t5_3b_qgen / trainer_state.json
learn3r's picture
End of training
357df63
{
"best_metric": 1.0960041284561157,
"best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/question_gen/t5_3b_epoch_10/checkpoint-342",
"epoch": 9.994520547945205,
"global_step": 3420,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 9.97076023391813e-05,
"loss": 1.7391,
"step": 10
},
{
"epoch": 0.06,
"learning_rate": 9.941520467836257e-05,
"loss": 1.3208,
"step": 20
},
{
"epoch": 0.09,
"learning_rate": 9.912280701754386e-05,
"loss": 1.2679,
"step": 30
},
{
"epoch": 0.12,
"learning_rate": 9.883040935672515e-05,
"loss": 1.2292,
"step": 40
},
{
"epoch": 0.15,
"learning_rate": 9.853801169590643e-05,
"loss": 1.2356,
"step": 50
},
{
"epoch": 0.18,
"learning_rate": 9.824561403508771e-05,
"loss": 1.2327,
"step": 60
},
{
"epoch": 0.2,
"learning_rate": 9.795321637426901e-05,
"loss": 1.2397,
"step": 70
},
{
"epoch": 0.23,
"learning_rate": 9.76608187134503e-05,
"loss": 1.1967,
"step": 80
},
{
"epoch": 0.26,
"learning_rate": 9.736842105263158e-05,
"loss": 1.1946,
"step": 90
},
{
"epoch": 0.29,
"learning_rate": 9.707602339181286e-05,
"loss": 1.2225,
"step": 100
},
{
"epoch": 0.32,
"learning_rate": 9.678362573099416e-05,
"loss": 1.2038,
"step": 110
},
{
"epoch": 0.35,
"learning_rate": 9.649122807017544e-05,
"loss": 1.1921,
"step": 120
},
{
"epoch": 0.38,
"learning_rate": 9.619883040935673e-05,
"loss": 1.1498,
"step": 130
},
{
"epoch": 0.41,
"learning_rate": 9.590643274853801e-05,
"loss": 1.1471,
"step": 140
},
{
"epoch": 0.44,
"learning_rate": 9.56140350877193e-05,
"loss": 1.1799,
"step": 150
},
{
"epoch": 0.47,
"learning_rate": 9.532163742690059e-05,
"loss": 1.1812,
"step": 160
},
{
"epoch": 0.5,
"learning_rate": 9.502923976608188e-05,
"loss": 1.1693,
"step": 170
},
{
"epoch": 0.53,
"learning_rate": 9.473684210526316e-05,
"loss": 1.1852,
"step": 180
},
{
"epoch": 0.56,
"learning_rate": 9.444444444444444e-05,
"loss": 1.1751,
"step": 190
},
{
"epoch": 0.58,
"learning_rate": 9.415204678362574e-05,
"loss": 1.1588,
"step": 200
},
{
"epoch": 0.61,
"learning_rate": 9.385964912280703e-05,
"loss": 1.1564,
"step": 210
},
{
"epoch": 0.64,
"learning_rate": 9.35672514619883e-05,
"loss": 1.181,
"step": 220
},
{
"epoch": 0.67,
"learning_rate": 9.327485380116959e-05,
"loss": 1.1835,
"step": 230
},
{
"epoch": 0.7,
"learning_rate": 9.298245614035089e-05,
"loss": 1.1421,
"step": 240
},
{
"epoch": 0.73,
"learning_rate": 9.269005847953217e-05,
"loss": 1.1774,
"step": 250
},
{
"epoch": 0.76,
"learning_rate": 9.239766081871345e-05,
"loss": 1.1355,
"step": 260
},
{
"epoch": 0.79,
"learning_rate": 9.210526315789474e-05,
"loss": 1.1467,
"step": 270
},
{
"epoch": 0.82,
"learning_rate": 9.181286549707603e-05,
"loss": 1.1534,
"step": 280
},
{
"epoch": 0.85,
"learning_rate": 9.152046783625731e-05,
"loss": 1.1597,
"step": 290
},
{
"epoch": 0.88,
"learning_rate": 9.12280701754386e-05,
"loss": 1.1345,
"step": 300
},
{
"epoch": 0.91,
"learning_rate": 9.093567251461989e-05,
"loss": 1.1458,
"step": 310
},
{
"epoch": 0.94,
"learning_rate": 9.064327485380117e-05,
"loss": 1.1065,
"step": 320
},
{
"epoch": 0.96,
"learning_rate": 9.035087719298246e-05,
"loss": 1.1485,
"step": 330
},
{
"epoch": 0.99,
"learning_rate": 9.005847953216374e-05,
"loss": 1.147,
"step": 340
},
{
"epoch": 1.0,
"eval_loss": 1.0960041284561157,
"eval_runtime": 132.8151,
"eval_samples_per_second": 79.584,
"eval_steps_per_second": 9.954,
"step": 342
},
{
"epoch": 1.02,
"learning_rate": 8.976608187134502e-05,
"loss": 1.0368,
"step": 350
},
{
"epoch": 1.05,
"learning_rate": 8.947368421052632e-05,
"loss": 0.9974,
"step": 360
},
{
"epoch": 1.08,
"learning_rate": 8.918128654970761e-05,
"loss": 0.9777,
"step": 370
},
{
"epoch": 1.11,
"learning_rate": 8.888888888888889e-05,
"loss": 0.9939,
"step": 380
},
{
"epoch": 1.14,
"learning_rate": 8.859649122807017e-05,
"loss": 0.9984,
"step": 390
},
{
"epoch": 1.17,
"learning_rate": 8.830409356725147e-05,
"loss": 0.9941,
"step": 400
},
{
"epoch": 1.2,
"learning_rate": 8.801169590643276e-05,
"loss": 1.0135,
"step": 410
},
{
"epoch": 1.23,
"learning_rate": 8.771929824561403e-05,
"loss": 0.9955,
"step": 420
},
{
"epoch": 1.26,
"learning_rate": 8.742690058479532e-05,
"loss": 0.984,
"step": 430
},
{
"epoch": 1.29,
"learning_rate": 8.713450292397662e-05,
"loss": 1.0194,
"step": 440
},
{
"epoch": 1.32,
"learning_rate": 8.68421052631579e-05,
"loss": 0.9848,
"step": 450
},
{
"epoch": 1.34,
"learning_rate": 8.654970760233918e-05,
"loss": 0.9885,
"step": 460
},
{
"epoch": 1.37,
"learning_rate": 8.625730994152047e-05,
"loss": 0.995,
"step": 470
},
{
"epoch": 1.4,
"learning_rate": 8.596491228070177e-05,
"loss": 1.0075,
"step": 480
},
{
"epoch": 1.43,
"learning_rate": 8.567251461988305e-05,
"loss": 1.0036,
"step": 490
},
{
"epoch": 1.46,
"learning_rate": 8.538011695906433e-05,
"loss": 0.9981,
"step": 500
},
{
"epoch": 1.49,
"learning_rate": 8.508771929824562e-05,
"loss": 0.9783,
"step": 510
},
{
"epoch": 1.52,
"learning_rate": 8.47953216374269e-05,
"loss": 0.9974,
"step": 520
},
{
"epoch": 1.55,
"learning_rate": 8.45029239766082e-05,
"loss": 0.9985,
"step": 530
},
{
"epoch": 1.58,
"learning_rate": 8.421052631578948e-05,
"loss": 1.0136,
"step": 540
},
{
"epoch": 1.61,
"learning_rate": 8.391812865497076e-05,
"loss": 1.0077,
"step": 550
},
{
"epoch": 1.64,
"learning_rate": 8.362573099415205e-05,
"loss": 0.9968,
"step": 560
},
{
"epoch": 1.67,
"learning_rate": 8.333333333333334e-05,
"loss": 1.0064,
"step": 570
},
{
"epoch": 1.69,
"learning_rate": 8.304093567251462e-05,
"loss": 0.994,
"step": 580
},
{
"epoch": 1.72,
"learning_rate": 8.27485380116959e-05,
"loss": 0.9906,
"step": 590
},
{
"epoch": 1.75,
"learning_rate": 8.24561403508772e-05,
"loss": 0.9842,
"step": 600
},
{
"epoch": 1.78,
"learning_rate": 8.216374269005849e-05,
"loss": 0.9906,
"step": 610
},
{
"epoch": 1.81,
"learning_rate": 8.187134502923976e-05,
"loss": 0.9925,
"step": 620
},
{
"epoch": 1.84,
"learning_rate": 8.157894736842105e-05,
"loss": 0.9849,
"step": 630
},
{
"epoch": 1.87,
"learning_rate": 8.128654970760235e-05,
"loss": 0.9992,
"step": 640
},
{
"epoch": 1.9,
"learning_rate": 8.099415204678363e-05,
"loss": 0.9861,
"step": 650
},
{
"epoch": 1.93,
"learning_rate": 8.070175438596491e-05,
"loss": 0.9953,
"step": 660
},
{
"epoch": 1.96,
"learning_rate": 8.04093567251462e-05,
"loss": 0.9758,
"step": 670
},
{
"epoch": 1.99,
"learning_rate": 8.01169590643275e-05,
"loss": 0.9928,
"step": 680
},
{
"epoch": 2.0,
"eval_loss": 1.102066159248352,
"eval_runtime": 133.0084,
"eval_samples_per_second": 79.469,
"eval_steps_per_second": 9.939,
"step": 684
},
{
"epoch": 2.02,
"learning_rate": 7.982456140350878e-05,
"loss": 0.9297,
"step": 690
},
{
"epoch": 2.05,
"learning_rate": 7.953216374269006e-05,
"loss": 0.8537,
"step": 700
},
{
"epoch": 2.07,
"learning_rate": 7.923976608187135e-05,
"loss": 0.8556,
"step": 710
},
{
"epoch": 2.1,
"learning_rate": 7.894736842105263e-05,
"loss": 0.8347,
"step": 720
},
{
"epoch": 2.13,
"learning_rate": 7.865497076023393e-05,
"loss": 0.8416,
"step": 730
},
{
"epoch": 2.16,
"learning_rate": 7.836257309941521e-05,
"loss": 0.8436,
"step": 740
},
{
"epoch": 2.19,
"learning_rate": 7.807017543859649e-05,
"loss": 0.8352,
"step": 750
},
{
"epoch": 2.22,
"learning_rate": 7.777777777777778e-05,
"loss": 0.8387,
"step": 760
},
{
"epoch": 2.25,
"learning_rate": 7.748538011695908e-05,
"loss": 0.8336,
"step": 770
},
{
"epoch": 2.28,
"learning_rate": 7.719298245614036e-05,
"loss": 0.8653,
"step": 780
},
{
"epoch": 2.31,
"learning_rate": 7.690058479532164e-05,
"loss": 0.8509,
"step": 790
},
{
"epoch": 2.34,
"learning_rate": 7.660818713450293e-05,
"loss": 0.8525,
"step": 800
},
{
"epoch": 2.37,
"learning_rate": 7.631578947368422e-05,
"loss": 0.8481,
"step": 810
},
{
"epoch": 2.4,
"learning_rate": 7.602339181286549e-05,
"loss": 0.8615,
"step": 820
},
{
"epoch": 2.43,
"learning_rate": 7.573099415204679e-05,
"loss": 0.8697,
"step": 830
},
{
"epoch": 2.45,
"learning_rate": 7.543859649122808e-05,
"loss": 0.8498,
"step": 840
},
{
"epoch": 2.48,
"learning_rate": 7.514619883040936e-05,
"loss": 0.8691,
"step": 850
},
{
"epoch": 2.51,
"learning_rate": 7.485380116959064e-05,
"loss": 0.8661,
"step": 860
},
{
"epoch": 2.54,
"learning_rate": 7.456140350877193e-05,
"loss": 0.8747,
"step": 870
},
{
"epoch": 2.57,
"learning_rate": 7.426900584795321e-05,
"loss": 0.8701,
"step": 880
},
{
"epoch": 2.6,
"learning_rate": 7.397660818713451e-05,
"loss": 0.8556,
"step": 890
},
{
"epoch": 2.63,
"learning_rate": 7.368421052631579e-05,
"loss": 0.8589,
"step": 900
},
{
"epoch": 2.66,
"learning_rate": 7.339181286549708e-05,
"loss": 0.8526,
"step": 910
},
{
"epoch": 2.69,
"learning_rate": 7.309941520467836e-05,
"loss": 0.8838,
"step": 920
},
{
"epoch": 2.72,
"learning_rate": 7.280701754385966e-05,
"loss": 0.872,
"step": 930
},
{
"epoch": 2.75,
"learning_rate": 7.251461988304094e-05,
"loss": 0.858,
"step": 940
},
{
"epoch": 2.78,
"learning_rate": 7.222222222222222e-05,
"loss": 0.852,
"step": 950
},
{
"epoch": 2.81,
"learning_rate": 7.192982456140351e-05,
"loss": 0.8461,
"step": 960
},
{
"epoch": 2.83,
"learning_rate": 7.163742690058481e-05,
"loss": 0.8773,
"step": 970
},
{
"epoch": 2.86,
"learning_rate": 7.134502923976609e-05,
"loss": 0.8413,
"step": 980
},
{
"epoch": 2.89,
"learning_rate": 7.105263157894737e-05,
"loss": 0.8462,
"step": 990
},
{
"epoch": 2.92,
"learning_rate": 7.076023391812866e-05,
"loss": 0.8504,
"step": 1000
},
{
"epoch": 2.95,
"learning_rate": 7.046783625730996e-05,
"loss": 0.8554,
"step": 1010
},
{
"epoch": 2.98,
"learning_rate": 7.017543859649122e-05,
"loss": 0.861,
"step": 1020
},
{
"epoch": 3.0,
"eval_loss": 1.1422148942947388,
"eval_runtime": 133.1909,
"eval_samples_per_second": 79.36,
"eval_steps_per_second": 9.926,
"step": 1026
},
{
"epoch": 3.01,
"learning_rate": 6.988304093567252e-05,
"loss": 0.8067,
"step": 1030
},
{
"epoch": 3.04,
"learning_rate": 6.959064327485381e-05,
"loss": 0.7203,
"step": 1040
},
{
"epoch": 3.07,
"learning_rate": 6.929824561403509e-05,
"loss": 0.7223,
"step": 1050
},
{
"epoch": 3.1,
"learning_rate": 6.900584795321637e-05,
"loss": 0.7184,
"step": 1060
},
{
"epoch": 3.13,
"learning_rate": 6.871345029239767e-05,
"loss": 0.7215,
"step": 1070
},
{
"epoch": 3.16,
"learning_rate": 6.842105263157895e-05,
"loss": 0.718,
"step": 1080
},
{
"epoch": 3.19,
"learning_rate": 6.812865497076024e-05,
"loss": 0.7349,
"step": 1090
},
{
"epoch": 3.21,
"learning_rate": 6.783625730994152e-05,
"loss": 0.7185,
"step": 1100
},
{
"epoch": 3.24,
"learning_rate": 6.754385964912281e-05,
"loss": 0.7486,
"step": 1110
},
{
"epoch": 3.27,
"learning_rate": 6.72514619883041e-05,
"loss": 0.7279,
"step": 1120
},
{
"epoch": 3.3,
"learning_rate": 6.695906432748539e-05,
"loss": 0.7447,
"step": 1130
},
{
"epoch": 3.33,
"learning_rate": 6.666666666666667e-05,
"loss": 0.7423,
"step": 1140
},
{
"epoch": 3.36,
"learning_rate": 6.637426900584795e-05,
"loss": 0.7391,
"step": 1150
},
{
"epoch": 3.39,
"learning_rate": 6.608187134502924e-05,
"loss": 0.7337,
"step": 1160
},
{
"epoch": 3.42,
"learning_rate": 6.578947368421054e-05,
"loss": 0.7282,
"step": 1170
},
{
"epoch": 3.45,
"learning_rate": 6.549707602339182e-05,
"loss": 0.7438,
"step": 1180
},
{
"epoch": 3.48,
"learning_rate": 6.52046783625731e-05,
"loss": 0.7302,
"step": 1190
},
{
"epoch": 3.51,
"learning_rate": 6.49122807017544e-05,
"loss": 0.7418,
"step": 1200
},
{
"epoch": 3.54,
"learning_rate": 6.461988304093567e-05,
"loss": 0.7289,
"step": 1210
},
{
"epoch": 3.57,
"learning_rate": 6.432748538011695e-05,
"loss": 0.7352,
"step": 1220
},
{
"epoch": 3.59,
"learning_rate": 6.403508771929825e-05,
"loss": 0.7342,
"step": 1230
},
{
"epoch": 3.62,
"learning_rate": 6.374269005847954e-05,
"loss": 0.7361,
"step": 1240
},
{
"epoch": 3.65,
"learning_rate": 6.345029239766082e-05,
"loss": 0.7489,
"step": 1250
},
{
"epoch": 3.68,
"learning_rate": 6.31578947368421e-05,
"loss": 0.7334,
"step": 1260
},
{
"epoch": 3.71,
"learning_rate": 6.28654970760234e-05,
"loss": 0.7474,
"step": 1270
},
{
"epoch": 3.74,
"learning_rate": 6.257309941520468e-05,
"loss": 0.7385,
"step": 1280
},
{
"epoch": 3.77,
"learning_rate": 6.228070175438597e-05,
"loss": 0.7334,
"step": 1290
},
{
"epoch": 3.8,
"learning_rate": 6.198830409356725e-05,
"loss": 0.7521,
"step": 1300
},
{
"epoch": 3.83,
"learning_rate": 6.169590643274855e-05,
"loss": 0.7486,
"step": 1310
},
{
"epoch": 3.86,
"learning_rate": 6.140350877192983e-05,
"loss": 0.7468,
"step": 1320
},
{
"epoch": 3.89,
"learning_rate": 6.111111111111112e-05,
"loss": 0.7351,
"step": 1330
},
{
"epoch": 3.92,
"learning_rate": 6.0818713450292395e-05,
"loss": 0.7363,
"step": 1340
},
{
"epoch": 3.95,
"learning_rate": 6.052631578947369e-05,
"loss": 0.7444,
"step": 1350
},
{
"epoch": 3.97,
"learning_rate": 6.0233918128654976e-05,
"loss": 0.7545,
"step": 1360
},
{
"epoch": 4.0,
"eval_loss": 1.2032016515731812,
"eval_runtime": 133.2982,
"eval_samples_per_second": 79.296,
"eval_steps_per_second": 9.918,
"step": 1368
},
{
"epoch": 4.0,
"learning_rate": 5.994152046783626e-05,
"loss": 0.73,
"step": 1370
},
{
"epoch": 4.03,
"learning_rate": 5.9649122807017544e-05,
"loss": 0.6411,
"step": 1380
},
{
"epoch": 4.06,
"learning_rate": 5.935672514619883e-05,
"loss": 0.6374,
"step": 1390
},
{
"epoch": 4.09,
"learning_rate": 5.9064327485380125e-05,
"loss": 0.636,
"step": 1400
},
{
"epoch": 4.12,
"learning_rate": 5.877192982456141e-05,
"loss": 0.6276,
"step": 1410
},
{
"epoch": 4.15,
"learning_rate": 5.847953216374269e-05,
"loss": 0.6348,
"step": 1420
},
{
"epoch": 4.18,
"learning_rate": 5.818713450292398e-05,
"loss": 0.6405,
"step": 1430
},
{
"epoch": 4.21,
"learning_rate": 5.789473684210527e-05,
"loss": 0.6305,
"step": 1440
},
{
"epoch": 4.24,
"learning_rate": 5.7602339181286554e-05,
"loss": 0.6175,
"step": 1450
},
{
"epoch": 4.27,
"learning_rate": 5.7309941520467835e-05,
"loss": 0.6242,
"step": 1460
},
{
"epoch": 4.3,
"learning_rate": 5.701754385964912e-05,
"loss": 0.6253,
"step": 1470
},
{
"epoch": 4.33,
"learning_rate": 5.6725146198830416e-05,
"loss": 0.6299,
"step": 1480
},
{
"epoch": 4.35,
"learning_rate": 5.64327485380117e-05,
"loss": 0.6331,
"step": 1490
},
{
"epoch": 4.38,
"learning_rate": 5.6140350877192984e-05,
"loss": 0.6449,
"step": 1500
},
{
"epoch": 4.41,
"learning_rate": 5.584795321637427e-05,
"loss": 0.6246,
"step": 1510
},
{
"epoch": 4.44,
"learning_rate": 5.555555555555556e-05,
"loss": 0.6248,
"step": 1520
},
{
"epoch": 4.47,
"learning_rate": 5.526315789473685e-05,
"loss": 0.6332,
"step": 1530
},
{
"epoch": 4.5,
"learning_rate": 5.4970760233918126e-05,
"loss": 0.6333,
"step": 1540
},
{
"epoch": 4.53,
"learning_rate": 5.467836257309942e-05,
"loss": 0.6364,
"step": 1550
},
{
"epoch": 4.56,
"learning_rate": 5.438596491228071e-05,
"loss": 0.6289,
"step": 1560
},
{
"epoch": 4.59,
"learning_rate": 5.4093567251461994e-05,
"loss": 0.635,
"step": 1570
},
{
"epoch": 4.62,
"learning_rate": 5.3801169590643275e-05,
"loss": 0.6447,
"step": 1580
},
{
"epoch": 4.65,
"learning_rate": 5.350877192982456e-05,
"loss": 0.6368,
"step": 1590
},
{
"epoch": 4.68,
"learning_rate": 5.3216374269005856e-05,
"loss": 0.649,
"step": 1600
},
{
"epoch": 4.71,
"learning_rate": 5.2923976608187143e-05,
"loss": 0.6354,
"step": 1610
},
{
"epoch": 4.73,
"learning_rate": 5.2631578947368424e-05,
"loss": 0.6386,
"step": 1620
},
{
"epoch": 4.76,
"learning_rate": 5.233918128654971e-05,
"loss": 0.6423,
"step": 1630
},
{
"epoch": 4.79,
"learning_rate": 5.2046783625731e-05,
"loss": 0.6382,
"step": 1640
},
{
"epoch": 4.82,
"learning_rate": 5.1754385964912286e-05,
"loss": 0.6348,
"step": 1650
},
{
"epoch": 4.85,
"learning_rate": 5.1461988304093566e-05,
"loss": 0.6391,
"step": 1660
},
{
"epoch": 4.88,
"learning_rate": 5.1169590643274853e-05,
"loss": 0.6573,
"step": 1670
},
{
"epoch": 4.91,
"learning_rate": 5.087719298245615e-05,
"loss": 0.6919,
"step": 1680
},
{
"epoch": 4.94,
"learning_rate": 5.0584795321637435e-05,
"loss": 0.6598,
"step": 1690
},
{
"epoch": 4.97,
"learning_rate": 5.0292397660818715e-05,
"loss": 0.6526,
"step": 1700
},
{
"epoch": 5.0,
"learning_rate": 5e-05,
"loss": 0.6504,
"step": 1710
},
{
"epoch": 5.0,
"eval_loss": 1.2719171047210693,
"eval_runtime": 133.3649,
"eval_samples_per_second": 79.256,
"eval_steps_per_second": 9.913,
"step": 1710
},
{
"epoch": 5.03,
"learning_rate": 4.970760233918128e-05,
"loss": 0.5766,
"step": 1720
},
{
"epoch": 5.06,
"learning_rate": 4.941520467836258e-05,
"loss": 0.5547,
"step": 1730
},
{
"epoch": 5.08,
"learning_rate": 4.912280701754386e-05,
"loss": 0.5504,
"step": 1740
},
{
"epoch": 5.11,
"learning_rate": 4.883040935672515e-05,
"loss": 0.5593,
"step": 1750
},
{
"epoch": 5.14,
"learning_rate": 4.853801169590643e-05,
"loss": 0.5595,
"step": 1760
},
{
"epoch": 5.17,
"learning_rate": 4.824561403508772e-05,
"loss": 0.5492,
"step": 1770
},
{
"epoch": 5.2,
"learning_rate": 4.7953216374269006e-05,
"loss": 0.5603,
"step": 1780
},
{
"epoch": 5.23,
"learning_rate": 4.7660818713450294e-05,
"loss": 0.5475,
"step": 1790
},
{
"epoch": 5.26,
"learning_rate": 4.736842105263158e-05,
"loss": 0.557,
"step": 1800
},
{
"epoch": 5.29,
"learning_rate": 4.707602339181287e-05,
"loss": 0.5563,
"step": 1810
},
{
"epoch": 5.32,
"learning_rate": 4.678362573099415e-05,
"loss": 0.5759,
"step": 1820
},
{
"epoch": 5.35,
"learning_rate": 4.649122807017544e-05,
"loss": 0.5608,
"step": 1830
},
{
"epoch": 5.38,
"learning_rate": 4.619883040935672e-05,
"loss": 0.5651,
"step": 1840
},
{
"epoch": 5.41,
"learning_rate": 4.590643274853802e-05,
"loss": 0.5627,
"step": 1850
},
{
"epoch": 5.44,
"learning_rate": 4.56140350877193e-05,
"loss": 0.5729,
"step": 1860
},
{
"epoch": 5.46,
"learning_rate": 4.5321637426900585e-05,
"loss": 0.5678,
"step": 1870
},
{
"epoch": 5.49,
"learning_rate": 4.502923976608187e-05,
"loss": 0.5661,
"step": 1880
},
{
"epoch": 5.52,
"learning_rate": 4.473684210526316e-05,
"loss": 0.5694,
"step": 1890
},
{
"epoch": 5.55,
"learning_rate": 4.4444444444444447e-05,
"loss": 0.5694,
"step": 1900
},
{
"epoch": 5.58,
"learning_rate": 4.4152046783625734e-05,
"loss": 0.5576,
"step": 1910
},
{
"epoch": 5.61,
"learning_rate": 4.3859649122807014e-05,
"loss": 0.565,
"step": 1920
},
{
"epoch": 5.64,
"learning_rate": 4.356725146198831e-05,
"loss": 0.5655,
"step": 1930
},
{
"epoch": 5.67,
"learning_rate": 4.327485380116959e-05,
"loss": 0.5611,
"step": 1940
},
{
"epoch": 5.7,
"learning_rate": 4.298245614035088e-05,
"loss": 0.5808,
"step": 1950
},
{
"epoch": 5.73,
"learning_rate": 4.269005847953216e-05,
"loss": 0.551,
"step": 1960
},
{
"epoch": 5.76,
"learning_rate": 4.239766081871345e-05,
"loss": 0.5609,
"step": 1970
},
{
"epoch": 5.79,
"learning_rate": 4.210526315789474e-05,
"loss": 0.5629,
"step": 1980
},
{
"epoch": 5.82,
"learning_rate": 4.1812865497076025e-05,
"loss": 0.5675,
"step": 1990
},
{
"epoch": 5.84,
"learning_rate": 4.152046783625731e-05,
"loss": 0.5697,
"step": 2000
},
{
"epoch": 5.87,
"learning_rate": 4.12280701754386e-05,
"loss": 0.5544,
"step": 2010
},
{
"epoch": 5.9,
"learning_rate": 4.093567251461988e-05,
"loss": 0.5676,
"step": 2020
},
{
"epoch": 5.93,
"learning_rate": 4.0643274853801174e-05,
"loss": 0.5622,
"step": 2030
},
{
"epoch": 5.96,
"learning_rate": 4.0350877192982455e-05,
"loss": 0.5656,
"step": 2040
},
{
"epoch": 5.99,
"learning_rate": 4.005847953216375e-05,
"loss": 0.5708,
"step": 2050
},
{
"epoch": 6.0,
"eval_loss": 1.340452790260315,
"eval_runtime": 133.0036,
"eval_samples_per_second": 79.472,
"eval_steps_per_second": 9.94,
"step": 2053
},
{
"epoch": 6.02,
"learning_rate": 3.976608187134503e-05,
"loss": 0.5084,
"step": 2060
},
{
"epoch": 6.05,
"learning_rate": 3.9473684210526316e-05,
"loss": 0.4929,
"step": 2070
},
{
"epoch": 6.08,
"learning_rate": 3.9181286549707604e-05,
"loss": 0.4962,
"step": 2080
},
{
"epoch": 6.11,
"learning_rate": 3.888888888888889e-05,
"loss": 0.4901,
"step": 2090
},
{
"epoch": 6.14,
"learning_rate": 3.859649122807018e-05,
"loss": 0.4999,
"step": 2100
},
{
"epoch": 6.17,
"learning_rate": 3.8304093567251465e-05,
"loss": 0.4905,
"step": 2110
},
{
"epoch": 6.2,
"learning_rate": 3.8011695906432746e-05,
"loss": 0.4871,
"step": 2120
},
{
"epoch": 6.22,
"learning_rate": 3.771929824561404e-05,
"loss": 0.4965,
"step": 2130
},
{
"epoch": 6.25,
"learning_rate": 3.742690058479532e-05,
"loss": 0.4862,
"step": 2140
},
{
"epoch": 6.28,
"learning_rate": 3.713450292397661e-05,
"loss": 0.4926,
"step": 2150
},
{
"epoch": 6.31,
"learning_rate": 3.6842105263157895e-05,
"loss": 0.4931,
"step": 2160
},
{
"epoch": 6.34,
"learning_rate": 3.654970760233918e-05,
"loss": 0.4931,
"step": 2170
},
{
"epoch": 6.37,
"learning_rate": 3.625730994152047e-05,
"loss": 0.5032,
"step": 2180
},
{
"epoch": 6.4,
"learning_rate": 3.5964912280701756e-05,
"loss": 0.5071,
"step": 2190
},
{
"epoch": 6.43,
"learning_rate": 3.5672514619883044e-05,
"loss": 0.4958,
"step": 2200
},
{
"epoch": 6.46,
"learning_rate": 3.538011695906433e-05,
"loss": 0.5028,
"step": 2210
},
{
"epoch": 6.49,
"learning_rate": 3.508771929824561e-05,
"loss": 0.5033,
"step": 2220
},
{
"epoch": 6.52,
"learning_rate": 3.4795321637426905e-05,
"loss": 0.5038,
"step": 2230
},
{
"epoch": 6.55,
"learning_rate": 3.4502923976608186e-05,
"loss": 0.4832,
"step": 2240
},
{
"epoch": 6.58,
"learning_rate": 3.421052631578947e-05,
"loss": 0.5049,
"step": 2250
},
{
"epoch": 6.6,
"learning_rate": 3.391812865497076e-05,
"loss": 0.496,
"step": 2260
},
{
"epoch": 6.63,
"learning_rate": 3.362573099415205e-05,
"loss": 0.5157,
"step": 2270
},
{
"epoch": 6.66,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.496,
"step": 2280
},
{
"epoch": 6.69,
"learning_rate": 3.304093567251462e-05,
"loss": 0.4893,
"step": 2290
},
{
"epoch": 6.72,
"learning_rate": 3.274853801169591e-05,
"loss": 0.4947,
"step": 2300
},
{
"epoch": 6.75,
"learning_rate": 3.24561403508772e-05,
"loss": 0.4953,
"step": 2310
},
{
"epoch": 6.78,
"learning_rate": 3.216374269005848e-05,
"loss": 0.5045,
"step": 2320
},
{
"epoch": 6.81,
"learning_rate": 3.187134502923977e-05,
"loss": 0.4985,
"step": 2330
},
{
"epoch": 6.84,
"learning_rate": 3.157894736842105e-05,
"loss": 0.4938,
"step": 2340
},
{
"epoch": 6.87,
"learning_rate": 3.128654970760234e-05,
"loss": 0.5037,
"step": 2350
},
{
"epoch": 6.9,
"learning_rate": 3.0994152046783626e-05,
"loss": 0.4967,
"step": 2360
},
{
"epoch": 6.93,
"learning_rate": 3.0701754385964913e-05,
"loss": 0.4956,
"step": 2370
},
{
"epoch": 6.96,
"learning_rate": 3.0409356725146197e-05,
"loss": 0.4982,
"step": 2380
},
{
"epoch": 6.98,
"learning_rate": 3.0116959064327488e-05,
"loss": 0.5077,
"step": 2390
},
{
"epoch": 7.0,
"eval_loss": 1.416868805885315,
"eval_runtime": 133.3362,
"eval_samples_per_second": 79.273,
"eval_steps_per_second": 9.915,
"step": 2395
},
{
"epoch": 7.01,
"learning_rate": 2.9824561403508772e-05,
"loss": 0.4655,
"step": 2400
},
{
"epoch": 7.04,
"learning_rate": 2.9532163742690062e-05,
"loss": 0.4391,
"step": 2410
},
{
"epoch": 7.07,
"learning_rate": 2.9239766081871346e-05,
"loss": 0.4432,
"step": 2420
},
{
"epoch": 7.1,
"learning_rate": 2.8947368421052634e-05,
"loss": 0.4383,
"step": 2430
},
{
"epoch": 7.13,
"learning_rate": 2.8654970760233917e-05,
"loss": 0.4335,
"step": 2440
},
{
"epoch": 7.16,
"learning_rate": 2.8362573099415208e-05,
"loss": 0.4526,
"step": 2450
},
{
"epoch": 7.19,
"learning_rate": 2.8070175438596492e-05,
"loss": 0.4407,
"step": 2460
},
{
"epoch": 7.22,
"learning_rate": 2.777777777777778e-05,
"loss": 0.4462,
"step": 2470
},
{
"epoch": 7.25,
"learning_rate": 2.7485380116959063e-05,
"loss": 0.4548,
"step": 2480
},
{
"epoch": 7.28,
"learning_rate": 2.7192982456140354e-05,
"loss": 0.4519,
"step": 2490
},
{
"epoch": 7.31,
"learning_rate": 2.6900584795321637e-05,
"loss": 0.4407,
"step": 2500
},
{
"epoch": 7.34,
"learning_rate": 2.6608187134502928e-05,
"loss": 0.4479,
"step": 2510
},
{
"epoch": 7.36,
"learning_rate": 2.6315789473684212e-05,
"loss": 0.456,
"step": 2520
},
{
"epoch": 7.39,
"learning_rate": 2.60233918128655e-05,
"loss": 0.4416,
"step": 2530
},
{
"epoch": 7.42,
"learning_rate": 2.5730994152046783e-05,
"loss": 0.4542,
"step": 2540
},
{
"epoch": 7.45,
"learning_rate": 2.5438596491228074e-05,
"loss": 0.4582,
"step": 2550
},
{
"epoch": 7.48,
"learning_rate": 2.5146198830409358e-05,
"loss": 0.4545,
"step": 2560
},
{
"epoch": 7.51,
"learning_rate": 2.485380116959064e-05,
"loss": 0.4476,
"step": 2570
},
{
"epoch": 7.54,
"learning_rate": 2.456140350877193e-05,
"loss": 0.4472,
"step": 2580
},
{
"epoch": 7.57,
"learning_rate": 2.4269005847953216e-05,
"loss": 0.4567,
"step": 2590
},
{
"epoch": 7.6,
"learning_rate": 2.3976608187134503e-05,
"loss": 0.4689,
"step": 2600
},
{
"epoch": 7.63,
"learning_rate": 2.368421052631579e-05,
"loss": 0.4587,
"step": 2610
},
{
"epoch": 7.66,
"learning_rate": 2.3391812865497074e-05,
"loss": 0.4494,
"step": 2620
},
{
"epoch": 7.69,
"learning_rate": 2.309941520467836e-05,
"loss": 0.4494,
"step": 2630
},
{
"epoch": 7.72,
"learning_rate": 2.280701754385965e-05,
"loss": 0.4447,
"step": 2640
},
{
"epoch": 7.74,
"learning_rate": 2.2514619883040936e-05,
"loss": 0.4506,
"step": 2650
},
{
"epoch": 7.77,
"learning_rate": 2.2222222222222223e-05,
"loss": 0.4435,
"step": 2660
},
{
"epoch": 7.8,
"learning_rate": 2.1929824561403507e-05,
"loss": 0.4374,
"step": 2670
},
{
"epoch": 7.83,
"learning_rate": 2.1637426900584794e-05,
"loss": 0.4503,
"step": 2680
},
{
"epoch": 7.86,
"learning_rate": 2.134502923976608e-05,
"loss": 0.4491,
"step": 2690
},
{
"epoch": 7.89,
"learning_rate": 2.105263157894737e-05,
"loss": 0.4479,
"step": 2700
},
{
"epoch": 7.92,
"learning_rate": 2.0760233918128656e-05,
"loss": 0.4404,
"step": 2710
},
{
"epoch": 7.95,
"learning_rate": 2.046783625730994e-05,
"loss": 0.4483,
"step": 2720
},
{
"epoch": 7.98,
"learning_rate": 2.0175438596491227e-05,
"loss": 0.452,
"step": 2730
},
{
"epoch": 8.0,
"eval_loss": 1.4924665689468384,
"eval_runtime": 133.3032,
"eval_samples_per_second": 79.293,
"eval_steps_per_second": 9.917,
"step": 2737
},
{
"epoch": 8.01,
"learning_rate": 1.9883040935672515e-05,
"loss": 0.4449,
"step": 2740
},
{
"epoch": 8.04,
"learning_rate": 1.9590643274853802e-05,
"loss": 0.4108,
"step": 2750
},
{
"epoch": 8.07,
"learning_rate": 1.929824561403509e-05,
"loss": 0.4182,
"step": 2760
},
{
"epoch": 8.09,
"learning_rate": 1.9005847953216373e-05,
"loss": 0.4163,
"step": 2770
},
{
"epoch": 8.12,
"learning_rate": 1.871345029239766e-05,
"loss": 0.3994,
"step": 2780
},
{
"epoch": 8.15,
"learning_rate": 1.8421052631578947e-05,
"loss": 0.4102,
"step": 2790
},
{
"epoch": 8.18,
"learning_rate": 1.8128654970760235e-05,
"loss": 0.4185,
"step": 2800
},
{
"epoch": 8.21,
"learning_rate": 1.7836257309941522e-05,
"loss": 0.4114,
"step": 2810
},
{
"epoch": 8.24,
"learning_rate": 1.7543859649122806e-05,
"loss": 0.4082,
"step": 2820
},
{
"epoch": 8.27,
"learning_rate": 1.7251461988304093e-05,
"loss": 0.4179,
"step": 2830
},
{
"epoch": 8.3,
"learning_rate": 1.695906432748538e-05,
"loss": 0.4098,
"step": 2840
},
{
"epoch": 8.33,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.4143,
"step": 2850
},
{
"epoch": 8.36,
"learning_rate": 1.6374269005847955e-05,
"loss": 0.4146,
"step": 2860
},
{
"epoch": 8.39,
"learning_rate": 1.608187134502924e-05,
"loss": 0.405,
"step": 2870
},
{
"epoch": 8.42,
"learning_rate": 1.5789473684210526e-05,
"loss": 0.4176,
"step": 2880
},
{
"epoch": 8.45,
"learning_rate": 1.5497076023391813e-05,
"loss": 0.4099,
"step": 2890
},
{
"epoch": 8.47,
"learning_rate": 1.5204678362573099e-05,
"loss": 0.4022,
"step": 2900
},
{
"epoch": 8.5,
"learning_rate": 1.4912280701754386e-05,
"loss": 0.4215,
"step": 2910
},
{
"epoch": 8.53,
"learning_rate": 1.4619883040935673e-05,
"loss": 0.4115,
"step": 2920
},
{
"epoch": 8.56,
"learning_rate": 1.4327485380116959e-05,
"loss": 0.4194,
"step": 2930
},
{
"epoch": 8.59,
"learning_rate": 1.4035087719298246e-05,
"loss": 0.4145,
"step": 2940
},
{
"epoch": 8.62,
"learning_rate": 1.3742690058479531e-05,
"loss": 0.4121,
"step": 2950
},
{
"epoch": 8.65,
"learning_rate": 1.3450292397660819e-05,
"loss": 0.409,
"step": 2960
},
{
"epoch": 8.68,
"learning_rate": 1.3157894736842106e-05,
"loss": 0.4139,
"step": 2970
},
{
"epoch": 8.71,
"learning_rate": 1.2865497076023392e-05,
"loss": 0.4091,
"step": 2980
},
{
"epoch": 8.74,
"learning_rate": 1.2573099415204679e-05,
"loss": 0.4111,
"step": 2990
},
{
"epoch": 8.77,
"learning_rate": 1.2280701754385964e-05,
"loss": 0.4187,
"step": 3000
},
{
"epoch": 8.8,
"learning_rate": 1.1988304093567252e-05,
"loss": 0.4153,
"step": 3010
},
{
"epoch": 8.83,
"learning_rate": 1.1695906432748537e-05,
"loss": 0.4184,
"step": 3020
},
{
"epoch": 8.85,
"learning_rate": 1.1403508771929824e-05,
"loss": 0.4155,
"step": 3030
},
{
"epoch": 8.88,
"learning_rate": 1.1111111111111112e-05,
"loss": 0.4108,
"step": 3040
},
{
"epoch": 8.91,
"learning_rate": 1.0818713450292397e-05,
"loss": 0.4184,
"step": 3050
},
{
"epoch": 8.94,
"learning_rate": 1.0526315789473684e-05,
"loss": 0.4152,
"step": 3060
},
{
"epoch": 8.97,
"learning_rate": 1.023391812865497e-05,
"loss": 0.4063,
"step": 3070
},
{
"epoch": 9.0,
"eval_loss": 1.5488338470458984,
"eval_runtime": 133.0253,
"eval_samples_per_second": 79.459,
"eval_steps_per_second": 9.938,
"step": 3079
},
{
"epoch": 9.0,
"learning_rate": 9.941520467836257e-06,
"loss": 0.4007,
"step": 3080
},
{
"epoch": 9.03,
"learning_rate": 9.649122807017545e-06,
"loss": 0.391,
"step": 3090
},
{
"epoch": 9.06,
"learning_rate": 9.35672514619883e-06,
"loss": 0.3912,
"step": 3100
},
{
"epoch": 9.09,
"learning_rate": 9.064327485380117e-06,
"loss": 0.3925,
"step": 3110
},
{
"epoch": 9.12,
"learning_rate": 8.771929824561403e-06,
"loss": 0.3948,
"step": 3120
},
{
"epoch": 9.15,
"learning_rate": 8.47953216374269e-06,
"loss": 0.3826,
"step": 3130
},
{
"epoch": 9.18,
"learning_rate": 8.187134502923977e-06,
"loss": 0.4017,
"step": 3140
},
{
"epoch": 9.21,
"learning_rate": 7.894736842105263e-06,
"loss": 0.4034,
"step": 3150
},
{
"epoch": 9.23,
"learning_rate": 7.602339181286549e-06,
"loss": 0.3979,
"step": 3160
},
{
"epoch": 9.26,
"learning_rate": 7.3099415204678366e-06,
"loss": 0.3977,
"step": 3170
},
{
"epoch": 9.29,
"learning_rate": 7.017543859649123e-06,
"loss": 0.3863,
"step": 3180
},
{
"epoch": 9.32,
"learning_rate": 6.725146198830409e-06,
"loss": 0.3948,
"step": 3190
},
{
"epoch": 9.35,
"learning_rate": 6.432748538011696e-06,
"loss": 0.3949,
"step": 3200
},
{
"epoch": 9.38,
"learning_rate": 6.140350877192982e-06,
"loss": 0.3886,
"step": 3210
},
{
"epoch": 9.41,
"learning_rate": 5.8479532163742686e-06,
"loss": 0.3825,
"step": 3220
},
{
"epoch": 9.44,
"learning_rate": 5.555555555555556e-06,
"loss": 0.3844,
"step": 3230
},
{
"epoch": 9.47,
"learning_rate": 5.263157894736842e-06,
"loss": 0.3958,
"step": 3240
},
{
"epoch": 9.5,
"learning_rate": 4.970760233918129e-06,
"loss": 0.3972,
"step": 3250
},
{
"epoch": 9.53,
"learning_rate": 4.678362573099415e-06,
"loss": 0.3875,
"step": 3260
},
{
"epoch": 9.56,
"learning_rate": 4.3859649122807014e-06,
"loss": 0.3966,
"step": 3270
},
{
"epoch": 9.59,
"learning_rate": 4.093567251461989e-06,
"loss": 0.3868,
"step": 3280
},
{
"epoch": 9.61,
"learning_rate": 3.8011695906432747e-06,
"loss": 0.3945,
"step": 3290
},
{
"epoch": 9.64,
"learning_rate": 3.5087719298245615e-06,
"loss": 0.3953,
"step": 3300
},
{
"epoch": 9.67,
"learning_rate": 3.216374269005848e-06,
"loss": 0.3939,
"step": 3310
},
{
"epoch": 9.7,
"learning_rate": 2.9239766081871343e-06,
"loss": 0.3935,
"step": 3320
},
{
"epoch": 9.73,
"learning_rate": 2.631578947368421e-06,
"loss": 0.3936,
"step": 3330
},
{
"epoch": 9.76,
"learning_rate": 2.3391812865497075e-06,
"loss": 0.3958,
"step": 3340
},
{
"epoch": 9.79,
"learning_rate": 2.0467836257309943e-06,
"loss": 0.3952,
"step": 3350
},
{
"epoch": 9.82,
"learning_rate": 1.7543859649122807e-06,
"loss": 0.3856,
"step": 3360
},
{
"epoch": 9.85,
"learning_rate": 1.4619883040935671e-06,
"loss": 0.382,
"step": 3370
},
{
"epoch": 9.88,
"learning_rate": 1.1695906432748538e-06,
"loss": 0.3845,
"step": 3380
},
{
"epoch": 9.91,
"learning_rate": 8.771929824561404e-07,
"loss": 0.3855,
"step": 3390
},
{
"epoch": 9.94,
"learning_rate": 5.847953216374269e-07,
"loss": 0.3948,
"step": 3400
},
{
"epoch": 9.97,
"learning_rate": 2.9239766081871344e-07,
"loss": 0.4005,
"step": 3410
},
{
"epoch": 9.99,
"learning_rate": 0.0,
"loss": 0.3863,
"step": 3420
},
{
"epoch": 9.99,
"eval_loss": 1.5926491022109985,
"eval_runtime": 133.4623,
"eval_samples_per_second": 79.198,
"eval_steps_per_second": 9.905,
"step": 3420
},
{
"epoch": 9.99,
"step": 3420,
"total_flos": 4.554369038996914e+18,
"train_loss": 0.6737501335422895,
"train_runtime": 54303.7835,
"train_samples_per_second": 16.131,
"train_steps_per_second": 0.063
}
],
"max_steps": 3420,
"num_train_epochs": 10,
"total_flos": 4.554369038996914e+18,
"trial_name": null,
"trial_params": null
}