hemlatachanne's picture
End of training
2a82b8e verified
{
"best_metric": 0.9951154529307282,
"best_model_checkpoint": "vit-hybrid-base-bit-384-finetuned-ibird/checkpoint-1131",
"epoch": 2.9973474801061006,
"eval_steps": 500,
"global_step": 1695,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"grad_norm": 10.60638427734375,
"learning_rate": 2.9411764705882355e-06,
"loss": 3.3842,
"step": 10
},
{
"epoch": 0.04,
"grad_norm": 10.456100463867188,
"learning_rate": 5.882352941176471e-06,
"loss": 3.1282,
"step": 20
},
{
"epoch": 0.05,
"grad_norm": 9.986790657043457,
"learning_rate": 8.823529411764707e-06,
"loss": 2.8815,
"step": 30
},
{
"epoch": 0.07,
"grad_norm": 10.580012321472168,
"learning_rate": 1.1764705882352942e-05,
"loss": 2.3087,
"step": 40
},
{
"epoch": 0.09,
"grad_norm": 7.9744133949279785,
"learning_rate": 1.4705882352941177e-05,
"loss": 1.5822,
"step": 50
},
{
"epoch": 0.11,
"grad_norm": 6.688446044921875,
"learning_rate": 1.7647058823529414e-05,
"loss": 0.9381,
"step": 60
},
{
"epoch": 0.12,
"grad_norm": 4.64487361907959,
"learning_rate": 2.058823529411765e-05,
"loss": 0.4115,
"step": 70
},
{
"epoch": 0.14,
"grad_norm": 5.120860576629639,
"learning_rate": 2.3529411764705884e-05,
"loss": 0.1915,
"step": 80
},
{
"epoch": 0.16,
"grad_norm": 2.6903035640716553,
"learning_rate": 2.647058823529412e-05,
"loss": 0.0935,
"step": 90
},
{
"epoch": 0.18,
"grad_norm": 2.5217204093933105,
"learning_rate": 2.9411764705882354e-05,
"loss": 0.0949,
"step": 100
},
{
"epoch": 0.19,
"grad_norm": 0.7664793133735657,
"learning_rate": 3.235294117647059e-05,
"loss": 0.0545,
"step": 110
},
{
"epoch": 0.21,
"grad_norm": 4.868194580078125,
"learning_rate": 3.529411764705883e-05,
"loss": 0.0495,
"step": 120
},
{
"epoch": 0.23,
"grad_norm": 3.152761459350586,
"learning_rate": 3.8235294117647055e-05,
"loss": 0.0566,
"step": 130
},
{
"epoch": 0.25,
"grad_norm": 3.1319901943206787,
"learning_rate": 4.11764705882353e-05,
"loss": 0.0522,
"step": 140
},
{
"epoch": 0.27,
"grad_norm": 2.599245309829712,
"learning_rate": 4.411764705882353e-05,
"loss": 0.0605,
"step": 150
},
{
"epoch": 0.28,
"grad_norm": 0.845979630947113,
"learning_rate": 4.705882352941177e-05,
"loss": 0.0521,
"step": 160
},
{
"epoch": 0.3,
"grad_norm": 6.3421759605407715,
"learning_rate": 5e-05,
"loss": 0.0849,
"step": 170
},
{
"epoch": 0.32,
"grad_norm": 2.421968698501587,
"learning_rate": 4.967213114754098e-05,
"loss": 0.0502,
"step": 180
},
{
"epoch": 0.34,
"grad_norm": 0.6349968910217285,
"learning_rate": 4.934426229508197e-05,
"loss": 0.0161,
"step": 190
},
{
"epoch": 0.35,
"grad_norm": 0.3493717908859253,
"learning_rate": 4.9016393442622957e-05,
"loss": 0.027,
"step": 200
},
{
"epoch": 0.37,
"grad_norm": 0.24897021055221558,
"learning_rate": 4.868852459016394e-05,
"loss": 0.0481,
"step": 210
},
{
"epoch": 0.39,
"grad_norm": 0.10974353551864624,
"learning_rate": 4.836065573770492e-05,
"loss": 0.017,
"step": 220
},
{
"epoch": 0.41,
"grad_norm": 2.431795358657837,
"learning_rate": 4.8032786885245904e-05,
"loss": 0.0494,
"step": 230
},
{
"epoch": 0.42,
"grad_norm": 3.4805123805999756,
"learning_rate": 4.770491803278689e-05,
"loss": 0.0388,
"step": 240
},
{
"epoch": 0.44,
"grad_norm": 2.0460708141326904,
"learning_rate": 4.737704918032787e-05,
"loss": 0.0507,
"step": 250
},
{
"epoch": 0.46,
"grad_norm": 0.7923760414123535,
"learning_rate": 4.704918032786885e-05,
"loss": 0.0781,
"step": 260
},
{
"epoch": 0.48,
"grad_norm": 2.829329252243042,
"learning_rate": 4.672131147540984e-05,
"loss": 0.0446,
"step": 270
},
{
"epoch": 0.5,
"grad_norm": 0.8546658158302307,
"learning_rate": 4.6393442622950825e-05,
"loss": 0.0543,
"step": 280
},
{
"epoch": 0.51,
"grad_norm": 0.2265116274356842,
"learning_rate": 4.6065573770491805e-05,
"loss": 0.0528,
"step": 290
},
{
"epoch": 0.53,
"grad_norm": 2.7868409156799316,
"learning_rate": 4.5737704918032786e-05,
"loss": 0.0442,
"step": 300
},
{
"epoch": 0.55,
"grad_norm": 0.05122404918074608,
"learning_rate": 4.540983606557377e-05,
"loss": 0.0473,
"step": 310
},
{
"epoch": 0.57,
"grad_norm": 4.265463829040527,
"learning_rate": 4.508196721311476e-05,
"loss": 0.0358,
"step": 320
},
{
"epoch": 0.58,
"grad_norm": 0.3292916417121887,
"learning_rate": 4.475409836065574e-05,
"loss": 0.0291,
"step": 330
},
{
"epoch": 0.6,
"grad_norm": 1.2873409986495972,
"learning_rate": 4.442622950819673e-05,
"loss": 0.0681,
"step": 340
},
{
"epoch": 0.62,
"grad_norm": 0.5806503891944885,
"learning_rate": 4.409836065573771e-05,
"loss": 0.0838,
"step": 350
},
{
"epoch": 0.64,
"grad_norm": 4.996669769287109,
"learning_rate": 4.377049180327869e-05,
"loss": 0.0182,
"step": 360
},
{
"epoch": 0.65,
"grad_norm": 0.058223385363817215,
"learning_rate": 4.3442622950819674e-05,
"loss": 0.0454,
"step": 370
},
{
"epoch": 0.67,
"grad_norm": 0.5088571906089783,
"learning_rate": 4.311475409836066e-05,
"loss": 0.0444,
"step": 380
},
{
"epoch": 0.69,
"grad_norm": 6.739996433258057,
"learning_rate": 4.278688524590164e-05,
"loss": 0.0445,
"step": 390
},
{
"epoch": 0.71,
"grad_norm": 0.22746062278747559,
"learning_rate": 4.245901639344262e-05,
"loss": 0.0258,
"step": 400
},
{
"epoch": 0.73,
"grad_norm": 0.04663905128836632,
"learning_rate": 4.213114754098361e-05,
"loss": 0.0023,
"step": 410
},
{
"epoch": 0.74,
"grad_norm": 0.3145718574523926,
"learning_rate": 4.1803278688524595e-05,
"loss": 0.0205,
"step": 420
},
{
"epoch": 0.76,
"grad_norm": 0.06759347021579742,
"learning_rate": 4.1475409836065575e-05,
"loss": 0.0779,
"step": 430
},
{
"epoch": 0.78,
"grad_norm": 0.7269498705863953,
"learning_rate": 4.1147540983606556e-05,
"loss": 0.0194,
"step": 440
},
{
"epoch": 0.8,
"grad_norm": 0.6663650274276733,
"learning_rate": 4.081967213114754e-05,
"loss": 0.0495,
"step": 450
},
{
"epoch": 0.81,
"grad_norm": 0.015244757756590843,
"learning_rate": 4.049180327868853e-05,
"loss": 0.0217,
"step": 460
},
{
"epoch": 0.83,
"grad_norm": 0.05687205493450165,
"learning_rate": 4.016393442622951e-05,
"loss": 0.0157,
"step": 470
},
{
"epoch": 0.85,
"grad_norm": 0.06807345896959305,
"learning_rate": 3.983606557377049e-05,
"loss": 0.0021,
"step": 480
},
{
"epoch": 0.87,
"grad_norm": 6.138463497161865,
"learning_rate": 3.950819672131148e-05,
"loss": 0.0473,
"step": 490
},
{
"epoch": 0.88,
"grad_norm": 0.23685503005981445,
"learning_rate": 3.9180327868852464e-05,
"loss": 0.0289,
"step": 500
},
{
"epoch": 0.9,
"grad_norm": 0.19919660687446594,
"learning_rate": 3.8852459016393444e-05,
"loss": 0.0659,
"step": 510
},
{
"epoch": 0.92,
"grad_norm": 2.63227915763855,
"learning_rate": 3.8524590163934424e-05,
"loss": 0.0618,
"step": 520
},
{
"epoch": 0.94,
"grad_norm": 0.18995435535907745,
"learning_rate": 3.819672131147541e-05,
"loss": 0.0151,
"step": 530
},
{
"epoch": 0.95,
"grad_norm": 0.05399240553379059,
"learning_rate": 3.78688524590164e-05,
"loss": 0.0168,
"step": 540
},
{
"epoch": 0.97,
"grad_norm": 0.13513301312923431,
"learning_rate": 3.754098360655738e-05,
"loss": 0.0492,
"step": 550
},
{
"epoch": 0.99,
"grad_norm": 8.097681999206543,
"learning_rate": 3.721311475409836e-05,
"loss": 0.0997,
"step": 560
},
{
"epoch": 1.0,
"eval_accuracy": 0.9906749555950266,
"eval_loss": 0.031034117564558983,
"eval_runtime": 124.3045,
"eval_samples_per_second": 18.117,
"eval_steps_per_second": 2.269,
"step": 565
},
{
"epoch": 1.01,
"grad_norm": 0.08757388591766357,
"learning_rate": 3.6885245901639346e-05,
"loss": 0.0243,
"step": 570
},
{
"epoch": 1.03,
"grad_norm": 0.03581409156322479,
"learning_rate": 3.655737704918033e-05,
"loss": 0.0101,
"step": 580
},
{
"epoch": 1.04,
"grad_norm": 0.01882335916161537,
"learning_rate": 3.622950819672131e-05,
"loss": 0.0034,
"step": 590
},
{
"epoch": 1.06,
"grad_norm": 0.12845346331596375,
"learning_rate": 3.590163934426229e-05,
"loss": 0.006,
"step": 600
},
{
"epoch": 1.08,
"grad_norm": 6.836170196533203,
"learning_rate": 3.557377049180328e-05,
"loss": 0.0054,
"step": 610
},
{
"epoch": 1.1,
"grad_norm": 0.014669831842184067,
"learning_rate": 3.524590163934427e-05,
"loss": 0.0007,
"step": 620
},
{
"epoch": 1.11,
"grad_norm": 0.0303161833435297,
"learning_rate": 3.491803278688525e-05,
"loss": 0.0272,
"step": 630
},
{
"epoch": 1.13,
"grad_norm": 0.010958039201796055,
"learning_rate": 3.459016393442623e-05,
"loss": 0.0016,
"step": 640
},
{
"epoch": 1.15,
"grad_norm": 0.10876976698637009,
"learning_rate": 3.4262295081967214e-05,
"loss": 0.0113,
"step": 650
},
{
"epoch": 1.17,
"grad_norm": 0.024475092068314552,
"learning_rate": 3.39344262295082e-05,
"loss": 0.0199,
"step": 660
},
{
"epoch": 1.18,
"grad_norm": 3.7424721717834473,
"learning_rate": 3.360655737704918e-05,
"loss": 0.0153,
"step": 670
},
{
"epoch": 1.2,
"grad_norm": 3.548234224319458,
"learning_rate": 3.327868852459017e-05,
"loss": 0.0124,
"step": 680
},
{
"epoch": 1.22,
"grad_norm": 0.018901199102401733,
"learning_rate": 3.295081967213115e-05,
"loss": 0.0059,
"step": 690
},
{
"epoch": 1.24,
"grad_norm": 0.01598968915641308,
"learning_rate": 3.2622950819672136e-05,
"loss": 0.0272,
"step": 700
},
{
"epoch": 1.26,
"grad_norm": 0.01613859459757805,
"learning_rate": 3.2295081967213116e-05,
"loss": 0.013,
"step": 710
},
{
"epoch": 1.27,
"grad_norm": 0.030256085097789764,
"learning_rate": 3.19672131147541e-05,
"loss": 0.0061,
"step": 720
},
{
"epoch": 1.29,
"grad_norm": 0.08354157209396362,
"learning_rate": 3.163934426229508e-05,
"loss": 0.0008,
"step": 730
},
{
"epoch": 1.31,
"grad_norm": 0.1932225227355957,
"learning_rate": 3.131147540983606e-05,
"loss": 0.0019,
"step": 740
},
{
"epoch": 1.33,
"grad_norm": 0.4999471604824066,
"learning_rate": 3.098360655737705e-05,
"loss": 0.012,
"step": 750
},
{
"epoch": 1.34,
"grad_norm": 0.016076169908046722,
"learning_rate": 3.065573770491804e-05,
"loss": 0.0006,
"step": 760
},
{
"epoch": 1.36,
"grad_norm": 0.010563875548541546,
"learning_rate": 3.0327868852459017e-05,
"loss": 0.0034,
"step": 770
},
{
"epoch": 1.38,
"grad_norm": 0.020218759775161743,
"learning_rate": 3e-05,
"loss": 0.0016,
"step": 780
},
{
"epoch": 1.4,
"grad_norm": 0.016868168488144875,
"learning_rate": 2.967213114754098e-05,
"loss": 0.0033,
"step": 790
},
{
"epoch": 1.41,
"grad_norm": 0.013298319652676582,
"learning_rate": 2.934426229508197e-05,
"loss": 0.0037,
"step": 800
},
{
"epoch": 1.43,
"grad_norm": 0.062481291592121124,
"learning_rate": 2.901639344262295e-05,
"loss": 0.0118,
"step": 810
},
{
"epoch": 1.45,
"grad_norm": 1.4717094898223877,
"learning_rate": 2.8688524590163935e-05,
"loss": 0.0104,
"step": 820
},
{
"epoch": 1.47,
"grad_norm": 1.5437506437301636,
"learning_rate": 2.8360655737704922e-05,
"loss": 0.0062,
"step": 830
},
{
"epoch": 1.49,
"grad_norm": 0.03338282182812691,
"learning_rate": 2.8032786885245906e-05,
"loss": 0.0025,
"step": 840
},
{
"epoch": 1.5,
"grad_norm": 1.8838506937026978,
"learning_rate": 2.7704918032786886e-05,
"loss": 0.0193,
"step": 850
},
{
"epoch": 1.52,
"grad_norm": 0.007599662058055401,
"learning_rate": 2.737704918032787e-05,
"loss": 0.0083,
"step": 860
},
{
"epoch": 1.54,
"grad_norm": 0.0383157841861248,
"learning_rate": 2.7049180327868856e-05,
"loss": 0.0059,
"step": 870
},
{
"epoch": 1.56,
"grad_norm": 0.004423599224537611,
"learning_rate": 2.6721311475409837e-05,
"loss": 0.0019,
"step": 880
},
{
"epoch": 1.57,
"grad_norm": 7.753259658813477,
"learning_rate": 2.639344262295082e-05,
"loss": 0.0197,
"step": 890
},
{
"epoch": 1.59,
"grad_norm": 0.0051222569309175014,
"learning_rate": 2.6065573770491804e-05,
"loss": 0.0128,
"step": 900
},
{
"epoch": 1.61,
"grad_norm": 0.006367386784404516,
"learning_rate": 2.573770491803279e-05,
"loss": 0.0097,
"step": 910
},
{
"epoch": 1.63,
"grad_norm": 0.016590917482972145,
"learning_rate": 2.540983606557377e-05,
"loss": 0.0163,
"step": 920
},
{
"epoch": 1.64,
"grad_norm": 0.006148540414869785,
"learning_rate": 2.5081967213114754e-05,
"loss": 0.0107,
"step": 930
},
{
"epoch": 1.66,
"grad_norm": 0.021444685757160187,
"learning_rate": 2.4754098360655738e-05,
"loss": 0.008,
"step": 940
},
{
"epoch": 1.68,
"grad_norm": 0.27239102125167847,
"learning_rate": 2.442622950819672e-05,
"loss": 0.0009,
"step": 950
},
{
"epoch": 1.7,
"grad_norm": 2.461951971054077,
"learning_rate": 2.4098360655737705e-05,
"loss": 0.0224,
"step": 960
},
{
"epoch": 1.72,
"grad_norm": 0.009482527151703835,
"learning_rate": 2.377049180327869e-05,
"loss": 0.0032,
"step": 970
},
{
"epoch": 1.73,
"grad_norm": 0.0694417804479599,
"learning_rate": 2.3442622950819672e-05,
"loss": 0.001,
"step": 980
},
{
"epoch": 1.75,
"grad_norm": 0.008940008468925953,
"learning_rate": 2.311475409836066e-05,
"loss": 0.0063,
"step": 990
},
{
"epoch": 1.77,
"grad_norm": 0.008408569730818272,
"learning_rate": 2.278688524590164e-05,
"loss": 0.0026,
"step": 1000
},
{
"epoch": 1.79,
"grad_norm": 0.00473896786570549,
"learning_rate": 2.2459016393442626e-05,
"loss": 0.0162,
"step": 1010
},
{
"epoch": 1.8,
"grad_norm": 0.0444784052670002,
"learning_rate": 2.2131147540983607e-05,
"loss": 0.0004,
"step": 1020
},
{
"epoch": 1.82,
"grad_norm": 0.002952872309833765,
"learning_rate": 2.1803278688524594e-05,
"loss": 0.0079,
"step": 1030
},
{
"epoch": 1.84,
"grad_norm": 0.009093045257031918,
"learning_rate": 2.1475409836065574e-05,
"loss": 0.0049,
"step": 1040
},
{
"epoch": 1.86,
"grad_norm": 5.8584818840026855,
"learning_rate": 2.114754098360656e-05,
"loss": 0.0048,
"step": 1050
},
{
"epoch": 1.87,
"grad_norm": 0.025469429790973663,
"learning_rate": 2.081967213114754e-05,
"loss": 0.0118,
"step": 1060
},
{
"epoch": 1.89,
"grad_norm": 0.005341151263564825,
"learning_rate": 2.0491803278688525e-05,
"loss": 0.0016,
"step": 1070
},
{
"epoch": 1.91,
"grad_norm": 0.006102704908698797,
"learning_rate": 2.0163934426229508e-05,
"loss": 0.001,
"step": 1080
},
{
"epoch": 1.93,
"grad_norm": 0.009585659019649029,
"learning_rate": 1.9836065573770492e-05,
"loss": 0.0003,
"step": 1090
},
{
"epoch": 1.95,
"grad_norm": 0.003977091982960701,
"learning_rate": 1.9508196721311475e-05,
"loss": 0.0034,
"step": 1100
},
{
"epoch": 1.96,
"grad_norm": 0.008223881945014,
"learning_rate": 1.918032786885246e-05,
"loss": 0.0016,
"step": 1110
},
{
"epoch": 1.98,
"grad_norm": 0.014407954178750515,
"learning_rate": 1.8852459016393442e-05,
"loss": 0.0013,
"step": 1120
},
{
"epoch": 2.0,
"grad_norm": 0.3026231825351715,
"learning_rate": 1.8524590163934426e-05,
"loss": 0.0005,
"step": 1130
},
{
"epoch": 2.0,
"eval_accuracy": 0.9951154529307282,
"eval_loss": 0.01963997818529606,
"eval_runtime": 122.5197,
"eval_samples_per_second": 18.381,
"eval_steps_per_second": 2.302,
"step": 1131
},
{
"epoch": 2.02,
"grad_norm": 0.005760945845395327,
"learning_rate": 1.8196721311475413e-05,
"loss": 0.0028,
"step": 1140
},
{
"epoch": 2.03,
"grad_norm": 0.00677720969542861,
"learning_rate": 1.7868852459016393e-05,
"loss": 0.0008,
"step": 1150
},
{
"epoch": 2.05,
"grad_norm": 0.004843315575271845,
"learning_rate": 1.754098360655738e-05,
"loss": 0.002,
"step": 1160
},
{
"epoch": 2.07,
"grad_norm": 0.003445906564593315,
"learning_rate": 1.721311475409836e-05,
"loss": 0.0002,
"step": 1170
},
{
"epoch": 2.09,
"grad_norm": 0.0024833856150507927,
"learning_rate": 1.6885245901639347e-05,
"loss": 0.0028,
"step": 1180
},
{
"epoch": 2.1,
"grad_norm": 0.024947261437773705,
"learning_rate": 1.6557377049180328e-05,
"loss": 0.0003,
"step": 1190
},
{
"epoch": 2.12,
"grad_norm": 0.0022959029302001,
"learning_rate": 1.6229508196721314e-05,
"loss": 0.0002,
"step": 1200
},
{
"epoch": 2.14,
"grad_norm": 0.01967487297952175,
"learning_rate": 1.5901639344262295e-05,
"loss": 0.0003,
"step": 1210
},
{
"epoch": 2.16,
"grad_norm": 0.357111394405365,
"learning_rate": 1.557377049180328e-05,
"loss": 0.0005,
"step": 1220
},
{
"epoch": 2.18,
"grad_norm": 0.0058966344222426414,
"learning_rate": 1.5245901639344262e-05,
"loss": 0.0001,
"step": 1230
},
{
"epoch": 2.19,
"grad_norm": 0.0033721905201673508,
"learning_rate": 1.4918032786885247e-05,
"loss": 0.0002,
"step": 1240
},
{
"epoch": 2.21,
"grad_norm": 0.0022145358379930258,
"learning_rate": 1.4590163934426229e-05,
"loss": 0.0002,
"step": 1250
},
{
"epoch": 2.23,
"grad_norm": 0.035282738506793976,
"learning_rate": 1.4262295081967214e-05,
"loss": 0.0002,
"step": 1260
},
{
"epoch": 2.25,
"grad_norm": 0.0015229666605591774,
"learning_rate": 1.3934426229508196e-05,
"loss": 0.0002,
"step": 1270
},
{
"epoch": 2.26,
"grad_norm": 0.005502559244632721,
"learning_rate": 1.3606557377049181e-05,
"loss": 0.0021,
"step": 1280
},
{
"epoch": 2.28,
"grad_norm": 0.0072832051664590836,
"learning_rate": 1.3278688524590163e-05,
"loss": 0.0001,
"step": 1290
},
{
"epoch": 2.3,
"grad_norm": 0.001554520451463759,
"learning_rate": 1.2950819672131149e-05,
"loss": 0.0002,
"step": 1300
},
{
"epoch": 2.32,
"grad_norm": 0.0054123131558299065,
"learning_rate": 1.2622950819672132e-05,
"loss": 0.0002,
"step": 1310
},
{
"epoch": 2.33,
"grad_norm": 0.013851379975676537,
"learning_rate": 1.2295081967213116e-05,
"loss": 0.0066,
"step": 1320
},
{
"epoch": 2.35,
"grad_norm": 0.0056920950300991535,
"learning_rate": 1.19672131147541e-05,
"loss": 0.0002,
"step": 1330
},
{
"epoch": 2.37,
"grad_norm": 0.00168787338770926,
"learning_rate": 1.1639344262295083e-05,
"loss": 0.0002,
"step": 1340
},
{
"epoch": 2.39,
"grad_norm": 0.005322947632521391,
"learning_rate": 1.1311475409836065e-05,
"loss": 0.0002,
"step": 1350
},
{
"epoch": 2.4,
"grad_norm": 0.002933880779892206,
"learning_rate": 1.0983606557377048e-05,
"loss": 0.0002,
"step": 1360
},
{
"epoch": 2.42,
"grad_norm": 0.0015972905093804002,
"learning_rate": 1.0655737704918032e-05,
"loss": 0.0003,
"step": 1370
},
{
"epoch": 2.44,
"grad_norm": 0.0015742258401587605,
"learning_rate": 1.0327868852459017e-05,
"loss": 0.0003,
"step": 1380
},
{
"epoch": 2.46,
"grad_norm": 0.003087802790105343,
"learning_rate": 1e-05,
"loss": 0.0002,
"step": 1390
},
{
"epoch": 2.48,
"grad_norm": 0.002106766914948821,
"learning_rate": 9.672131147540984e-06,
"loss": 0.0008,
"step": 1400
},
{
"epoch": 2.49,
"grad_norm": 0.058892298489809036,
"learning_rate": 9.344262295081968e-06,
"loss": 0.0003,
"step": 1410
},
{
"epoch": 2.51,
"grad_norm": 0.0024353417102247477,
"learning_rate": 9.016393442622952e-06,
"loss": 0.0002,
"step": 1420
},
{
"epoch": 2.53,
"grad_norm": 0.003295810893177986,
"learning_rate": 8.688524590163935e-06,
"loss": 0.0028,
"step": 1430
},
{
"epoch": 2.55,
"grad_norm": 0.006564658600836992,
"learning_rate": 8.360655737704919e-06,
"loss": 0.0001,
"step": 1440
},
{
"epoch": 2.56,
"grad_norm": 0.17080430686473846,
"learning_rate": 8.032786885245902e-06,
"loss": 0.0005,
"step": 1450
},
{
"epoch": 2.58,
"grad_norm": 0.012207315303385258,
"learning_rate": 7.704918032786886e-06,
"loss": 0.0008,
"step": 1460
},
{
"epoch": 2.6,
"grad_norm": 0.07983198761940002,
"learning_rate": 7.3770491803278695e-06,
"loss": 0.0003,
"step": 1470
},
{
"epoch": 2.62,
"grad_norm": 0.0021029170602560043,
"learning_rate": 7.049180327868852e-06,
"loss": 0.0002,
"step": 1480
},
{
"epoch": 2.63,
"grad_norm": 0.001665195683017373,
"learning_rate": 6.721311475409836e-06,
"loss": 0.0002,
"step": 1490
},
{
"epoch": 2.65,
"grad_norm": 0.0021821956615895033,
"learning_rate": 6.393442622950819e-06,
"loss": 0.0003,
"step": 1500
},
{
"epoch": 2.67,
"grad_norm": 0.0023076001089066267,
"learning_rate": 6.065573770491804e-06,
"loss": 0.0007,
"step": 1510
},
{
"epoch": 2.69,
"grad_norm": 0.002858547493815422,
"learning_rate": 5.737704918032787e-06,
"loss": 0.0002,
"step": 1520
},
{
"epoch": 2.71,
"grad_norm": 0.014774510636925697,
"learning_rate": 5.409836065573771e-06,
"loss": 0.0002,
"step": 1530
},
{
"epoch": 2.72,
"grad_norm": 0.0014713788405060768,
"learning_rate": 5.0819672131147545e-06,
"loss": 0.0001,
"step": 1540
},
{
"epoch": 2.74,
"grad_norm": 0.0027496495749801397,
"learning_rate": 4.754098360655738e-06,
"loss": 0.0002,
"step": 1550
},
{
"epoch": 2.76,
"grad_norm": 0.002773697255179286,
"learning_rate": 4.426229508196722e-06,
"loss": 0.0002,
"step": 1560
},
{
"epoch": 2.78,
"grad_norm": 0.002084126928821206,
"learning_rate": 4.098360655737704e-06,
"loss": 0.0001,
"step": 1570
},
{
"epoch": 2.79,
"grad_norm": 0.0016491630813106894,
"learning_rate": 3.770491803278689e-06,
"loss": 0.0002,
"step": 1580
},
{
"epoch": 2.81,
"grad_norm": 0.003069406608119607,
"learning_rate": 3.4426229508196724e-06,
"loss": 0.0074,
"step": 1590
},
{
"epoch": 2.83,
"grad_norm": 0.001330082886852324,
"learning_rate": 3.114754098360656e-06,
"loss": 0.0003,
"step": 1600
},
{
"epoch": 2.85,
"grad_norm": 0.9783486127853394,
"learning_rate": 2.7868852459016396e-06,
"loss": 0.0007,
"step": 1610
},
{
"epoch": 2.86,
"grad_norm": 0.001621302799321711,
"learning_rate": 2.459016393442623e-06,
"loss": 0.0001,
"step": 1620
},
{
"epoch": 2.88,
"grad_norm": 0.05432431399822235,
"learning_rate": 2.1311475409836067e-06,
"loss": 0.0002,
"step": 1630
},
{
"epoch": 2.9,
"grad_norm": 0.001987756695598364,
"learning_rate": 1.8032786885245903e-06,
"loss": 0.0003,
"step": 1640
},
{
"epoch": 2.92,
"grad_norm": 0.0034532160498201847,
"learning_rate": 1.4754098360655739e-06,
"loss": 0.0001,
"step": 1650
},
{
"epoch": 2.94,
"grad_norm": 0.0017541700508445501,
"learning_rate": 1.1475409836065575e-06,
"loss": 0.0112,
"step": 1660
},
{
"epoch": 2.95,
"grad_norm": 0.02740754559636116,
"learning_rate": 8.19672131147541e-07,
"loss": 0.0002,
"step": 1670
},
{
"epoch": 2.97,
"grad_norm": 0.0015147700905799866,
"learning_rate": 4.918032786885246e-07,
"loss": 0.0001,
"step": 1680
},
{
"epoch": 2.99,
"grad_norm": 2.0458450317382812,
"learning_rate": 1.639344262295082e-07,
"loss": 0.0014,
"step": 1690
},
{
"epoch": 3.0,
"eval_accuracy": 0.9942273534635879,
"eval_loss": 0.02003500424325466,
"eval_runtime": 123.0739,
"eval_samples_per_second": 18.298,
"eval_steps_per_second": 2.291,
"step": 1695
},
{
"epoch": 3.0,
"step": 1695,
"total_flos": 1.4134330246261506e+19,
"train_loss": 0.10345972847079468,
"train_runtime": 4872.3509,
"train_samples_per_second": 11.138,
"train_steps_per_second": 0.348
}
],
"logging_steps": 10,
"max_steps": 1695,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 1.4134330246261506e+19,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}