
{
"epoch": 38.58,
"learning_rate": 0.0001,
"loss": 0.0868,
"step": 116
},
{
"epoch": 39.87,
"learning_rate": 0.0001,
"loss": 0.0851,
"step": 120
},
{
"epoch": 41.29,
"learning_rate": 0.0001,
"loss": 0.0968,
"step": 124
},
{
"epoch": 42.58,
"learning_rate": 0.0001,
"loss": 0.0861,
"step": 128
},
{
"epoch": 42.58,
"eval_exact_match": 0.5193423597678917,
"eval_loss": 0.18803898990154266,
"eval_runtime": 182.6967,
"eval_samples_per_second": 5.66,
"step": 128
},
{
"epoch": 43.87,
"learning_rate": 0.0001,
"loss": 0.0817,
"step": 132
},
{
"epoch": 45.29,
"learning_rate": 0.0001,
"loss": 0.0813,
"step": 136
},
{
"epoch": 46.58,
"learning_rate": 0.0001,
"loss": 0.0753,
"step": 140
},
{
"epoch": 47.87,
"learning_rate": 0.0001,
"loss": 0.0706,
"step": 144
},
{
"epoch": 49.29,
"learning_rate": 0.0001,
"loss": 0.069,
"step": 148
},
{
"epoch": 50.58,
"learning_rate": 0.0001,
"loss": 0.0621,
"step": 152
},
{
"epoch": 51.87,
"learning_rate": 0.0001,
"loss": 0.0612,
"step": 156
},
{
"epoch": 53.29,
"learning_rate": 0.0001,
"loss": 0.0642,
"step": 160
},
{
"epoch": 54.58,
"learning_rate": 0.0001,
"loss": 0.0571,
"step": 164
},
{
"epoch": 55.87,
"learning_rate": 0.0001,
"loss": 0.0578,
"step": 168
},
{
"epoch": 57.29,
"learning_rate": 0.0001,
"loss": 0.0594,
"step": 172
},
{
"epoch": 58.58,
"learning_rate": 0.0001,
"loss": 0.0492,
"step": 176
},
{
"epoch": 59.87,
"learning_rate": 0.0001,
"loss": 0.0483,
"step": 180
},
{
"epoch": 61.29,
"learning_rate": 0.0001,
"loss": 0.054,
"step": 184
},
{
"epoch": 62.58,
"learning_rate": 0.0001,
"loss": 0.0417,
"step": 188
},
{
"epoch": 63.87,
"learning_rate": 0.0001,
"loss": 0.0393,
"step": 192
},
{
"epoch": 63.87,
"eval_exact_match": 0.5735009671179884,
"eval_loss": 0.20905810594558716,
"eval_runtime": 185.8951,
"eval_samples_per_second": 5.562,
"step": 192
},
{
"epoch": 65.29,
"learning_rate": 0.0001,
"loss": 0.0385,
"step": 196
},
{
"epoch": 66.58,
"learning_rate": 0.0001,
"loss": 0.0354,
"step": 200
},
{
"epoch": 67.87,
"learning_rate": 0.0001,
"loss": 0.0354,
"step": 204
},
{
"epoch": 69.29,
"learning_rate": 0.0001,
"loss": 0.0439,
"step": 208
},
{
"epoch": 70.58,
"learning_rate": 0.0001,
"loss": 0.0322,
"step": 212
},
{
"epoch": 71.87,
"learning_rate": 0.0001,
"loss": 0.0394,
"step": 216
},
{
"epoch": 73.29,
"learning_rate": 0.0001,
"loss": 0.035,
"step": 220
},
{
"epoch": 74.58,
"learning_rate": 0.0001,
"loss": 0.0319,
"step": 224
},
{
"epoch": 75.87,
"learning_rate": 0.0001,
"loss": 0.032,
"step": 228
},
{
"epoch": 77.29,
"learning_rate": 0.0001,
"loss": 0.0322,
"step": 232
},
{
"epoch": 78.58,
"learning_rate": 0.0001,
"loss": 0.0301,
"step": 236
},
{
"epoch": 79.87,
"learning_rate": 0.0001,
"loss": 0.0232,
"step": 240
},
{
"epoch": 81.29,
"learning_rate": 0.0001,
"loss": 0.0263,
"step": 244
},
{
"epoch": 82.58,
"learning_rate": 0.0001,
"loss": 0.0255,
"step": 248
},
{
"epoch": 83.87,
"learning_rate": 0.0001,
"loss": 0.0235,
"step": 252
},
{
"epoch": 85.29,
"learning_rate": 0.0001,
"loss": 0.0295,
"step": 256
},
{
"epoch": 85.29,
"eval_exact_match": 0.43617021276595747,
"eval_loss": 0.24603009223937988,
"eval_runtime": 212.814,
"eval_samples_per_second": 4.859,
"step": 256
},
{
"epoch": 86.58,
"learning_rate": 0.0001,
"loss": 0.0312,
"step": 260
},
{
"epoch": 87.87,
"learning_rate": 0.0001,
"loss": 0.0204,
"step": 264
},
{
"epoch": 89.29,
"learning_rate": 0.0001,
"loss": 0.0214,
"step": 268
},
{
"epoch": 90.58,
"learning_rate": 0.0001,
"loss": 0.0196,
"step": 272
},
{
"epoch": 91.87,
"learning_rate": 0.0001,
"loss": 0.0177,
"step": 276
},
{
"epoch": 93.29,
"learning_rate": 0.0001,
"loss": 0.0176,
"step": 280
},
{
"epoch": 94.58,
"learning_rate": 0.0001,
"loss": 0.0217,
"step": 284
},
{
"epoch": 95.87,
"learning_rate": 0.0001,
"loss": 0.0163,
"step": 288
},
{
"epoch": 97.29,
"learning_rate": 0.0001,
"loss": 0.0227,
"step": 292
},
{
"epoch": 98.58,
"learning_rate": 0.0001,
"loss": 0.014,
"step": 296
},
{
"epoch": 99.87,
"learning_rate": 0.0001,
"loss": 0.0136,
"step": 300
},
{
"epoch": 101.29,
"learning_rate": 0.0001,
"loss": 0.0149,
"step": 304
},
{
"epoch": 102.58,
"learning_rate": 0.0001,
"loss": 0.0141,
"step": 308
},
{
"epoch": 103.87,
"learning_rate": 0.0001,
"loss": 0.0186,
"step": 312
},
{
"epoch": 105.29,
"learning_rate": 0.0001,
"loss": 0.02,
"step": 316
},
{
"epoch": 106.58,
"learning_rate": 0.0001,
"loss": 0.0157,
"step": 320
},
{
"epoch": 106.58,
"eval_exact_match": 0.6015473887814313,
"eval_loss": 0.2599264681339264,
"eval_runtime": 164.4683,
"eval_samples_per_second": 6.287,
"step": 320
},
{
"epoch": 107.87,
"learning_rate": 0.0001,
"loss": 0.0129,
"step": 324
},
{
"epoch": 109.29,
"learning_rate": 0.0001,
"loss": 0.0131,
"step": 328
},
{
"epoch": 110.58,
"learning_rate": 0.0001,
"loss": 0.0114,
"step": 332
},
{
"epoch": 111.87,
"learning_rate": 0.0001,
"loss": 0.0103,
"step": 336
},
{
"epoch": 113.29,
"learning_rate": 0.0001,
"loss": 0.01,
"step": 340
},
{
"epoch": 114.58,
"learning_rate": 0.0001,
"loss": 0.0098,
"step": 344
},
{
"epoch": 115.87,
"learning_rate": 0.0001,
"loss": 0.0093,
"step": 348
},
{
"epoch": 117.29,
"learning_rate": 0.0001,
"loss": 0.0113,
"step": 352
},
{
"epoch": 118.58,
"learning_rate": 0.0001,
"loss": 0.0094,
"step": 356
},
{
"epoch": 119.87,
"learning_rate": 0.0001,
"loss": 0.0091,
"step": 360
},
{
"epoch": 121.29,
"learning_rate": 0.0001,
"loss": 0.0086,
"step": 364
},
{
"epoch": 122.58,
"learning_rate": 0.0001,
"loss": 0.0087,
"step": 368
},
{
"epoch": 123.87,
"learning_rate": 0.0001,
"loss": 0.0081,
"step": 372
},
{
"epoch": 125.29,
"learning_rate": 0.0001,
"loss": 0.013,
"step": 376
},
{
"epoch": 126.58,
"learning_rate": 0.0001,
"loss": 0.008,
"step": 380
},
{
"epoch": 127.87,
"learning_rate": 0.0001,
"loss": 0.0072,
"step": 384
},
{
"epoch": 127.87,
"eval_exact_match": 0.6112185686653772,
"eval_loss": 0.28473153710365295,
"eval_runtime": 194.1529,
"eval_samples_per_second": 5.326,
"step": 384
},
{
"epoch": 129.29,
"learning_rate": 0.0001,
"loss": 0.007,
"step": 388
},
{
"epoch": 130.58,
"learning_rate": 0.0001,
"loss": 0.0074,
"step": 392
},
{
"epoch": 131.87,
"learning_rate": 0.0001,
"loss": 0.006,
"step": 396
},
{
"epoch": 133.29,
"learning_rate": 0.0001,
"loss": 0.01,
"step": 400
},
{
"epoch": 134.58,
"learning_rate": 0.0001,
"loss": 0.009,
"step": 404
},
{
"epoch": 135.87,
"learning_rate": 0.0001,
"loss": 0.01,
"step": 408
},
{
"epoch": 137.29,
"learning_rate": 0.0001,
"loss": 0.0127,
"step": 412
},
{
"epoch": 138.58,
"learning_rate": 0.0001,
"loss": 0.0066,
"step": 416
},
{
"epoch": 139.87,
"learning_rate": 0.0001,
"loss": 0.0074,
"step": 420
},
{
"epoch": 141.29,
"learning_rate": 0.0001,
"loss": 0.0058,
"step": 424
},
{
"epoch": 142.58,
"learning_rate": 0.0001,
"loss": 0.0047,
"step": 428
},
{
"epoch": 143.87,
"learning_rate": 0.0001,
"loss": 0.0047,
"step": 432
},
{
"epoch": 145.29,
"learning_rate": 0.0001,
"loss": 0.0054,
"step": 436
},
{
"epoch": 146.58,
"learning_rate": 0.0001,
"loss": 0.005,
"step": 440
},
{
"epoch": 147.87,
"learning_rate": 0.0001,
"loss": 0.0046,
"step": 444
},
{
"epoch": 149.29,
"learning_rate": 0.0001,
"loss": 0.0047,
"step": 448
},
{
"epoch": 149.29,
"eval_exact_match": 0.6141199226305609,
"eval_loss": 0.323922336101532,
"eval_runtime": 195.6739,
"eval_samples_per_second": 5.284,
"step": 448
},
{
"epoch": 150.58,
"learning_rate": 0.0001,
"loss": 0.0043,
"step": 452
},
{
"epoch": 151.87,
"learning_rate": 0.0001,
"loss": 0.0041,
"step": 456
},
{
"epoch": 153.29,
"learning_rate": 0.0001,
"loss": 0.005,
"step": 460
},
{
"epoch": 154.58,
"learning_rate": 0.0001,
"loss": 0.0053,
"step": 464
},
{
"epoch": 155.87,
"learning_rate": 0.0001,
"loss": 0.0036,
"step": 468
},
{
"epoch": 157.29,
"learning_rate": 0.0001,
"loss": 0.0044,
"step": 472
},
{
"epoch": 158.58,
"learning_rate": 0.0001,
"loss": 0.0036,
"step": 476
},
{
"epoch": 159.87,
"learning_rate": 0.0001,
"loss": 0.0041,
"step": 480
},
{
"epoch": 161.29,
"learning_rate": 0.0001,
"loss": 0.004,
"step": 484
},
{
"epoch": 162.58,
"learning_rate": 0.0001,
"loss": 0.0037,
"step": 488
},
{
"epoch": 163.87,
"learning_rate": 0.0001,
"loss": 0.0031,
"step": 492
},
{
"epoch": 165.29,
"learning_rate": 0.0001,
"loss": 0.004,
"step": 496
},
{
"epoch": 166.58,
"learning_rate": 0.0001,
"loss": 0.0031,
"step": 500
},
{
"epoch": 167.87,
"learning_rate": 0.0001,
"loss": 0.0031,
"step": 504
},
{
"epoch": 169.29,
"learning_rate": 0.0001,
"loss": 0.003,
"step": 508
},
{
"epoch": 170.58,
"learning_rate": 0.0001,
"loss": 0.0036,
"step": 512
},
{
"epoch": 170.58,
"eval_exact_match": 0.6305609284332688,
"eval_loss": 0.33440926671028137,
"eval_runtime": 186.0075,
"eval_samples_per_second": 5.559,
"step": 512
},
{
"epoch": 171.87,
"learning_rate": 0.0001,
"loss": 0.0031,
"step": 516
},
{
"epoch": 173.29,
"learning_rate": 0.0001,
"loss": 0.0037,
"step": 520
},
{
"epoch": 174.58,
"learning_rate": 0.0001,
"loss": 0.0032,
"step": 524
},
{
"epoch": 175.87,
"learning_rate": 0.0001,
"loss": 0.0107,
"step": 528
},
{
"epoch": 177.29,
"learning_rate": 0.0001,
"loss": 0.0075,
"step": 532
},
{
"epoch": 178.58,
"learning_rate": 0.0001,
"loss": 0.0027,
"step": 536
},
{
"epoch": 179.87,
"learning_rate": 0.0001,
"loss": 0.0027,
"step": 540
},
{
"epoch": 181.29,
"learning_rate": 0.0001,
"loss": 0.0028,
"step": 544
},
{
"epoch": 182.58,
"learning_rate": 0.0001,
"loss": 0.0024,
"step": 548
},
{
"epoch": 183.87,
"learning_rate": 0.0001,
"loss": 0.0022,
"step": 552
},
{
"epoch": 185.29,
"learning_rate": 0.0001,
"loss": 0.0022,
"step": 556
},
{
"epoch": 186.58,
"learning_rate": 0.0001,
"loss": 0.0021,
"step": 560
},
{
"epoch": 187.87,
"learning_rate": 0.0001,
"loss": 0.0029,
"step": 564
},
{
"epoch": 189.29,
"learning_rate": 0.0001,
"loss": 0.0028,
"step": 568
},
{
"epoch": 190.58,
"learning_rate": 0.0001,
"loss": 0.0025,
"step": 572
},
{
"epoch": 191.87,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 576
},
{
"epoch": 191.87,
"eval_exact_match": 0.6150870406189555,
"eval_loss": 0.3535918891429901,
"eval_runtime": 185.3001,
"eval_samples_per_second": 5.58,
"step": 576
},
{
"epoch": 193.29,
"learning_rate": 0.0001,
"loss": 0.0028,
"step": 580
},
{
"epoch": 194.58,
"learning_rate": 0.0001,
"loss": 0.0024,
"step": 584
},
{
"epoch": 195.87,
"learning_rate": 0.0001,
"loss": 0.0021,
"step": 588
},
{
"epoch": 197.29,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 592
},
{
"epoch": 198.58,
"learning_rate": 0.0001,
"loss": 0.0023,
"step": 596
},
{
"epoch": 199.87,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 600
},
{
"epoch": 201.29,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 604
},
{
"epoch": 202.58,
"learning_rate": 0.0001,
"loss": 0.0098,
"step": 608
},
{
"epoch": 203.87,
"learning_rate": 0.0001,
"loss": 0.0074,
"step": 612
},
{
"epoch": 205.29,
"learning_rate": 0.0001,
"loss": 0.0024,
"step": 616
},
{
"epoch": 206.58,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 620
},
{
"epoch": 207.87,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 624
},
{
"epoch": 209.29,
"learning_rate": 0.0001,
"loss": 0.0024,
"step": 628
},
{
"epoch": 210.58,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 632
},
{
"epoch": 211.87,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 636
},
{
"epoch": 213.29,
"learning_rate": 0.0001,
"loss": 0.0017,
"step": 640
},
{
"epoch": 213.29,
"eval_exact_match": 0.6131528046421664,
"eval_loss": 0.35180971026420593,
"eval_runtime": 192.056,
"eval_samples_per_second": 5.384,
"step": 640
},
{
"epoch": 214.58,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 644
},
{
"epoch": 215.87,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 648
},
{
"epoch": 217.29,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 652
},
{
"epoch": 218.58,
"learning_rate": 0.0001,
"loss": 0.0018,
"step": 656
},
{
"epoch": 219.87,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 660
},
{
"epoch": 221.29,
"learning_rate": 0.0001,
"loss": 0.0018,
"step": 664
},
{
"epoch": 222.58,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 668
},
{
"epoch": 223.87,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 672
},
{
"epoch": 225.29,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 676
},
{
"epoch": 226.58,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 680
},
{
"epoch": 227.87,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 684
},
{
"epoch": 229.29,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 688
},
{
"epoch": 230.58,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 692
},
{
"epoch": 231.87,
"learning_rate": 0.0001,
"loss": 0.0017,
"step": 696
},
{
"epoch": 233.29,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 700
},
{
"epoch": 234.58,
"learning_rate": 0.0001,
"loss": 0.0017,
"step": 704
},
{
"epoch": 234.58,
"eval_exact_match": 0.620889748549323,
"eval_loss": 0.35935279726982117,
"eval_runtime": 197.9548,
"eval_samples_per_second": 5.223,
"step": 704
},
{
"epoch": 235.87,
"learning_rate": 0.0001,
"loss": 0.0053,
"step": 708
},
{
"epoch": 237.29,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 712
},
{
"epoch": 238.58,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 716
},
{
"epoch": 239.87,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 720
},
{
"epoch": 241.29,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 724
},
{
"epoch": 242.58,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 728
},
{
"epoch": 243.87,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 732
},
{
"epoch": 245.29,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 736
},
{
"epoch": 246.58,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 740
},
{
"epoch": 247.87,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 744
},
{
"epoch": 249.29,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 748
},
{
"epoch": 250.58,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 752
},
{
"epoch": 251.87,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 756
},
{
"epoch": 253.29,
"learning_rate": 0.0001,
"loss": 0.0025,
"step": 760
},
{
"epoch": 254.58,
"learning_rate": 0.0001,
"loss": 0.0024,
"step": 764
},
{
"epoch": 255.87,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 768
},
{
"epoch": 255.87,
"eval_exact_match": 0.6141199226305609,
"eval_loss": 0.35694804787635803,
"eval_runtime": 194.1968,
"eval_samples_per_second": 5.324,
"step": 768
},
{
"epoch": 257.29,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 772
},
{
"epoch": 258.58,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 776
},
{
"epoch": 259.87,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 780
},
{
"epoch": 261.29,
"learning_rate": 0.0001,
"loss": 0.0017,
"step": 784
},
{
"epoch": 262.58,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 788
},
{
"epoch": 263.87,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 792
},
{
"epoch": 265.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 796
},
{
"epoch": 266.58,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 800
},
{
"epoch": 267.87,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 804
},
{
"epoch": 269.29,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 808
},
{
"epoch": 270.58,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 812
},
{
"epoch": 271.87,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 816
},
{
"epoch": 273.29,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 820
},
{
"epoch": 274.58,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 824
},
{
"epoch": 275.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 828
},
{
"epoch": 277.29,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 832
},
{
"epoch": 277.29,
"eval_exact_match": 0.6121856866537717,
"eval_loss": 0.38368162512779236,
"eval_runtime": 198.324,
"eval_samples_per_second": 5.214,
"step": 832
},
{
"epoch": 278.58,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 836
},
{
"epoch": 279.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 840
},
{
"epoch": 281.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 844
},
{
"epoch": 282.58,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 848
},
{
"epoch": 283.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 852
},
{
"epoch": 285.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 856
},
{
"epoch": 286.58,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 860
},
{
"epoch": 287.87,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 864
},
{
"epoch": 289.29,
"learning_rate": 0.0001,
"loss": 0.0022,
"step": 868
},
{
"epoch": 290.58,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 872
},
{
"epoch": 291.87,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 876
},
{
"epoch": 293.29,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 880
},
{
"epoch": 294.58,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 884
},
{
"epoch": 295.87,
"learning_rate": 0.0001,
"loss": 0.0052,
"step": 888
},
{
"epoch": 297.29,
"learning_rate": 0.0001,
"loss": 0.0031,
"step": 892
},
{
"epoch": 298.58,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 896
},
{
"epoch": 298.58,
"eval_exact_match": 0.6266924564796905,
"eval_loss": 0.36901429295539856,
"eval_runtime": 197.2225,
"eval_samples_per_second": 5.243,
"step": 896
},
{
"epoch": 299.87,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 900
},
{
"epoch": 301.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 904
},
{
"epoch": 302.58,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 908
},
{
"epoch": 303.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 912
},
{
"epoch": 305.29,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 916
},
{
"epoch": 306.58,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 920
},
{
"epoch": 307.87,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 924
},
{
"epoch": 309.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 928
},
{
"epoch": 310.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 932
},
{
"epoch": 311.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 936
},
{
"epoch": 313.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 940
},
{
"epoch": 314.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 944
},
{
"epoch": 315.87,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 948
},
{
"epoch": 317.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 952
},
{
"epoch": 318.58,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 956
},
{
"epoch": 319.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 960
},
{
"epoch": 319.87,
"eval_exact_match": 0.6160541586073501,
"eval_loss": 0.3921535909175873,
"eval_runtime": 198.7943,
"eval_samples_per_second": 5.201,
"step": 960
},
{
"epoch": 321.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 964
},
{
"epoch": 322.58,
"learning_rate": 0.0001,
"loss": 0.0026,
"step": 968
},
{
"epoch": 323.87,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 972
},
{
"epoch": 325.29,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 976
},
{
"epoch": 326.58,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 980
},
{
"epoch": 327.87,
"learning_rate": 0.0001,
"loss": 0.0017,
"step": 984
},
{
"epoch": 329.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 988
},
{
"epoch": 330.58,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 992
},
{
"epoch": 331.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 996
},
{
"epoch": 333.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1000
},
{
"epoch": 334.58,
"learning_rate": 0.0001,
"loss": 0.005,
"step": 1004
},
{
"epoch": 335.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 1008
},
{
"epoch": 337.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1012
},
{
"epoch": 338.58,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 1016
},
{
"epoch": 339.87,
"learning_rate": 0.0001,
"loss": 0.0038,
"step": 1020
},
{
"epoch": 341.29,
"learning_rate": 0.0001,
"loss": 0.0054,
"step": 1024
},
{
"epoch": 341.29,
"eval_exact_match": 0.6470019342359767,
"eval_loss": 0.3395414650440216,
"eval_runtime": 199.1545,
"eval_samples_per_second": 5.192,
"step": 1024
},
{
"epoch": 342.58,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 1028
},
{
"epoch": 343.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 1032
},
{
"epoch": 345.29,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1036
},
{
"epoch": 346.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1040
},
{
"epoch": 347.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1044
},
{
"epoch": 349.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1048
},
{
"epoch": 350.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1052
},
{
"epoch": 351.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1056
},
{
"epoch": 353.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 1060
},
{
"epoch": 354.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1064
},
{
"epoch": 355.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 1068
},
{
"epoch": 357.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 1072
},
{
"epoch": 358.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1076
},
{
"epoch": 359.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1080
},
{
"epoch": 361.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 1084
},
{
"epoch": 362.58,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1088
},
{
"epoch": 362.58,
"eval_exact_match": 0.6315280464216635,
"eval_loss": 0.3889999985694885,
"eval_runtime": 196.7486,
"eval_samples_per_second": 5.255,
"step": 1088
},
{
"epoch": 363.87,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 1092
},
{
"epoch": 365.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1096
},
{
"epoch": 366.58,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 1100
},
{
"epoch": 367.87,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1104
},
{
"epoch": 369.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1108
},
{
"epoch": 370.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1112
},
{
"epoch": 371.87,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 1116
},
{
"epoch": 373.29,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 1120
},
{
"epoch": 374.58,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1124
},
{
"epoch": 375.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1128
},
{
"epoch": 377.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1132
},
{
"epoch": 378.58,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1136
},
{
"epoch": 379.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1140
},
{
"epoch": 381.29,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1144
},
{
"epoch": 382.58,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1148
},
{
"epoch": 383.87,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 1152
},
{
"epoch": 383.87,
"eval_exact_match": 0.6392649903288201,
"eval_loss": 0.3856046497821808,
"eval_runtime": 188.1717,
"eval_samples_per_second": 5.495,
"step": 1152
},
{
"epoch": 385.29,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1156
},
{
"epoch": 386.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1160
},
{
"epoch": 387.87,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 1164
},
{
"epoch": 389.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 1168
},
{
"epoch": 390.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1172
},
{
"epoch": 391.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1176
},
{
"epoch": 393.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1180
},
{
"epoch": 394.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1184
},
{
"epoch": 395.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1188
},
{
"epoch": 397.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1192
},
{
"epoch": 398.58,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 1196
},
{
"epoch": 399.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1200
},
{
"epoch": 401.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1204
},
{
"epoch": 402.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1208
},
{
"epoch": 403.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1212
},
{
"epoch": 405.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1216
},
{
"epoch": 405.29,
"eval_exact_match": 0.6315280464216635,
"eval_loss": 0.40906959772109985,
"eval_runtime": 198.153,
"eval_samples_per_second": 5.218,
"step": 1216
},
{
"epoch": 406.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1220
},
{
"epoch": 407.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1224
},
{
"epoch": 409.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1228
},
{
"epoch": 410.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1232
},
{
"epoch": 411.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1236
},
{
"epoch": 413.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1240
},
{
"epoch": 414.58,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 1244
},
{
"epoch": 415.87,
"learning_rate": 0.0001,
"loss": 0.0012,
"step": 1248
},
{
"epoch": 417.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1252
},
{
"epoch": 418.58,
"learning_rate": 0.0001,
"loss": 0.0042,
"step": 1256
},
{
"epoch": 419.87,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 1260
},
{
"epoch": 421.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1264
},
{
"epoch": 422.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1268
},
{
"epoch": 423.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1272
},
{
"epoch": 425.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1276
},
{
"epoch": 426.58,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1280
},
{
"epoch": 426.58,
"eval_exact_match": 0.6073500967117988,
"eval_loss": 0.3927420973777771,
"eval_runtime": 197.5888,
"eval_samples_per_second": 5.233,
"step": 1280
},
{
"epoch": 427.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1284
},
{
"epoch": 429.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1288
},
{
"epoch": 430.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1292
},
{
"epoch": 431.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1296
},
{
"epoch": 433.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1300
},
{
"epoch": 434.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1304
},
{
"epoch": 435.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1308
},
{
"epoch": 437.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 1312
},
{
"epoch": 438.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1316
},
{
"epoch": 439.87,
"learning_rate": 0.0001,
"loss": 0.0075,
"step": 1320
},
{
"epoch": 441.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 1324
},
{
"epoch": 442.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1328
},
{
"epoch": 443.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1332
},
{
"epoch": 445.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1336
},
{
"epoch": 446.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1340
},
{
"epoch": 447.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1344
},
{
"epoch": 447.87,
"eval_exact_match": 0.6141199226305609,
"eval_loss": 0.39868542551994324,
"eval_runtime": 191.3189,
"eval_samples_per_second": 5.405,
"step": 1344
},
{
"epoch": 449.29,
"learning_rate": 0.0001,
"loss": 0.0015,
"step": 1348
},
{
"epoch": 450.58,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1352
},
{
"epoch": 451.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1356
},
{
"epoch": 453.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1360
},
{
"epoch": 454.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1364
},
{
"epoch": 455.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1368
},
{
"epoch": 457.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1372
},
{
"epoch": 458.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1376
},
{
"epoch": 459.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1380
},
{
"epoch": 461.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1384
},
{
"epoch": 462.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1388
},
{
"epoch": 463.87,
"learning_rate": 0.0001,
"loss": 0.0028,
"step": 1392
},
{
"epoch": 465.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1396
},
{
"epoch": 466.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1400
},
{
"epoch": 467.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1404
},
{
"epoch": 469.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1408
},
{
"epoch": 469.29,
"eval_exact_match": 0.6131528046421664,
"eval_loss": 0.41037216782569885,
"eval_runtime": 178.412,
"eval_samples_per_second": 5.796,
"step": 1408
},
{
"epoch": 470.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1412
},
{
"epoch": 471.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1416
},
{
"epoch": 473.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1420
},
{
"epoch": 474.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1424
},
{
"epoch": 475.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1428
},
{
"epoch": 477.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1432
},
{
"epoch": 478.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1436
},
{
"epoch": 479.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1440
},
{
"epoch": 481.29,
"learning_rate": 0.0001,
"loss": 0.0026,
"step": 1444
},
{
"epoch": 482.58,
"learning_rate": 0.0001,
"loss": 0.0197,
"step": 1448
},
{
"epoch": 483.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1452
},
{
"epoch": 485.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1456
},
{
"epoch": 486.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1460
},
{
"epoch": 487.87,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 1464
},
{
"epoch": 489.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1468
},
{
"epoch": 490.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1472
},
{
"epoch": 490.58,
"eval_exact_match": 0.620889748549323,
"eval_loss": 0.40761619806289673,
"eval_runtime": 192.2803,
"eval_samples_per_second": 5.378,
"step": 1472
},
{
"epoch": 491.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1476
},
{
"epoch": 493.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1480
},
{
"epoch": 494.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1484
},
{
"epoch": 495.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1488
},
{
"epoch": 497.29,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 1492
},
{
"epoch": 498.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1496
},
{
"epoch": 499.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1500
},
{
"epoch": 501.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1504
},
{
"epoch": 502.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1508
},
{
"epoch": 503.87,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 1512
},
{
"epoch": 505.29,
"learning_rate": 0.0001,
"loss": 0.0021,
"step": 1516
},
{
"epoch": 506.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1520
},
{
"epoch": 507.87,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 1524
},
{
"epoch": 509.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1528
},
{
"epoch": 510.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1532
},
{
"epoch": 511.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1536
},
{
"epoch": 511.87,
"eval_exact_match": 0.6189555125725339,
"eval_loss": 0.4127079248428345,
"eval_runtime": 194.9448,
"eval_samples_per_second": 5.304,
"step": 1536
},
{
"epoch": 513.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1540
},
{
"epoch": 514.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1544
},
{
"epoch": 515.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1548
},
{
"epoch": 517.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1552
},
{
"epoch": 518.58,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1556
},
{
"epoch": 519.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1560
},
{
"epoch": 521.29,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1564
},
{
"epoch": 522.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1568
},
{
"epoch": 523.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1572
},
{
"epoch": 525.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1576
},
{
"epoch": 526.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1580
},
{
"epoch": 527.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1584
},
{
"epoch": 529.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1588
},
{
"epoch": 530.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1592
},
{
"epoch": 531.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1596
},
{
"epoch": 533.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1600
},
{
"epoch": 533.29,
"eval_exact_match": 0.6266924564796905,
"eval_loss": 0.4237005412578583,
"eval_runtime": 188.9181,
"eval_samples_per_second": 5.473,
"step": 1600
},
{
"epoch": 534.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1604
},
{
"epoch": 535.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1608
},
{
"epoch": 537.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1612
},
{
"epoch": 538.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1616
},
{
"epoch": 539.87,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 1620
},
{
"epoch": 541.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1624
},
{
"epoch": 542.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1628
},
{
"epoch": 543.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1632
},
{
"epoch": 545.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1636
},
{
"epoch": 546.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1640
},
{
"epoch": 547.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 1644
},
{
"epoch": 549.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1648
},
{
"epoch": 550.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1652
},
{
"epoch": 551.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1656
},
{
"epoch": 553.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1660
},
{
"epoch": 554.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1664
},
{
"epoch": 554.58,
"eval_exact_match": 0.6179883945841392,
"eval_loss": 0.41781872510910034,
"eval_runtime": 190.2976,
"eval_samples_per_second": 5.434,
"step": 1664
},
{
"epoch": 555.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1668
},
{
"epoch": 557.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1672
},
{
"epoch": 558.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1676
},
{
"epoch": 559.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1680
},
{
"epoch": 561.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1684
},
{
"epoch": 562.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1688
},
{
"epoch": 563.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1692
},
{
"epoch": 565.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1696
},
{
"epoch": 566.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1700
},
{
"epoch": 567.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1704
},
{
"epoch": 569.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 1708
},
{
"epoch": 570.58,
"learning_rate": 0.0001,
"loss": 0.0028,
"step": 1712
},
{
"epoch": 571.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1716
},
{
"epoch": 573.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1720
},
{
"epoch": 574.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1724
},
{
"epoch": 575.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1728
},
{
"epoch": 575.87,
"eval_exact_match": 0.632495164410058,
"eval_loss": 0.4037138521671295,
"eval_runtime": 202.986,
"eval_samples_per_second": 5.094,
"step": 1728
},
{
"epoch": 577.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1732
},
{
"epoch": 578.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1736
},
{
"epoch": 579.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1740
},
{
"epoch": 581.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1744
},
{
"epoch": 582.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1748
},
{
"epoch": 583.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1752
},
{
"epoch": 585.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1756
},
{
"epoch": 586.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1760
},
{
"epoch": 587.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1764
},
{
"epoch": 589.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1768
},
{
"epoch": 590.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1772
},
{
"epoch": 591.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1776
},
{
"epoch": 593.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1780
},
{
"epoch": 594.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1784
},
{
"epoch": 595.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1788
},
{
"epoch": 597.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1792
},
{
"epoch": 597.29,
"eval_exact_match": 0.620889748549323,
"eval_loss": 0.4223831295967102,
"eval_runtime": 185.5321,
"eval_samples_per_second": 5.573,
"step": 1792
},
{
"epoch": 598.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1796
},
{
"epoch": 599.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1800
},
{
"epoch": 601.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1804
},
{
"epoch": 602.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1808
},
{
"epoch": 603.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1812
},
{
"epoch": 605.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1816
},
{
"epoch": 606.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1820
},
{
"epoch": 607.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1824
},
{
"epoch": 609.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1828
},
{
"epoch": 610.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1832
},
{
"epoch": 611.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1836
},
{
"epoch": 613.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1840
},
{
"epoch": 614.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1844
},
{
"epoch": 615.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1848
},
{
"epoch": 617.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1852
},
{
"epoch": 618.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1856
},
{
"epoch": 618.58,
"eval_exact_match": 0.6286266924564797,
"eval_loss": 0.4336487054824829,
"eval_runtime": 195.847,
"eval_samples_per_second": 5.28,
"step": 1856
},
{
"epoch": 619.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1860
},
{
"epoch": 621.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1864
},
{
"epoch": 622.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1868
},
{
"epoch": 623.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1872
},
{
"epoch": 625.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1876
},
{
"epoch": 626.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1880
},
{
"epoch": 627.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1884
},
{
"epoch": 629.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1888
},
{
"epoch": 630.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1892
},
{
"epoch": 631.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 1896
},
{
"epoch": 633.29,
"learning_rate": 0.0001,
"loss": 0.0046,
"step": 1900
},
{
"epoch": 634.58,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 1904
},
{
"epoch": 635.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1908
},
{
"epoch": 637.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1912
},
{
"epoch": 638.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1916
},
{
"epoch": 639.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1920
},
{
"epoch": 639.87,
"eval_exact_match": 0.6295938104448743,
"eval_loss": 0.39955005049705505,
"eval_runtime": 187.1183,
"eval_samples_per_second": 5.526,
"step": 1920
},
{
"epoch": 641.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1924
},
{
"epoch": 642.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1928
},
{
"epoch": 643.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1932
},
{
"epoch": 645.29,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 1936
},
{
"epoch": 646.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1940
},
{
"epoch": 647.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1944
},
{
"epoch": 649.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1948
},
{
"epoch": 650.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1952
},
{
"epoch": 651.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 1956
},
{
"epoch": 653.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 1960
},
{
"epoch": 654.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1964
},
{
"epoch": 655.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1968
},
{
"epoch": 657.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 1972
},
{
"epoch": 658.58,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 1976
},
{
"epoch": 659.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1980
},
{
"epoch": 661.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1984
},
{
"epoch": 661.29,
"eval_exact_match": 0.6276595744680851,
"eval_loss": 0.4064112603664398,
"eval_runtime": 194.3821,
"eval_samples_per_second": 5.319,
"step": 1984
},
{
"epoch": 662.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1988
},
{
"epoch": 663.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 1992
},
{
"epoch": 665.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 1996
},
{
"epoch": 666.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2000
},
{
"epoch": 667.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2004
},
{
"epoch": 669.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 2008
},
{
"epoch": 670.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2012
},
{
"epoch": 671.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2016
},
{
"epoch": 673.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2020
},
{
"epoch": 674.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 2024
},
{
"epoch": 675.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2028
},
{
"epoch": 677.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2032
},
{
"epoch": 678.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2036
},
{
"epoch": 679.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2040
},
{
"epoch": 681.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2044
},
{
"epoch": 682.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2048
},
{
"epoch": 682.58,
"eval_exact_match": 0.6160541586073501,
"eval_loss": 0.4323204755783081,
"eval_runtime": 197.4132,
"eval_samples_per_second": 5.238,
"step": 2048
},
{
"epoch": 683.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2052
},
{
"epoch": 685.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2056
},
{
"epoch": 686.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2060
},
{
"epoch": 687.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2064
},
{
"epoch": 689.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2068
},
{
"epoch": 690.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2072
},
{
"epoch": 691.87,
"learning_rate": 0.0001,
"loss": 0.0016,
"step": 2076
},
{
"epoch": 693.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2080
},
{
"epoch": 694.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2084
},
{
"epoch": 695.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2088
},
{
"epoch": 697.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2092
},
{
"epoch": 698.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2096
},
{
"epoch": 699.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2100
},
{
"epoch": 701.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2104
},
{
"epoch": 702.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2108
},
{
"epoch": 703.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2112
},
{
"epoch": 703.87,
"eval_exact_match": 0.6228239845261122,
"eval_loss": 0.4344477951526642,
"eval_runtime": 180.0925,
"eval_samples_per_second": 5.741,
"step": 2112
},
{
"epoch": 705.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2116
},
{
"epoch": 706.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2120
},
{
"epoch": 707.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2124
},
{
"epoch": 709.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 2128
},
{
"epoch": 710.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2132
},
{
"epoch": 711.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2136
},
{
"epoch": 713.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2140
},
{
"epoch": 714.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2144
},
{
"epoch": 715.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2148
},
{
"epoch": 717.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 2152
},
{
"epoch": 718.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2156
},
{
"epoch": 719.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2160
},
{
"epoch": 721.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 2164
},
{
"epoch": 722.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2168
},
{
"epoch": 723.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 2172
},
{
"epoch": 725.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2176
},
{
"epoch": 725.29,
"eval_exact_match": 0.6141199226305609,
"eval_loss": 0.42048418521881104,
"eval_runtime": 191.4987,
"eval_samples_per_second": 5.4,
"step": 2176
},
{
"epoch": 726.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2180
},
{
"epoch": 727.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2184
},
{
"epoch": 729.29,
"learning_rate": 0.0001,
"loss": 0.0056,
"step": 2188
},
{
"epoch": 730.58,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 2192
},
{
"epoch": 731.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2196
},
{
"epoch": 733.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 2200
},
{
"epoch": 734.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2204
},
{
"epoch": 735.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2208
},
{
"epoch": 737.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2212
},
{
"epoch": 738.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2216
},
{
"epoch": 739.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2220
},
{
"epoch": 741.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2224
},
{
"epoch": 742.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2228
},
{
"epoch": 743.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2232
},
{
"epoch": 745.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2236
},
{
"epoch": 746.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2240
},
{
"epoch": 746.58,
"eval_exact_match": 0.6199226305609284,
"eval_loss": 0.42163723707199097,
"eval_runtime": 193.8256,
"eval_samples_per_second": 5.335,
"step": 2240
},
{
"epoch": 747.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2244
},
{
"epoch": 749.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2248
},
{
"epoch": 750.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2252
},
{
"epoch": 751.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2256
},
{
"epoch": 753.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2260
},
{
"epoch": 754.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2264
},
{
"epoch": 755.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2268
},
{
"epoch": 757.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2272
},
{
"epoch": 758.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2276
},
{
"epoch": 759.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2280
},
{
"epoch": 761.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2284
},
{
"epoch": 762.58,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 2288
},
{
"epoch": 763.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2292
},
{
"epoch": 765.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2296
},
{
"epoch": 766.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2300
},
{
"epoch": 767.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2304
},
{
"epoch": 767.87,
"eval_exact_match": 0.6247582205029013,
"eval_loss": 0.43437090516090393,
"eval_runtime": 182.8367,
"eval_samples_per_second": 5.655,
"step": 2304
},
{
"epoch": 769.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2308
},
{
"epoch": 770.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2312
},
{
"epoch": 771.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2316
},
{
"epoch": 773.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2320
},
{
"epoch": 774.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 2324
},
{
"epoch": 775.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2328
},
{
"epoch": 777.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2332
},
{
"epoch": 778.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2336
},
{
"epoch": 779.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2340
},
{
"epoch": 781.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2344
},
{
"epoch": 782.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2348
},
{
"epoch": 783.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2352
},
{
"epoch": 785.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2356
},
{
"epoch": 786.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2360
},
{
"epoch": 787.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2364
},
{
"epoch": 789.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2368
},
{
"epoch": 789.29,
"eval_exact_match": 0.6247582205029013,
"eval_loss": 0.44927769899368286,
"eval_runtime": 198.3875,
"eval_samples_per_second": 5.212,
"step": 2368
},
{
"epoch": 790.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2372
},
{
"epoch": 791.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2376
},
{
"epoch": 793.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 2380
},
{
"epoch": 794.58,
"learning_rate": 0.0001,
"loss": 0.0058,
"step": 2384
},
{
"epoch": 795.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 2388
},
{
"epoch": 797.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2392
},
{
"epoch": 798.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2396
},
{
"epoch": 799.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2400
},
{
"epoch": 801.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2404
},
{
"epoch": 802.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2408
},
{
"epoch": 803.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2412
},
{
"epoch": 805.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2416
},
{
"epoch": 806.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2420
},
{
"epoch": 807.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2424
},
{
"epoch": 809.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2428
},
{
"epoch": 810.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2432
},
{
"epoch": 810.58,
"eval_exact_match": 0.6266924564796905,
"eval_loss": 0.43350014090538025,
"eval_runtime": 195.7441,
"eval_samples_per_second": 5.282,
"step": 2432
},
{
"epoch": 811.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 2436
},
{
"epoch": 813.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2440
},
{
"epoch": 814.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2444
},
{
"epoch": 815.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2448
},
{
"epoch": 817.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2452
},
{
"epoch": 818.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2456
},
{
"epoch": 819.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2460
},
{
"epoch": 821.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2464
},
{
"epoch": 822.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2468
},
{
"epoch": 823.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2472
},
{
"epoch": 825.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2476
},
{
"epoch": 826.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2480
},
{
"epoch": 827.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2484
},
{
"epoch": 829.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2488
},
{
"epoch": 830.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2492
},
{
"epoch": 831.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2496
},
{
"epoch": 831.87,
"eval_exact_match": 0.6392649903288201,
"eval_loss": 0.4426421821117401,
"eval_runtime": 200.6137,
"eval_samples_per_second": 5.154,
"step": 2496
},
{
"epoch": 833.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2500
},
{
"epoch": 834.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2504
},
{
"epoch": 835.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2508
},
{
"epoch": 837.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2512
},
{
"epoch": 838.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2516
},
{
"epoch": 839.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2520
},
{
"epoch": 841.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2524
},
{
"epoch": 842.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2528
},
{
"epoch": 843.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2532
},
{
"epoch": 845.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 2536
},
{
"epoch": 846.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2540
},
{
"epoch": 847.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2544
},
{
"epoch": 849.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 2548
},
{
"epoch": 850.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2552
},
{
"epoch": 851.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2556
},
{
"epoch": 853.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2560
},
{
"epoch": 853.29,
"eval_exact_match": 0.6373307543520309,
"eval_loss": 0.43845412135124207,
"eval_runtime": 215.8522,
"eval_samples_per_second": 4.79,
"step": 2560
},
{
"epoch": 854.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2564
},
{
"epoch": 855.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2568
},
{
"epoch": 857.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 2572
},
{
"epoch": 858.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2576
},
{
"epoch": 859.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2580
},
{
"epoch": 861.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2584
},
{
"epoch": 862.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2588
},
{
"epoch": 863.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2592
},
{
"epoch": 865.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2596
},
{
"epoch": 866.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2600
},
{
"epoch": 867.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2604
},
{
"epoch": 869.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2608
},
{
"epoch": 870.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2612
},
{
"epoch": 871.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2616
},
{
"epoch": 873.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2620
},
{
"epoch": 874.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2624
},
{
"epoch": 874.58,
"eval_exact_match": 0.6315280464216635,
"eval_loss": 0.455252081155777,
"eval_runtime": 209.6498,
"eval_samples_per_second": 4.932,
"step": 2624
},
{
"epoch": 875.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2628
},
{
"epoch": 877.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2632
},
{
"epoch": 878.58,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 2636
},
{
"epoch": 879.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2640
},
{
"epoch": 881.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2644
},
{
"epoch": 882.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2648
},
{
"epoch": 883.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2652
},
{
"epoch": 885.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2656
},
{
"epoch": 886.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2660
},
{
"epoch": 887.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2664
},
{
"epoch": 889.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 2668
},
{
"epoch": 890.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2672
},
{
"epoch": 891.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2676
},
{
"epoch": 893.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2680
},
{
"epoch": 894.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 2684
},
{
"epoch": 895.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2688
},
{
"epoch": 895.87,
"eval_exact_match": 0.632495164410058,
"eval_loss": 0.4402284026145935,
"eval_runtime": 210.5896,
"eval_samples_per_second": 4.91,
"step": 2688
},
{
"epoch": 897.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2692
},
{
"epoch": 898.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2696
},
{
"epoch": 899.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2700
},
{
"epoch": 901.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2704
},
{
"epoch": 902.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2708
},
{
"epoch": 903.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2712
},
{
"epoch": 905.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2716
},
{
"epoch": 906.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2720
},
{
"epoch": 907.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2724
},
{
"epoch": 909.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2728
},
{
"epoch": 910.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2732
},
{
"epoch": 911.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2736
},
{
"epoch": 913.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2740
},
{
"epoch": 914.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2744
},
{
"epoch": 915.87,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 2748
},
{
"epoch": 917.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2752
},
{
"epoch": 917.29,
"eval_exact_match": 0.625725338491296,
"eval_loss": 0.45130491256713867,
"eval_runtime": 203.4647,
"eval_samples_per_second": 5.082,
"step": 2752
},
{
"epoch": 918.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2756
},
{
"epoch": 919.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2760
},
{
"epoch": 921.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2764
},
{
"epoch": 922.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2768
},
{
"epoch": 923.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2772
},
{
"epoch": 925.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2776
},
{
"epoch": 926.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2780
},
{
"epoch": 927.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2784
},
{
"epoch": 929.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2788
},
{
"epoch": 930.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2792
},
{
"epoch": 931.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2796
},
{
"epoch": 933.29,
"learning_rate": 0.0001,
"loss": 0.0022,
"step": 2800
},
{
"epoch": 934.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2804
},
{
"epoch": 935.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2808
},
{
"epoch": 937.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2812
},
{
"epoch": 938.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2816
},
{
"epoch": 938.58,
"eval_exact_match": 0.6344294003868471,
"eval_loss": 0.4340243935585022,
"eval_runtime": 218.0667,
"eval_samples_per_second": 4.742,
"step": 2816
},
{
"epoch": 939.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2820
},
{
"epoch": 941.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2824
},
{
"epoch": 942.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2828
},
{
"epoch": 943.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2832
},
{
"epoch": 945.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2836
},
{
"epoch": 946.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2840
},
{
"epoch": 947.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2844
},
{
"epoch": 949.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2848
},
{
"epoch": 950.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2852
},
{
"epoch": 951.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2856
},
{
"epoch": 953.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2860
},
{
"epoch": 954.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2864
},
{
"epoch": 955.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2868
},
{
"epoch": 957.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2872
},
{
"epoch": 958.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2876
},
{
"epoch": 959.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2880
},
{
"epoch": 959.87,
"eval_exact_match": 0.6295938104448743,
"eval_loss": 0.4441409111022949,
"eval_runtime": 199.2,
"eval_samples_per_second": 5.191,
"step": 2880
},
{
"epoch": 961.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2884
},
{
"epoch": 962.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 2888
},
{
"epoch": 963.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2892
},
{
"epoch": 965.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2896
},
{
"epoch": 966.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 2900
},
{
"epoch": 967.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2904
},
{
"epoch": 969.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2908
},
{
"epoch": 970.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2912
},
{
"epoch": 971.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2916
},
{
"epoch": 973.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2920
},
{
"epoch": 974.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2924
},
{
"epoch": 975.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2928
},
{
"epoch": 977.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2932
},
{
"epoch": 978.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 2936
},
{
"epoch": 979.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2940
},
{
"epoch": 981.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 2944
},
{
"epoch": 981.29,
"eval_exact_match": 0.625725338491296,
"eval_loss": 0.4419667720794678,
"eval_runtime": 191.9537,
"eval_samples_per_second": 5.387,
"step": 2944
},
{
"epoch": 982.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2948
},
{
"epoch": 983.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2952
},
{
"epoch": 985.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2956
},
{
"epoch": 986.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2960
},
{
"epoch": 987.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 2964
},
{
"epoch": 989.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2968
},
{
"epoch": 990.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 2972
},
{
"epoch": 991.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2976
},
{
"epoch": 993.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2980
},
{
"epoch": 994.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2984
},
{
"epoch": 995.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 2988
},
{
"epoch": 997.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 2992
},
{
"epoch": 998.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 2996
},
{
"epoch": 999.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3000
},
{
"epoch": 1001.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3004
},
{
"epoch": 1002.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3008
},
{
"epoch": 1002.58,
"eval_exact_match": 0.6363636363636364,
"eval_loss": 0.43674689531326294,
"eval_runtime": 192.5462,
"eval_samples_per_second": 5.37,
"step": 3008
},
{
"epoch": 1003.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3012
},
{
"epoch": 1005.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3016
},
{
"epoch": 1006.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3020
},
{
"epoch": 1007.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3024
},
{
"epoch": 1009.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3028
},
{
"epoch": 1010.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3032
},
{
"epoch": 1011.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3036
},
{
"epoch": 1013.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3040
},
{
"epoch": 1014.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 3044
},
{
"epoch": 1015.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3048
},
{
"epoch": 1017.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3052
},
{
"epoch": 1018.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3056
},
{
"epoch": 1019.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3060
},
{
"epoch": 1021.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3064
},
{
"epoch": 1022.58,
"learning_rate": 0.0001,
"loss": 0.0033,
"step": 3068
},
{
"epoch": 1023.87,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 3072
},
{
"epoch": 1023.87,
"eval_exact_match": 0.6295938104448743,
"eval_loss": 0.4049379229545593,
"eval_runtime": 183.2179,
"eval_samples_per_second": 5.644,
"step": 3072
},
{
"epoch": 1025.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3076
},
{
"epoch": 1026.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3080
},
{
"epoch": 1027.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3084
},
{
"epoch": 1029.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3088
},
{
"epoch": 1030.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3092
},
{
"epoch": 1031.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3096
},
{
"epoch": 1033.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3100
},
{
"epoch": 1034.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3104
},
{
"epoch": 1035.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3108
},
{
"epoch": 1037.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3112
},
{
"epoch": 1038.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3116
},
{
"epoch": 1039.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3120
},
{
"epoch": 1041.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3124
},
{
"epoch": 1042.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3128
},
{
"epoch": 1043.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3132
},
{
"epoch": 1045.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3136
},
{
"epoch": 1045.29,
"eval_exact_match": 0.6102514506769826,
"eval_loss": 0.4436005651950836,
"eval_runtime": 189.9177,
"eval_samples_per_second": 5.444,
"step": 3136
},
{
"epoch": 1046.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3140
},
{
"epoch": 1047.87,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 3144
},
{
"epoch": 1049.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3148
},
{
"epoch": 1050.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3152
},
{
"epoch": 1051.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3156
},
{
"epoch": 1053.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3160
},
{
"epoch": 1054.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3164
},
{
"epoch": 1055.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 3168
},
{
"epoch": 1057.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3172
},
{
"epoch": 1058.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3176
},
{
"epoch": 1059.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3180
},
{
"epoch": 1061.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3184
},
{
"epoch": 1062.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3188
},
{
"epoch": 1063.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3192
},
{
"epoch": 1065.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3196
},
{
"epoch": 1066.58,
"learning_rate": 0.0001,
"loss": 0.0026,
"step": 3200
},
{
"epoch": 1066.58,
"eval_exact_match": 0.6179883945841392,
"eval_loss": 0.4263126850128174,
"eval_runtime": 191.6463,
"eval_samples_per_second": 5.395,
"step": 3200
},
{
"epoch": 1067.87,
"learning_rate": 0.0001,
"loss": 0.0025,
"step": 3204
},
{
"epoch": 1069.29,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 3208
},
{
"epoch": 1070.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3212
},
{
"epoch": 1071.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3216
},
{
"epoch": 1073.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3220
},
{
"epoch": 1074.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3224
},
{
"epoch": 1075.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3228
},
{
"epoch": 1077.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3232
},
{
"epoch": 1078.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3236
},
{
"epoch": 1079.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3240
},
{
"epoch": 1081.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3244
},
{
"epoch": 1082.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3248
},
{
"epoch": 1083.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3252
},
{
"epoch": 1085.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3256
},
{
"epoch": 1086.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3260
},
{
"epoch": 1087.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3264
},
{
"epoch": 1087.87,
"eval_exact_match": 0.6286266924564797,
"eval_loss": 0.4278923273086548,
"eval_runtime": 198.7655,
"eval_samples_per_second": 5.202,
"step": 3264
},
{
"epoch": 1089.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3268
},
{
"epoch": 1090.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3272
},
{
"epoch": 1091.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3276
},
{
"epoch": 1093.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3280
},
{
"epoch": 1094.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3284
},
{
"epoch": 1095.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3288
},
{
"epoch": 1097.29,
"learning_rate": 0.0001,
"loss": 0.001,
"step": 3292
},
{
"epoch": 1098.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3296
},
{
"epoch": 1099.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3300
},
{
"epoch": 1101.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3304
},
{
"epoch": 1102.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3308
},
{
"epoch": 1103.87,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 3312
},
{
"epoch": 1105.29,
"learning_rate": 0.0001,
"loss": 0.0018,
"step": 3316
},
{
"epoch": 1106.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3320
},
{
"epoch": 1107.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3324
},
{
"epoch": 1109.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3328
},
{
"epoch": 1109.29,
"eval_exact_match": 0.620889748549323,
"eval_loss": 0.4213942289352417,
"eval_runtime": 195.4858,
"eval_samples_per_second": 5.289,
"step": 3328
},
{
"epoch": 1110.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3332
},
{
"epoch": 1111.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3336
},
{
"epoch": 1113.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3340
},
{
"epoch": 1114.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3344
},
{
"epoch": 1115.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3348
},
{
"epoch": 1117.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3352
},
{
"epoch": 1118.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3356
},
{
"epoch": 1119.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3360
},
{
"epoch": 1121.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3364
},
{
"epoch": 1122.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3368
},
{
"epoch": 1123.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3372
},
{
"epoch": 1125.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3376
},
{
"epoch": 1126.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3380
},
{
"epoch": 1127.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3384
},
{
"epoch": 1129.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3388
},
{
"epoch": 1130.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3392
},
{
"epoch": 1130.58,
"eval_exact_match": 0.6276595744680851,
"eval_loss": 0.4281994104385376,
"eval_runtime": 190.1723,
"eval_samples_per_second": 5.437,
"step": 3392
},
{
"epoch": 1131.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3396
},
{
"epoch": 1133.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3400
},
{
"epoch": 1134.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3404
},
{
"epoch": 1135.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3408
},
{
"epoch": 1137.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3412
},
{
"epoch": 1138.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3416
},
{
"epoch": 1139.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3420
},
{
"epoch": 1141.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3424
},
{
"epoch": 1142.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 3428
},
{
"epoch": 1143.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3432
},
{
"epoch": 1145.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3436
},
{
"epoch": 1146.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3440
},
{
"epoch": 1147.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3444
},
{
"epoch": 1149.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3448
},
{
"epoch": 1150.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 3452
},
{
"epoch": 1151.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3456
},
{
"epoch": 1151.87,
"eval_exact_match": 0.6150870406189555,
"eval_loss": 0.4299115240573883,
"eval_runtime": 193.9237,
"eval_samples_per_second": 5.332,
"step": 3456
},
{
"epoch": 1153.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 3460
},
{
"epoch": 1154.58,
"learning_rate": 0.0001,
"loss": 0.0038,
"step": 3464
},
{
"epoch": 1155.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3468
},
{
"epoch": 1157.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3472
},
{
"epoch": 1158.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3476
},
{
"epoch": 1159.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3480
},
{
"epoch": 1161.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3484
},
{
"epoch": 1162.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3488
},
{
"epoch": 1163.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3492
},
{
"epoch": 1165.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3496
},
{
"epoch": 1166.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3500
},
{
"epoch": 1167.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3504
},
{
"epoch": 1169.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3508
},
{
"epoch": 1170.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3512
},
{
"epoch": 1171.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3516
},
{
"epoch": 1173.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3520
},
{
"epoch": 1173.29,
"eval_exact_match": 0.6228239845261122,
"eval_loss": 0.44940173625946045,
"eval_runtime": 196.4426,
"eval_samples_per_second": 5.264,
"step": 3520
},
{
"epoch": 1174.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3524
},
{
"epoch": 1175.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3528
},
{
"epoch": 1177.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3532
},
{
"epoch": 1178.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3536
},
{
"epoch": 1179.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3540
},
{
"epoch": 1181.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3544
},
{
"epoch": 1182.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3548
},
{
"epoch": 1183.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3552
},
{
"epoch": 1185.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3556
},
{
"epoch": 1186.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3560
},
{
"epoch": 1187.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3564
},
{
"epoch": 1189.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3568
},
{
"epoch": 1190.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3572
},
{
"epoch": 1191.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3576
},
{
"epoch": 1193.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3580
},
{
"epoch": 1194.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3584
},
{
"epoch": 1194.58,
"eval_exact_match": 0.6286266924564797,
"eval_loss": 0.4586807191371918,
"eval_runtime": 199.1549,
"eval_samples_per_second": 5.192,
"step": 3584
},
{
"epoch": 1195.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3588
},
{
"epoch": 1197.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3592
},
{
"epoch": 1198.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3596
},
{
"epoch": 1199.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3600
},
{
"epoch": 1201.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3604
},
{
"epoch": 1202.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3608
},
{
"epoch": 1203.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3612
},
{
"epoch": 1205.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3616
},
{
"epoch": 1206.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3620
},
{
"epoch": 1207.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3624
},
{
"epoch": 1209.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3628
},
{
"epoch": 1210.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3632
},
{
"epoch": 1211.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3636
},
{
"epoch": 1213.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3640
},
{
"epoch": 1214.58,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 3644
},
{
"epoch": 1215.87,
"learning_rate": 0.0001,
"loss": 0.0007,
"step": 3648
},
{
"epoch": 1215.87,
"eval_exact_match": 0.6295938104448743,
"eval_loss": 0.41312792897224426,
"eval_runtime": 195.607,
"eval_samples_per_second": 5.286,
"step": 3648
},
{
"epoch": 1217.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3652
},
{
"epoch": 1218.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3656
},
{
"epoch": 1219.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3660
},
{
"epoch": 1221.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3664
},
{
"epoch": 1222.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3668
},
{
"epoch": 1223.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3672
},
{
"epoch": 1225.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3676
},
{
"epoch": 1226.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3680
},
{
"epoch": 1227.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3684
},
{
"epoch": 1229.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3688
},
{
"epoch": 1230.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3692
},
{
"epoch": 1231.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3696
},
{
"epoch": 1233.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3700
},
{
"epoch": 1234.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3704
},
{
"epoch": 1235.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3708
},
{
"epoch": 1237.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3712
},
{
"epoch": 1237.29,
"eval_exact_match": 0.632495164410058,
"eval_loss": 0.44447311758995056,
"eval_runtime": 188.9293,
"eval_samples_per_second": 5.473,
"step": 3712
},
{
"epoch": 1238.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3716
},
{
"epoch": 1239.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3720
},
{
"epoch": 1241.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3724
},
{
"epoch": 1242.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3728
},
{
"epoch": 1243.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3732
},
{
"epoch": 1245.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3736
},
{
"epoch": 1246.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3740
},
{
"epoch": 1247.87,
"learning_rate": 0.0001,
"loss": 0.0014,
"step": 3744
},
{
"epoch": 1249.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3748
},
{
"epoch": 1250.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3752
},
{
"epoch": 1251.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3756
},
{
"epoch": 1253.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3760
},
{
"epoch": 1254.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3764
},
{
"epoch": 1255.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3768
},
{
"epoch": 1257.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3772
},
{
"epoch": 1258.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3776
},
{
"epoch": 1258.58,
"eval_exact_match": 0.6315280464216635,
"eval_loss": 0.4454585909843445,
"eval_runtime": 200.6809,
"eval_samples_per_second": 5.152,
"step": 3776
},
{
"epoch": 1259.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3780
},
{
"epoch": 1261.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3784
},
{
"epoch": 1262.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3788
},
{
"epoch": 1263.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3792
},
{
"epoch": 1265.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3796
},
{
"epoch": 1266.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3800
},
{
"epoch": 1267.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3804
},
{
"epoch": 1269.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3808
},
{
"epoch": 1270.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3812
},
{
"epoch": 1271.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3816
},
{
"epoch": 1273.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3820
},
{
"epoch": 1274.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3824
},
{
"epoch": 1275.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3828
},
{
"epoch": 1277.29,
"learning_rate": 0.0001,
"loss": 0.0009,
"step": 3832
},
{
"epoch": 1278.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3836
},
{
"epoch": 1279.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3840
},
{
"epoch": 1279.87,
"eval_exact_match": 0.6266924564796905,
"eval_loss": 0.4396822452545166,
"eval_runtime": 195.0331,
"eval_samples_per_second": 5.302,
"step": 3840
},
{
"epoch": 1281.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3844
},
{
"epoch": 1282.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3848
},
{
"epoch": 1283.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3852
},
{
"epoch": 1285.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3856
},
{
"epoch": 1286.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3860
},
{
"epoch": 1287.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3864
},
{
"epoch": 1289.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3868
},
{
"epoch": 1290.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3872
},
{
"epoch": 1291.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3876
},
{
"epoch": 1293.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3880
},
{
"epoch": 1294.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3884
},
{
"epoch": 1295.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3888
},
{
"epoch": 1297.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3892
},
{
"epoch": 1298.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3896
},
{
"epoch": 1299.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3900
},
{
"epoch": 1301.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3904
},
{
"epoch": 1301.29,
"eval_exact_match": 0.6353965183752418,
"eval_loss": 0.4471467435359955,
"eval_runtime": 188.9551,
"eval_samples_per_second": 5.472,
"step": 3904
},
{
"epoch": 1302.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3908
},
{
"epoch": 1303.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3912
},
{
"epoch": 1305.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3916
},
{
"epoch": 1306.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3920
},
{
"epoch": 1307.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3924
},
{
"epoch": 1309.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3928
},
{
"epoch": 1310.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 3932
},
{
"epoch": 1311.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3936
},
{
"epoch": 1313.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3940
},
{
"epoch": 1314.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3944
},
{
"epoch": 1315.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3948
},
{
"epoch": 1317.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3952
},
{
"epoch": 1318.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3956
},
{
"epoch": 1319.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3960
},
{
"epoch": 1321.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3964
},
{
"epoch": 1322.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3968
},
{
"epoch": 1322.58,
"eval_exact_match": 0.6344294003868471,
"eval_loss": 0.46103620529174805,
"eval_runtime": 188.0154,
"eval_samples_per_second": 5.5,
"step": 3968
},
{
"epoch": 1323.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3972
},
{
"epoch": 1325.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 3976
},
{
"epoch": 1326.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3980
},
{
"epoch": 1327.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 3984
},
{
"epoch": 1329.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 3988
},
{
"epoch": 1330.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 3992
},
{
"epoch": 1331.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 3996
},
{
"epoch": 1333.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4000
},
{
"epoch": 1334.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4004
},
{
"epoch": 1335.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4008
},
{
"epoch": 1337.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4012
},
{
"epoch": 1338.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4016
},
{
"epoch": 1339.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4020
},
{
"epoch": 1341.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 4024
},
{
"epoch": 1342.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4028
},
{
"epoch": 1343.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4032
},
{
"epoch": 1343.87,
"eval_exact_match": 0.6228239845261122,
"eval_loss": 0.44995567202568054,
"eval_runtime": 191.2283,
"eval_samples_per_second": 5.407,
"step": 4032
},
{
"epoch": 1345.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4036
},
{
"epoch": 1346.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4040
},
{
"epoch": 1347.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4044
},
{
"epoch": 1349.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4048
},
{
"epoch": 1350.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4052
},
{
"epoch": 1351.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4056
},
{
"epoch": 1353.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4060
},
{
"epoch": 1354.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4064
},
{
"epoch": 1355.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4068
},
{
"epoch": 1357.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 4072
},
{
"epoch": 1358.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4076
},
{
"epoch": 1359.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4080
},
{
"epoch": 1361.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4084
},
{
"epoch": 1362.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4088
},
{
"epoch": 1363.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4092
},
{
"epoch": 1365.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4096
},
{
"epoch": 1365.29,
"eval_exact_match": 0.6295938104448743,
"eval_loss": 0.44842708110809326,
"eval_runtime": 188.5729,
"eval_samples_per_second": 5.483,
"step": 4096
},
{
"epoch": 1366.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4100
},
{
"epoch": 1367.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4104
},
{
"epoch": 1369.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 4108
},
{
"epoch": 1370.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4112
},
{
"epoch": 1371.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4116
},
{
"epoch": 1373.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4120
},
{
"epoch": 1374.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 4124
},
{
"epoch": 1375.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4128
},
{
"epoch": 1377.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4132
},
{
"epoch": 1378.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4136
},
{
"epoch": 1379.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4140
},
{
"epoch": 1381.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4144
},
{
"epoch": 1382.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4148
},
{
"epoch": 1383.87,
"learning_rate": 0.0001,
"loss": 0.0019,
"step": 4152
},
{
"epoch": 1385.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4156
},
{
"epoch": 1386.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4160
},
{
"epoch": 1386.58,
"eval_exact_match": 0.6228239845261122,
"eval_loss": 0.44423192739486694,
"eval_runtime": 190.5483,
"eval_samples_per_second": 5.426,
"step": 4160
},
{
"epoch": 1387.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4164
},
{
"epoch": 1389.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4168
},
{
"epoch": 1390.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4172
},
{
"epoch": 1391.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4176
},
{
"epoch": 1393.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 4180
},
{
"epoch": 1394.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4184
},
{
"epoch": 1395.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4188
},
{
"epoch": 1397.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4192
},
{
"epoch": 1398.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4196
},
{
"epoch": 1399.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4200
},
{
"epoch": 1401.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4204
},
{
"epoch": 1402.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4208
},
{
"epoch": 1403.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4212
},
{
"epoch": 1405.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4216
},
{
"epoch": 1406.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4220
},
{
"epoch": 1407.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4224
},
{
"epoch": 1407.87,
"eval_exact_match": 0.6276595744680851,
"eval_loss": 0.43081170320510864,
"eval_runtime": 188.0185,
"eval_samples_per_second": 5.499,
"step": 4224
},
{
"epoch": 1409.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4228
},
{
"epoch": 1410.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4232
},
{
"epoch": 1411.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4236
},
{
"epoch": 1413.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4240
},
{
"epoch": 1414.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4244
},
{
"epoch": 1415.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4248
},
{
"epoch": 1417.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4252
},
{
"epoch": 1418.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4256
},
{
"epoch": 1419.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4260
},
{
"epoch": 1421.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4264
},
{
"epoch": 1422.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4268
},
{
"epoch": 1423.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 4272
},
{
"epoch": 1425.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4276
},
{
"epoch": 1426.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4280
},
{
"epoch": 1427.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4284
},
{
"epoch": 1429.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4288
},
{
"epoch": 1429.29,
"eval_exact_match": 0.6228239845261122,
"eval_loss": 0.46322837471961975,
"eval_runtime": 197.7854,
"eval_samples_per_second": 5.228,
"step": 4288
},
{
"epoch": 1430.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 4292
},
{
"epoch": 1431.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4296
},
{
"epoch": 1433.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4300
},
{
"epoch": 1434.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4304
},
{
"epoch": 1435.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4308
},
{
"epoch": 1437.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4312
},
{
"epoch": 1438.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 4316
},
{
"epoch": 1439.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4320
},
{
"epoch": 1441.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4324
},
{
"epoch": 1442.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4328
},
{
"epoch": 1443.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4332
},
{
"epoch": 1445.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4336
},
{
"epoch": 1446.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4340
},
{
"epoch": 1447.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4344
},
{
"epoch": 1449.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4348
},
{
"epoch": 1450.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 4352
},
{
"epoch": 1450.58,
"eval_exact_match": 0.6228239845261122,
"eval_loss": 0.4470163583755493,
"eval_runtime": 192.0339,
"eval_samples_per_second": 5.384,
"step": 4352
},
{
"epoch": 1451.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4356
},
{
"epoch": 1453.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4360
},
{
"epoch": 1454.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4364
},
{
"epoch": 1455.87,
"learning_rate": 0.0001,
"loss": 0.0024,
"step": 4368
},
{
"epoch": 1457.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4372
},
{
"epoch": 1458.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4376
},
{
"epoch": 1459.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4380
},
{
"epoch": 1461.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4384
},
{
"epoch": 1462.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4388
},
{
"epoch": 1463.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4392
},
{
"epoch": 1465.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4396
},
{
"epoch": 1466.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4400
},
{
"epoch": 1467.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4404
},
{
"epoch": 1469.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4408
},
{
"epoch": 1470.58,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 4412
},
{
"epoch": 1471.87,
"learning_rate": 0.0001,
"loss": 0.0032,
"step": 4416
},
{
"epoch": 1471.87,
"eval_exact_match": 0.6295938104448743,
"eval_loss": 0.42921268939971924,
"eval_runtime": 198.8466,
"eval_samples_per_second": 5.2,
"step": 4416
},
{
"epoch": 1473.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4420
},
{
"epoch": 1474.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4424
},
{
"epoch": 1475.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4428
},
{
"epoch": 1477.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4432
},
{
"epoch": 1478.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4436
},
{
"epoch": 1479.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4440
},
{
"epoch": 1481.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4444
},
{
"epoch": 1482.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4448
},
{
"epoch": 1483.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4452
},
{
"epoch": 1485.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4456
},
{
"epoch": 1486.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4460
},
{
"epoch": 1487.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4464
},
{
"epoch": 1489.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4468
},
{
"epoch": 1490.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4472
},
{
"epoch": 1491.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4476
},
{
"epoch": 1493.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 4480
},
{
"epoch": 1493.29,
"eval_exact_match": 0.6363636363636364,
"eval_loss": 0.44782036542892456,
"eval_runtime": 189.1196,
"eval_samples_per_second": 5.467,
"step": 4480
},
{
"epoch": 1494.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4484
},
{
"epoch": 1495.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4488
},
{
"epoch": 1497.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4492
},
{
"epoch": 1498.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4496
},
{
"epoch": 1499.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4500
},
{
"epoch": 1501.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4504
},
{
"epoch": 1502.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4508
},
{
"epoch": 1503.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4512
},
{
"epoch": 1505.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4516
},
{
"epoch": 1506.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4520
},
{
"epoch": 1507.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4524
},
{
"epoch": 1509.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4528
},
{
"epoch": 1510.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4532
},
{
"epoch": 1511.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4536
},
{
"epoch": 1513.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4540
},
{
"epoch": 1514.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4544
},
{
"epoch": 1514.58,
"eval_exact_match": 0.6411992263056093,
"eval_loss": 0.455540269613266,
"eval_runtime": 188.1434,
"eval_samples_per_second": 5.496,
"step": 4544
},
{
"epoch": 1515.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4548
},
{
"epoch": 1517.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4552
},
{
"epoch": 1518.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4556
},
{
"epoch": 1519.87,
"learning_rate": 0.0001,
"loss": 0.0008,
"step": 4560
},
{
"epoch": 1521.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4564
},
{
"epoch": 1522.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4568
},
{
"epoch": 1523.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4572
},
{
"epoch": 1525.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 4576
},
{
"epoch": 1526.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4580
},
{
"epoch": 1527.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4584
},
{
"epoch": 1529.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4588
},
{
"epoch": 1530.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4592
},
{
"epoch": 1531.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4596
},
{
"epoch": 1533.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4600
},
{
"epoch": 1534.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4604
},
{
"epoch": 1535.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4608
},
{
"epoch": 1535.87,
"eval_exact_match": 0.6499032882011605,
"eval_loss": 0.4536188542842865,
"eval_runtime": 195.0315,
"eval_samples_per_second": 5.302,
"step": 4608
},
{
"epoch": 1537.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4612
},
{
"epoch": 1538.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4616
},
{
"epoch": 1539.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4620
},
{
"epoch": 1541.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 4624
},
{
"epoch": 1542.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 4628
},
{
"epoch": 1543.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4632
},
{
"epoch": 1545.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4636
},
{
"epoch": 1546.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4640
},
{
"epoch": 1547.87,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 4644
},
{
"epoch": 1549.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4648
},
{
"epoch": 1550.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4652
},
{
"epoch": 1551.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4656
},
{
"epoch": 1553.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4660
},
{
"epoch": 1554.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4664
},
{
"epoch": 1555.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4668
},
{
"epoch": 1557.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4672
},
{
"epoch": 1557.29,
"eval_exact_match": 0.6363636363636364,
"eval_loss": 0.44341862201690674,
"eval_runtime": 203.5035,
"eval_samples_per_second": 5.081,
"step": 4672
},
{
"epoch": 1558.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4676
},
{
"epoch": 1559.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4680
},
{
"epoch": 1561.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4684
},
{
"epoch": 1562.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4688
},
{
"epoch": 1563.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4692
},
{
"epoch": 1565.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4696
},
{
"epoch": 1566.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4700
},
{
"epoch": 1567.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4704
},
{
"epoch": 1569.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4708
},
{
"epoch": 1570.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4712
},
{
"epoch": 1571.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4716
},
{
"epoch": 1573.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4720
},
{
"epoch": 1574.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4724
},
{
"epoch": 1575.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4728
},
{
"epoch": 1577.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 4732
},
{
"epoch": 1578.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4736
},
{
"epoch": 1578.58,
"eval_exact_match": 0.6334622823984526,
"eval_loss": 0.4519117474555969,
"eval_runtime": 202.255,
"eval_samples_per_second": 5.112,
"step": 4736
},
{
"epoch": 1579.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4740
},
{
"epoch": 1581.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4744
},
{
"epoch": 1582.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4748
},
{
"epoch": 1583.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4752
},
{
"epoch": 1585.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4756
},
{
"epoch": 1586.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4760
},
{
"epoch": 1587.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4764
},
{
"epoch": 1589.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4768
},
{
"epoch": 1590.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4772
},
{
"epoch": 1591.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4776
},
{
"epoch": 1593.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4780
},
{
"epoch": 1594.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4784
},
{
"epoch": 1595.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4788
},
{
"epoch": 1597.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4792
},
{
"epoch": 1598.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4796
},
{
"epoch": 1599.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4800
},
{
"epoch": 1599.87,
"eval_exact_match": 0.6353965183752418,
"eval_loss": 0.4388425946235657,
"eval_runtime": 189.1795,
"eval_samples_per_second": 5.466,
"step": 4800
},
{
"epoch": 1601.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4804
},
{
"epoch": 1602.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 4808
},
{
"epoch": 1603.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4812
},
{
"epoch": 1605.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4816
},
{
"epoch": 1606.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4820
},
{
"epoch": 1607.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4824
},
{
"epoch": 1609.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4828
},
{
"epoch": 1610.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4832
},
{
"epoch": 1611.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4836
},
{
"epoch": 1613.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4840
},
{
"epoch": 1614.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4844
},
{
"epoch": 1615.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4848
},
{
"epoch": 1617.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4852
},
{
"epoch": 1618.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4856
},
{
"epoch": 1619.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4860
},
{
"epoch": 1621.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4864
},
{
"epoch": 1621.29,
"eval_exact_match": 0.6402321083172147,
"eval_loss": 0.4638407528400421,
"eval_runtime": 188.0386,
"eval_samples_per_second": 5.499,
"step": 4864
},
{
"epoch": 1622.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4868
},
{
"epoch": 1623.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4872
},
{
"epoch": 1625.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4876
},
{
"epoch": 1626.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4880
},
{
"epoch": 1627.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4884
},
{
"epoch": 1629.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4888
},
{
"epoch": 1630.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4892
},
{
"epoch": 1631.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4896
},
{
"epoch": 1633.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4900
},
{
"epoch": 1634.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4904
},
{
"epoch": 1635.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 4908
},
{
"epoch": 1637.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4912
},
{
"epoch": 1638.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4916
},
{
"epoch": 1639.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4920
},
{
"epoch": 1641.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4924
},
{
"epoch": 1642.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4928
},
{
"epoch": 1642.58,
"eval_exact_match": 0.6431334622823984,
"eval_loss": 0.4654744267463684,
"eval_runtime": 194.7711,
"eval_samples_per_second": 5.309,
"step": 4928
},
{
"epoch": 1643.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4932
},
{
"epoch": 1645.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4936
},
{
"epoch": 1646.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4940
},
{
"epoch": 1647.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4944
},
{
"epoch": 1649.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4948
},
{
"epoch": 1650.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 4952
},
{
"epoch": 1651.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4956
},
{
"epoch": 1653.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4960
},
{
"epoch": 1654.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4964
},
{
"epoch": 1655.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4968
},
{
"epoch": 1657.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4972
},
{
"epoch": 1658.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4976
},
{
"epoch": 1659.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4980
},
{
"epoch": 1661.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 4984
},
{
"epoch": 1662.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4988
},
{
"epoch": 1663.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4992
},
{
"epoch": 1663.87,
"eval_exact_match": 0.6353965183752418,
"eval_loss": 0.4679672122001648,
"eval_runtime": 189.9449,
"eval_samples_per_second": 5.444,
"step": 4992
},
{
"epoch": 1665.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 4996
},
{
"epoch": 1666.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5000
},
{
"epoch": 1667.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5004
},
{
"epoch": 1669.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5008
},
{
"epoch": 1670.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5012
},
{
"epoch": 1671.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5016
},
{
"epoch": 1673.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5020
},
{
"epoch": 1674.58,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 5024
},
{
"epoch": 1675.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 5028
},
{
"epoch": 1677.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5032
},
{
"epoch": 1678.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5036
},
{
"epoch": 1679.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5040
},
{
"epoch": 1681.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5044
},
{
"epoch": 1682.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5048
},
{
"epoch": 1683.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5052
},
{
"epoch": 1685.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5056
},
{
"epoch": 1685.29,
"eval_exact_match": 0.6334622823984526,
"eval_loss": 0.43442824482917786,
"eval_runtime": 208.3421,
"eval_samples_per_second": 4.963,
"step": 5056
},
{
"epoch": 1686.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5060
},
{
"epoch": 1687.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 5064
},
{
"epoch": 1689.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5068
},
{
"epoch": 1690.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5072
},
{
"epoch": 1691.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5076
},
{
"epoch": 1693.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5080
},
{
"epoch": 1694.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5084
},
{
"epoch": 1695.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5088
},
{
"epoch": 1697.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5092
},
{
"epoch": 1698.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5096
},
{
"epoch": 1699.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5100
},
{
"epoch": 1701.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5104
},
{
"epoch": 1702.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5108
},
{
"epoch": 1703.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5112
},
{
"epoch": 1705.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5116
},
{
"epoch": 1706.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 5120
},
{
"epoch": 1706.58,
"eval_exact_match": 0.6411992263056093,
"eval_loss": 0.4362451732158661,
"eval_runtime": 187.5005,
"eval_samples_per_second": 5.515,
"step": 5120
},
{
"epoch": 1707.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5124
},
{
"epoch": 1709.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5128
},
{
"epoch": 1710.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5132
},
{
"epoch": 1711.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5136
},
{
"epoch": 1713.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5140
},
{
"epoch": 1714.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5144
},
{
"epoch": 1715.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 5148
},
{
"epoch": 1717.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5152
},
{
"epoch": 1718.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5156
},
{
"epoch": 1719.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5160
},
{
"epoch": 1721.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5164
},
{
"epoch": 1722.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5168
},
{
"epoch": 1723.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5172
},
{
"epoch": 1725.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5176
},
{
"epoch": 1726.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5180
},
{
"epoch": 1727.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5184
},
{
"epoch": 1727.87,
"eval_exact_match": 0.632495164410058,
"eval_loss": 0.46694526076316833,
"eval_runtime": 201.6818,
"eval_samples_per_second": 5.127,
"step": 5184
},
{
"epoch": 1729.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5188
},
{
"epoch": 1730.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5192
},
{
"epoch": 1731.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5196
},
{
"epoch": 1733.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5200
},
{
"epoch": 1734.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5204
},
{
"epoch": 1735.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5208
},
{
"epoch": 1737.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5212
},
{
"epoch": 1738.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5216
},
{
"epoch": 1739.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5220
},
{
"epoch": 1741.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5224
},
{
"epoch": 1742.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5228
},
{
"epoch": 1743.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5232
},
{
"epoch": 1745.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5236
},
{
"epoch": 1746.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5240
},
{
"epoch": 1747.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5244
},
{
"epoch": 1749.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5248
},
{
"epoch": 1749.29,
"eval_exact_match": 0.6334622823984526,
"eval_loss": 0.4616599678993225,
"eval_runtime": 203.8893,
"eval_samples_per_second": 5.071,
"step": 5248
},
{
"epoch": 1750.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5252
},
{
"epoch": 1751.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5256
},
{
"epoch": 1753.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 5260
},
{
"epoch": 1754.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5264
},
{
"epoch": 1755.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5268
},
{
"epoch": 1757.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5272
},
{
"epoch": 1758.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5276
},
{
"epoch": 1759.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5280
},
{
"epoch": 1761.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5284
},
{
"epoch": 1762.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5288
},
{
"epoch": 1763.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5292
},
{
"epoch": 1765.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5296
},
{
"epoch": 1766.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5300
},
{
"epoch": 1767.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5304
},
{
"epoch": 1769.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 5308
},
{
"epoch": 1770.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5312
},
{
"epoch": 1770.58,
"eval_exact_match": 0.6266924564796905,
"eval_loss": 0.46690136194229126,
"eval_runtime": 200.8681,
"eval_samples_per_second": 5.148,
"step": 5312
},
{
"epoch": 1771.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5316
},
{
"epoch": 1773.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5320
},
{
"epoch": 1774.58,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 5324
},
{
"epoch": 1775.87,
"learning_rate": 0.0001,
"loss": 0.0013,
"step": 5328
},
{
"epoch": 1777.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5332
},
{
"epoch": 1778.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5336
},
{
"epoch": 1779.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5340
},
{
"epoch": 1781.29,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 5344
},
{
"epoch": 1782.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5348
},
{
"epoch": 1783.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5352
},
{
"epoch": 1785.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5356
},
{
"epoch": 1786.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5360
},
{
"epoch": 1787.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 5364
},
{
"epoch": 1789.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5368
},
{
"epoch": 1790.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5372
},
{
"epoch": 1791.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5376
},
{
"epoch": 1791.87,
"eval_exact_match": 0.6334622823984526,
"eval_loss": 0.4663194417953491,
"eval_runtime": 196.1072,
"eval_samples_per_second": 5.273,
"step": 5376
},
{
"epoch": 1793.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5380
},
{
"epoch": 1794.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5384
},
{
"epoch": 1795.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5388
},
{
"epoch": 1797.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5392
},
{
"epoch": 1798.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5396
},
{
"epoch": 1799.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5400
},
{
"epoch": 1801.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5404
},
{
"epoch": 1802.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5408
},
{
"epoch": 1803.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5412
},
{
"epoch": 1805.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5416
},
{
"epoch": 1806.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5420
},
{
"epoch": 1807.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5424
},
{
"epoch": 1809.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5428
},
{
"epoch": 1810.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5432
},
{
"epoch": 1811.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5436
},
{
"epoch": 1813.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5440
},
{
"epoch": 1813.29,
"eval_exact_match": 0.6286266924564797,
"eval_loss": 0.4637957811355591,
"eval_runtime": 192.7867,
"eval_samples_per_second": 5.363,
"step": 5440
},
{
"epoch": 1814.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5444
},
{
"epoch": 1815.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5448
},
{
"epoch": 1817.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5452
},
{
"epoch": 1818.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5456
},
{
"epoch": 1819.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5460
},
{
"epoch": 1821.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5464
},
{
"epoch": 1822.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5468
},
{
"epoch": 1823.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 5472
},
{
"epoch": 1825.29,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 5476
},
{
"epoch": 1826.58,
"learning_rate": 0.0001,
"loss": 0.002,
"step": 5480
},
{
"epoch": 1827.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5484
},
{
"epoch": 1829.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5488
},
{
"epoch": 1830.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5492
},
{
"epoch": 1831.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5496
},
{
"epoch": 1833.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5500
},
{
"epoch": 1834.58,
"learning_rate": 0.0001,
"loss": 0.0011,
"step": 5504
},
{
"epoch": 1834.58,
"eval_exact_match": 0.6344294003868471,
"eval_loss": 0.4392802119255066,
"eval_runtime": 187.4432,
"eval_samples_per_second": 5.516,
"step": 5504
},
{
"epoch": 1835.87,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 5508
},
{
"epoch": 1837.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5512
},
{
"epoch": 1838.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5516
},
{
"epoch": 1839.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5520
},
{
"epoch": 1841.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5524
},
{
"epoch": 1842.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5528
},
{
"epoch": 1843.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5532
},
{
"epoch": 1845.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5536
},
{
"epoch": 1846.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5540
},
{
"epoch": 1847.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5544
},
{
"epoch": 1849.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5548
},
{
"epoch": 1850.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5552
},
{
"epoch": 1851.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5556
},
{
"epoch": 1853.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5560
},
{
"epoch": 1854.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5564
},
{
"epoch": 1855.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5568
},
{
"epoch": 1855.87,
"eval_exact_match": 0.632495164410058,
"eval_loss": 0.4490313231945038,
"eval_runtime": 204.1165,
"eval_samples_per_second": 5.066,
"step": 5568
},
{
"epoch": 1857.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5572
},
{
"epoch": 1858.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5576
},
{
"epoch": 1859.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5580
},
{
"epoch": 1861.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5584
},
{
"epoch": 1862.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5588
},
{
"epoch": 1863.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5592
},
{
"epoch": 1865.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5596
},
{
"epoch": 1866.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5600
},
{
"epoch": 1867.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5604
},
{
"epoch": 1869.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5608
},
{
"epoch": 1870.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5612
},
{
"epoch": 1871.87,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 5616
},
{
"epoch": 1873.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5620
},
{
"epoch": 1874.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5624
},
{
"epoch": 1875.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5628
},
{
"epoch": 1877.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5632
},
{
"epoch": 1877.29,
"eval_exact_match": 0.6441005802707931,
"eval_loss": 0.45752978324890137,
"eval_runtime": 198.5165,
"eval_samples_per_second": 5.209,
"step": 5632
},
{
"epoch": 1878.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5636
},
{
"epoch": 1879.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5640
},
{
"epoch": 1881.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5644
},
{
"epoch": 1882.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5648
},
{
"epoch": 1883.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5652
},
{
"epoch": 1885.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5656
},
{
"epoch": 1886.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5660
},
{
"epoch": 1887.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5664
},
{
"epoch": 1889.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5668
},
{
"epoch": 1890.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5672
},
{
"epoch": 1891.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5676
},
{
"epoch": 1893.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5680
},
{
"epoch": 1894.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5684
},
{
"epoch": 1895.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5688
},
{
"epoch": 1897.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5692
},
{
"epoch": 1898.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5696
},
{
"epoch": 1898.58,
"eval_exact_match": 0.6431334622823984,
"eval_loss": 0.479341596364975,
"eval_runtime": 190.8589,
"eval_samples_per_second": 5.418,
"step": 5696
},
{
"epoch": 1899.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5700
},
{
"epoch": 1901.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5704
},
{
"epoch": 1902.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5708
},
{
"epoch": 1903.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5712
},
{
"epoch": 1905.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5716
},
{
"epoch": 1906.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5720
},
{
"epoch": 1907.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5724
},
{
"epoch": 1909.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5728
},
{
"epoch": 1910.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5732
},
{
"epoch": 1911.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5736
},
{
"epoch": 1913.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5740
},
{
"epoch": 1914.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5744
},
{
"epoch": 1915.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5748
},
{
"epoch": 1917.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5752
},
{
"epoch": 1918.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5756
},
{
"epoch": 1919.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5760
},
{
"epoch": 1919.87,
"eval_exact_match": 0.6537717601547389,
"eval_loss": 0.4889405071735382,
"eval_runtime": 187.8265,
"eval_samples_per_second": 5.505,
"step": 5760
},
{
"epoch": 1921.29,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5764
},
{
"epoch": 1922.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5768
},
{
"epoch": 1923.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5772
},
{
"epoch": 1925.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5776
},
{
"epoch": 1926.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5780
},
{
"epoch": 1927.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5784
},
{
"epoch": 1929.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5788
},
{
"epoch": 1930.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5792
},
{
"epoch": 1931.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5796
},
{
"epoch": 1933.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5800
},
{
"epoch": 1934.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5804
},
{
"epoch": 1935.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5808
},
{
"epoch": 1937.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5812
},
{
"epoch": 1938.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5816
},
{
"epoch": 1939.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5820
},
{
"epoch": 1941.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5824
},
{
"epoch": 1941.29,
"eval_exact_match": 0.6421663442940039,
"eval_loss": 0.47956621646881104,
"eval_runtime": 187.9453,
"eval_samples_per_second": 5.502,
"step": 5824
},
{
"epoch": 1942.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5828
},
{
"epoch": 1943.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5832
},
{
"epoch": 1945.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5836
},
{
"epoch": 1946.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5840
},
{
"epoch": 1947.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5844
},
{
"epoch": 1949.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5848
},
{
"epoch": 1950.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5852
},
{
"epoch": 1951.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5856
},
{
"epoch": 1953.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5860
},
{
"epoch": 1954.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5864
},
{
"epoch": 1955.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5868
},
{
"epoch": 1957.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5872
},
{
"epoch": 1958.58,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5876
},
{
"epoch": 1959.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5880
},
{
"epoch": 1961.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5884
},
{
"epoch": 1962.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5888
},
{
"epoch": 1962.58,
"eval_exact_match": 0.6431334622823984,
"eval_loss": 0.49590054154396057,
"eval_runtime": 191.5599,
"eval_samples_per_second": 5.398,
"step": 5888
},
{
"epoch": 1963.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5892
},
{
"epoch": 1965.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5896
},
{
"epoch": 1966.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5900
},
{
"epoch": 1967.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5904
},
{
"epoch": 1969.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5908
},
{
"epoch": 1970.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5912
},
{
"epoch": 1971.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 5916
},
{
"epoch": 1973.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5920
},
{
"epoch": 1974.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5924
},
{
"epoch": 1975.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5928
},
{
"epoch": 1977.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5932
},
{
"epoch": 1978.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5936
},
{
"epoch": 1979.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5940
},
{
"epoch": 1981.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5944
},
{
"epoch": 1982.58,
"learning_rate": 0.0001,
"loss": 0.0005,
"step": 5948
},
{
"epoch": 1983.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5952
},
{
"epoch": 1983.87,
"eval_exact_match": 0.6450676982591876,
"eval_loss": 0.46603018045425415,
"eval_runtime": 192.1373,
"eval_samples_per_second": 5.382,
"step": 5952
},
{
"epoch": 1985.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5956
},
{
"epoch": 1986.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5960
},
{
"epoch": 1987.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5964
},
{
"epoch": 1989.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5968
},
{
"epoch": 1990.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5972
},
{
"epoch": 1991.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5976
},
{
"epoch": 1993.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5980
},
{
"epoch": 1994.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5984
},
{
"epoch": 1995.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 5988
},
{
"epoch": 1997.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 5992
},
{
"epoch": 1998.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 5996
},
{
"epoch": 1999.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6000
},
{
"epoch": 2001.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6004
},
{
"epoch": 2002.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6008
},
{
"epoch": 2003.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6012
},
{
"epoch": 2005.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6016
},
{
"epoch": 2005.29,
"eval_exact_match": 0.6402321083172147,
"eval_loss": 0.45267048478126526,
"eval_runtime": 193.3117,
"eval_samples_per_second": 5.349,
"step": 6016
},
{
"epoch": 2006.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6020
},
{
"epoch": 2007.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6024
},
{
"epoch": 2009.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6028
},
{
"epoch": 2010.58,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 6032
},
{
"epoch": 2011.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6036
},
{
"epoch": 2013.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6040
},
{
"epoch": 2014.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6044
},
{
"epoch": 2015.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6048
},
{
"epoch": 2017.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6052
},
{
"epoch": 2018.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6056
},
{
"epoch": 2019.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6060
},
{
"epoch": 2021.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 6064
},
{
"epoch": 2022.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6068
},
{
"epoch": 2023.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6072
},
{
"epoch": 2025.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6076
},
{
"epoch": 2026.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6080
},
{
"epoch": 2026.58,
"eval_exact_match": 0.6431334622823984,
"eval_loss": 0.4771885573863983,
"eval_runtime": 182.168,
"eval_samples_per_second": 5.676,
"step": 6080
},
{
"epoch": 2027.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 6084
},
{
"epoch": 2029.29,
"learning_rate": 0.0001,
"loss": 0.0004,
"step": 6088
},
{
"epoch": 2030.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6092
},
{
"epoch": 2031.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6096
},
{
"epoch": 2033.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6100
},
{
"epoch": 2034.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6104
},
{
"epoch": 2035.87,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 6108
},
{
"epoch": 2037.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6112
},
{
"epoch": 2038.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6116
},
{
"epoch": 2039.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6120
},
{
"epoch": 2041.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6124
},
{
"epoch": 2042.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6128
},
{
"epoch": 2043.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6132
},
{
"epoch": 2045.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6136
},
{
"epoch": 2046.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6140
},
{
"epoch": 2047.87,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6144
},
{
"epoch": 2047.87,
"eval_exact_match": 0.6479690522243714,
"eval_loss": 0.49728310108184814,
"eval_runtime": 184.1321,
"eval_samples_per_second": 5.616,
"step": 6144
},
{
"epoch": 2049.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 6148
},
{
"epoch": 2050.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6152
},
{
"epoch": 2051.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6156
},
{
"epoch": 2053.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6160
},
{
"epoch": 2054.58,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6164
},
{
"epoch": 2055.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6168
},
{
"epoch": 2057.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6172
},
{
"epoch": 2058.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6176
},
{
"epoch": 2059.87,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6180
},
{
"epoch": 2061.29,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6184
},
{
"epoch": 2062.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6188
},
{
"epoch": 2063.87,
"learning_rate": 0.0001,
"loss": 0.0006,
"step": 6192
},
{
"epoch": 2065.29,
"learning_rate": 0.0001,
"loss": 0.0002,
"step": 6196
},
{
"epoch": 2066.58,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 6200
},
{
"epoch": 2067.87,
"learning_rate": 0.0001,
"loss": 0.0003,
"step": 6204
},
{
"epoch": 2069.29,
"learning_rate": 0.0001,
"loss": 0.0001,
"step": 6208
},
{
"epoch": 2069.29,
"eval_exact_match": 0.6547388781431335,
"eval_loss": 0.47000861167907715,
"eval_runtime": 199.9619,
"eval_samples_per_second": 5.171,
"step": 6208
}
],
"max_steps": 9216,
"num_train_epochs": 3072,
"total_flos": 2.850364268144178e+19,
"trial_name": null,
"trial_params": null
}