videomae-base-finetuned-dd / trainer_state.json
RodrigoFardin's picture
End of training
fdf4d85 verified
raw
history blame
27.1 kB
{
"best_metric": 0.8978675645342312,
"best_model_checkpoint": "videomae-base-finetuned-dd\\checkpoint-1344",
"epoch": 4.005917159763314,
"eval_steps": 500,
"global_step": 1352,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0073964497041420114,
"grad_norm": 22.897619247436523,
"learning_rate": 3.6764705882352942e-06,
"loss": 0.7648,
"step": 10
},
{
"epoch": 0.014792899408284023,
"grad_norm": 6.239708423614502,
"learning_rate": 7.3529411764705884e-06,
"loss": 0.6619,
"step": 20
},
{
"epoch": 0.022189349112426034,
"grad_norm": 11.775946617126465,
"learning_rate": 1.1029411764705883e-05,
"loss": 0.6844,
"step": 30
},
{
"epoch": 0.029585798816568046,
"grad_norm": 12.948445320129395,
"learning_rate": 1.4705882352941177e-05,
"loss": 0.6158,
"step": 40
},
{
"epoch": 0.03698224852071006,
"grad_norm": 18.087614059448242,
"learning_rate": 1.8382352941176472e-05,
"loss": 0.6569,
"step": 50
},
{
"epoch": 0.04437869822485207,
"grad_norm": 10.075726509094238,
"learning_rate": 2.2058823529411766e-05,
"loss": 0.6572,
"step": 60
},
{
"epoch": 0.051775147928994084,
"grad_norm": 4.17144250869751,
"learning_rate": 2.5735294117647057e-05,
"loss": 0.3788,
"step": 70
},
{
"epoch": 0.05917159763313609,
"grad_norm": 24.00593376159668,
"learning_rate": 2.9411764705882354e-05,
"loss": 0.5752,
"step": 80
},
{
"epoch": 0.06656804733727811,
"grad_norm": 0.1779569834470749,
"learning_rate": 3.308823529411765e-05,
"loss": 0.488,
"step": 90
},
{
"epoch": 0.07396449704142012,
"grad_norm": 2.1989564895629883,
"learning_rate": 3.6764705882352945e-05,
"loss": 1.0225,
"step": 100
},
{
"epoch": 0.08136094674556213,
"grad_norm": 16.50670623779297,
"learning_rate": 4.044117647058824e-05,
"loss": 1.0776,
"step": 110
},
{
"epoch": 0.08875739644970414,
"grad_norm": 36.278076171875,
"learning_rate": 4.411764705882353e-05,
"loss": 1.1623,
"step": 120
},
{
"epoch": 0.09615384615384616,
"grad_norm": 8.829483032226562,
"learning_rate": 4.7794117647058826e-05,
"loss": 0.9484,
"step": 130
},
{
"epoch": 0.10355029585798817,
"grad_norm": 0.47137251496315,
"learning_rate": 4.983552631578948e-05,
"loss": 0.2775,
"step": 140
},
{
"epoch": 0.11094674556213018,
"grad_norm": 1.3937591314315796,
"learning_rate": 4.942434210526316e-05,
"loss": 0.959,
"step": 150
},
{
"epoch": 0.11834319526627218,
"grad_norm": 14.168859481811523,
"learning_rate": 4.901315789473684e-05,
"loss": 1.2776,
"step": 160
},
{
"epoch": 0.1257396449704142,
"grad_norm": 209.46127319335938,
"learning_rate": 4.860197368421053e-05,
"loss": 1.1028,
"step": 170
},
{
"epoch": 0.13313609467455623,
"grad_norm": 0.22284063696861267,
"learning_rate": 4.819078947368421e-05,
"loss": 0.2779,
"step": 180
},
{
"epoch": 0.14053254437869822,
"grad_norm": 0.16761070489883423,
"learning_rate": 4.7779605263157896e-05,
"loss": 1.05,
"step": 190
},
{
"epoch": 0.14792899408284024,
"grad_norm": 33.303768157958984,
"learning_rate": 4.736842105263158e-05,
"loss": 0.5783,
"step": 200
},
{
"epoch": 0.15532544378698224,
"grad_norm": 1.759582281112671,
"learning_rate": 4.6957236842105265e-05,
"loss": 0.5052,
"step": 210
},
{
"epoch": 0.16272189349112426,
"grad_norm": 0.16995219886302948,
"learning_rate": 4.654605263157895e-05,
"loss": 0.8945,
"step": 220
},
{
"epoch": 0.17011834319526628,
"grad_norm": 5.4278740882873535,
"learning_rate": 4.6134868421052635e-05,
"loss": 1.1972,
"step": 230
},
{
"epoch": 0.17751479289940827,
"grad_norm": 5.3494038581848145,
"learning_rate": 4.572368421052632e-05,
"loss": 0.5657,
"step": 240
},
{
"epoch": 0.1849112426035503,
"grad_norm": 38.29714584350586,
"learning_rate": 4.5312500000000004e-05,
"loss": 0.6919,
"step": 250
},
{
"epoch": 0.19230769230769232,
"grad_norm": 31.136438369750977,
"learning_rate": 4.490131578947369e-05,
"loss": 0.4502,
"step": 260
},
{
"epoch": 0.1997041420118343,
"grad_norm": 16.622379302978516,
"learning_rate": 4.449013157894737e-05,
"loss": 0.4002,
"step": 270
},
{
"epoch": 0.20710059171597633,
"grad_norm": 0.7369871735572815,
"learning_rate": 4.407894736842105e-05,
"loss": 0.7867,
"step": 280
},
{
"epoch": 0.21449704142011836,
"grad_norm": 1.566986083984375,
"learning_rate": 4.3667763157894735e-05,
"loss": 0.9427,
"step": 290
},
{
"epoch": 0.22189349112426035,
"grad_norm": 20.51339340209961,
"learning_rate": 4.3256578947368426e-05,
"loss": 1.3318,
"step": 300
},
{
"epoch": 0.22928994082840237,
"grad_norm": 16.819499969482422,
"learning_rate": 4.284539473684211e-05,
"loss": 0.7456,
"step": 310
},
{
"epoch": 0.23668639053254437,
"grad_norm": 0.06786404550075531,
"learning_rate": 4.2434210526315796e-05,
"loss": 0.0428,
"step": 320
},
{
"epoch": 0.2440828402366864,
"grad_norm": 18.684722900390625,
"learning_rate": 4.202302631578947e-05,
"loss": 0.5777,
"step": 330
},
{
"epoch": 0.2485207100591716,
"eval_accuracy": 0.8372615039281706,
"eval_loss": 0.8213610649108887,
"eval_runtime": 1015.4502,
"eval_samples_per_second": 0.877,
"eval_steps_per_second": 0.439,
"step": 336
},
{
"epoch": 1.0029585798816567,
"grad_norm": 0.05207992345094681,
"learning_rate": 4.161184210526316e-05,
"loss": 0.3122,
"step": 340
},
{
"epoch": 1.0103550295857988,
"grad_norm": 0.08761809021234512,
"learning_rate": 4.120065789473684e-05,
"loss": 0.8974,
"step": 350
},
{
"epoch": 1.017751479289941,
"grad_norm": 2.162889003753662,
"learning_rate": 4.078947368421053e-05,
"loss": 0.5468,
"step": 360
},
{
"epoch": 1.0251479289940828,
"grad_norm": 0.3578147888183594,
"learning_rate": 4.037828947368421e-05,
"loss": 0.4065,
"step": 370
},
{
"epoch": 1.032544378698225,
"grad_norm": 0.1941654235124588,
"learning_rate": 3.9967105263157896e-05,
"loss": 0.5691,
"step": 380
},
{
"epoch": 1.0399408284023668,
"grad_norm": 0.4009058475494385,
"learning_rate": 3.955592105263158e-05,
"loss": 1.0974,
"step": 390
},
{
"epoch": 1.047337278106509,
"grad_norm": 1.8374804258346558,
"learning_rate": 3.9144736842105265e-05,
"loss": 0.3989,
"step": 400
},
{
"epoch": 1.054733727810651,
"grad_norm": 0.1476195603609085,
"learning_rate": 3.873355263157895e-05,
"loss": 0.9985,
"step": 410
},
{
"epoch": 1.0621301775147929,
"grad_norm": 0.23249651491641998,
"learning_rate": 3.8322368421052634e-05,
"loss": 0.0116,
"step": 420
},
{
"epoch": 1.069526627218935,
"grad_norm": 120.95174407958984,
"learning_rate": 3.791118421052632e-05,
"loss": 0.2631,
"step": 430
},
{
"epoch": 1.0769230769230769,
"grad_norm": 0.18491333723068237,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.6313,
"step": 440
},
{
"epoch": 1.084319526627219,
"grad_norm": 139.33804321289062,
"learning_rate": 3.708881578947369e-05,
"loss": 0.8515,
"step": 450
},
{
"epoch": 1.0917159763313609,
"grad_norm": 0.22990567982196808,
"learning_rate": 3.6677631578947366e-05,
"loss": 0.7119,
"step": 460
},
{
"epoch": 1.099112426035503,
"grad_norm": 1.84480881690979,
"learning_rate": 3.626644736842105e-05,
"loss": 0.2762,
"step": 470
},
{
"epoch": 1.106508875739645,
"grad_norm": 12.12067699432373,
"learning_rate": 3.5855263157894735e-05,
"loss": 0.3234,
"step": 480
},
{
"epoch": 1.113905325443787,
"grad_norm": 25.058439254760742,
"learning_rate": 3.5444078947368426e-05,
"loss": 0.869,
"step": 490
},
{
"epoch": 1.121301775147929,
"grad_norm": 20.12900161743164,
"learning_rate": 3.503289473684211e-05,
"loss": 1.085,
"step": 500
},
{
"epoch": 1.128698224852071,
"grad_norm": 12.086252212524414,
"learning_rate": 3.4621710526315795e-05,
"loss": 0.7208,
"step": 510
},
{
"epoch": 1.136094674556213,
"grad_norm": 25.66025161743164,
"learning_rate": 3.421052631578947e-05,
"loss": 0.6842,
"step": 520
},
{
"epoch": 1.143491124260355,
"grad_norm": 15.72556209564209,
"learning_rate": 3.379934210526316e-05,
"loss": 0.7922,
"step": 530
},
{
"epoch": 1.150887573964497,
"grad_norm": 0.22508259117603302,
"learning_rate": 3.338815789473684e-05,
"loss": 0.4978,
"step": 540
},
{
"epoch": 1.1582840236686391,
"grad_norm": 0.3129032254219055,
"learning_rate": 3.297697368421053e-05,
"loss": 0.9733,
"step": 550
},
{
"epoch": 1.165680473372781,
"grad_norm": 2.5810675621032715,
"learning_rate": 3.256578947368421e-05,
"loss": 1.0189,
"step": 560
},
{
"epoch": 1.1730769230769231,
"grad_norm": 0.23227569460868835,
"learning_rate": 3.2154605263157896e-05,
"loss": 0.1902,
"step": 570
},
{
"epoch": 1.180473372781065,
"grad_norm": 41.80963134765625,
"learning_rate": 3.174342105263158e-05,
"loss": 0.676,
"step": 580
},
{
"epoch": 1.1878698224852071,
"grad_norm": 0.23230111598968506,
"learning_rate": 3.1332236842105265e-05,
"loss": 0.7534,
"step": 590
},
{
"epoch": 1.195266272189349,
"grad_norm": 0.22635290026664734,
"learning_rate": 3.092105263157895e-05,
"loss": 0.4239,
"step": 600
},
{
"epoch": 1.202662721893491,
"grad_norm": 1.2736923694610596,
"learning_rate": 3.0509868421052634e-05,
"loss": 0.1946,
"step": 610
},
{
"epoch": 1.2100591715976332,
"grad_norm": 138.90115356445312,
"learning_rate": 3.009868421052632e-05,
"loss": 0.3972,
"step": 620
},
{
"epoch": 1.217455621301775,
"grad_norm": 0.8171183466911316,
"learning_rate": 2.96875e-05,
"loss": 0.3475,
"step": 630
},
{
"epoch": 1.2248520710059172,
"grad_norm": 0.14913396537303925,
"learning_rate": 2.9276315789473684e-05,
"loss": 0.0097,
"step": 640
},
{
"epoch": 1.232248520710059,
"grad_norm": 0.16601596772670746,
"learning_rate": 2.886513157894737e-05,
"loss": 0.7906,
"step": 650
},
{
"epoch": 1.2396449704142012,
"grad_norm": 0.0358225516974926,
"learning_rate": 2.8453947368421054e-05,
"loss": 1.0754,
"step": 660
},
{
"epoch": 1.2470414201183433,
"grad_norm": 18.65558433532715,
"learning_rate": 2.8042763157894735e-05,
"loss": 1.144,
"step": 670
},
{
"epoch": 1.2492603550295858,
"eval_accuracy": 0.8754208754208754,
"eval_loss": 0.4453337490558624,
"eval_runtime": 1018.2198,
"eval_samples_per_second": 0.875,
"eval_steps_per_second": 0.438,
"step": 673
},
{
"epoch": 2.0051775147928996,
"grad_norm": 0.26922106742858887,
"learning_rate": 2.7631578947368426e-05,
"loss": 0.0545,
"step": 680
},
{
"epoch": 2.0125739644970415,
"grad_norm": 0.08718305081129074,
"learning_rate": 2.7220394736842107e-05,
"loss": 0.2609,
"step": 690
},
{
"epoch": 2.0199704142011834,
"grad_norm": 0.062128640711307526,
"learning_rate": 2.6809210526315792e-05,
"loss": 0.205,
"step": 700
},
{
"epoch": 2.0273668639053253,
"grad_norm": 0.03102479875087738,
"learning_rate": 2.6398026315789476e-05,
"loss": 0.2414,
"step": 710
},
{
"epoch": 2.0347633136094676,
"grad_norm": 0.1195179671049118,
"learning_rate": 2.598684210526316e-05,
"loss": 0.8337,
"step": 720
},
{
"epoch": 2.0421597633136095,
"grad_norm": 0.18848393857479095,
"learning_rate": 2.5575657894736842e-05,
"loss": 0.938,
"step": 730
},
{
"epoch": 2.0495562130177514,
"grad_norm": 0.30771324038505554,
"learning_rate": 2.5164473684210527e-05,
"loss": 0.5181,
"step": 740
},
{
"epoch": 2.0569526627218937,
"grad_norm": 45.85493469238281,
"learning_rate": 2.4753289473684215e-05,
"loss": 0.9593,
"step": 750
},
{
"epoch": 2.0643491124260356,
"grad_norm": 0.8830978870391846,
"learning_rate": 2.4342105263157896e-05,
"loss": 0.0696,
"step": 760
},
{
"epoch": 2.0717455621301775,
"grad_norm": 0.23656447231769562,
"learning_rate": 2.393092105263158e-05,
"loss": 0.3063,
"step": 770
},
{
"epoch": 2.0791420118343193,
"grad_norm": 14.187068939208984,
"learning_rate": 2.3519736842105265e-05,
"loss": 1.1477,
"step": 780
},
{
"epoch": 2.0865384615384617,
"grad_norm": 20.795085906982422,
"learning_rate": 2.3108552631578946e-05,
"loss": 0.6215,
"step": 790
},
{
"epoch": 2.0939349112426036,
"grad_norm": 0.27157312631607056,
"learning_rate": 2.2697368421052634e-05,
"loss": 0.271,
"step": 800
},
{
"epoch": 2.1013313609467454,
"grad_norm": 26.25171661376953,
"learning_rate": 2.228618421052632e-05,
"loss": 0.9012,
"step": 810
},
{
"epoch": 2.1087278106508878,
"grad_norm": 0.18786455690860748,
"learning_rate": 2.1875e-05,
"loss": 0.1285,
"step": 820
},
{
"epoch": 2.1161242603550297,
"grad_norm": 0.09338750690221786,
"learning_rate": 2.1463815789473684e-05,
"loss": 0.6236,
"step": 830
},
{
"epoch": 2.1235207100591715,
"grad_norm": 0.15658584237098694,
"learning_rate": 2.105263157894737e-05,
"loss": 0.2534,
"step": 840
},
{
"epoch": 2.1309171597633134,
"grad_norm": 0.03897108510136604,
"learning_rate": 2.0641447368421053e-05,
"loss": 0.0317,
"step": 850
},
{
"epoch": 2.1383136094674557,
"grad_norm": 0.17228557169437408,
"learning_rate": 2.0230263157894738e-05,
"loss": 0.5894,
"step": 860
},
{
"epoch": 2.1457100591715976,
"grad_norm": 37.38359069824219,
"learning_rate": 1.9819078947368423e-05,
"loss": 0.3589,
"step": 870
},
{
"epoch": 2.1531065088757395,
"grad_norm": 14.766074180603027,
"learning_rate": 1.9407894736842107e-05,
"loss": 0.2452,
"step": 880
},
{
"epoch": 2.160502958579882,
"grad_norm": 0.09724285453557968,
"learning_rate": 1.8996710526315788e-05,
"loss": 0.4429,
"step": 890
},
{
"epoch": 2.1678994082840237,
"grad_norm": 0.20985926687717438,
"learning_rate": 1.8585526315789476e-05,
"loss": 0.3037,
"step": 900
},
{
"epoch": 2.1752958579881656,
"grad_norm": 17.582971572875977,
"learning_rate": 1.8174342105263157e-05,
"loss": 0.8678,
"step": 910
},
{
"epoch": 2.1826923076923075,
"grad_norm": 0.10595466196537018,
"learning_rate": 1.7763157894736842e-05,
"loss": 0.5378,
"step": 920
},
{
"epoch": 2.19008875739645,
"grad_norm": 0.08267045021057129,
"learning_rate": 1.7351973684210527e-05,
"loss": 0.3613,
"step": 930
},
{
"epoch": 2.1974852071005917,
"grad_norm": 0.33162182569503784,
"learning_rate": 1.694078947368421e-05,
"loss": 0.4509,
"step": 940
},
{
"epoch": 2.2048816568047336,
"grad_norm": 0.070041224360466,
"learning_rate": 1.6529605263157896e-05,
"loss": 0.2242,
"step": 950
},
{
"epoch": 2.212278106508876,
"grad_norm": 0.07664915174245834,
"learning_rate": 1.611842105263158e-05,
"loss": 0.0046,
"step": 960
},
{
"epoch": 2.219674556213018,
"grad_norm": 84.51457977294922,
"learning_rate": 1.5707236842105265e-05,
"loss": 0.4331,
"step": 970
},
{
"epoch": 2.2270710059171597,
"grad_norm": 0.13595078885555267,
"learning_rate": 1.5296052631578946e-05,
"loss": 0.2439,
"step": 980
},
{
"epoch": 2.234467455621302,
"grad_norm": 107.51094818115234,
"learning_rate": 1.4884868421052634e-05,
"loss": 0.1195,
"step": 990
},
{
"epoch": 2.241863905325444,
"grad_norm": 6.981244087219238,
"learning_rate": 1.4473684210526317e-05,
"loss": 0.2644,
"step": 1000
},
{
"epoch": 2.2492603550295858,
"grad_norm": 0.01494936365634203,
"learning_rate": 1.4062500000000001e-05,
"loss": 0.2783,
"step": 1010
},
{
"epoch": 2.2492603550295858,
"eval_accuracy": 0.8170594837261503,
"eval_loss": 0.8930483460426331,
"eval_runtime": 977.9049,
"eval_samples_per_second": 0.911,
"eval_steps_per_second": 0.456,
"step": 1010
},
{
"epoch": 3.007396449704142,
"grad_norm": 2.669276714324951,
"learning_rate": 1.3651315789473684e-05,
"loss": 0.6067,
"step": 1020
},
{
"epoch": 3.014792899408284,
"grad_norm": 0.0429365374147892,
"learning_rate": 1.3240131578947369e-05,
"loss": 0.2312,
"step": 1030
},
{
"epoch": 3.022189349112426,
"grad_norm": 0.0598057359457016,
"learning_rate": 1.2828947368421055e-05,
"loss": 0.0344,
"step": 1040
},
{
"epoch": 3.029585798816568,
"grad_norm": 0.0926346629858017,
"learning_rate": 1.2417763157894738e-05,
"loss": 0.9614,
"step": 1050
},
{
"epoch": 3.03698224852071,
"grad_norm": 0.01585511490702629,
"learning_rate": 1.200657894736842e-05,
"loss": 0.7316,
"step": 1060
},
{
"epoch": 3.044378698224852,
"grad_norm": 0.10066540539264679,
"learning_rate": 1.1595394736842107e-05,
"loss": 0.0041,
"step": 1070
},
{
"epoch": 3.051775147928994,
"grad_norm": 0.06809567660093307,
"learning_rate": 1.118421052631579e-05,
"loss": 0.0092,
"step": 1080
},
{
"epoch": 3.059171597633136,
"grad_norm": 0.14008688926696777,
"learning_rate": 1.0773026315789474e-05,
"loss": 1.0007,
"step": 1090
},
{
"epoch": 3.0665680473372783,
"grad_norm": 0.030784226953983307,
"learning_rate": 1.0361842105263159e-05,
"loss": 0.4198,
"step": 1100
},
{
"epoch": 3.07396449704142,
"grad_norm": 0.07336018979549408,
"learning_rate": 9.950657894736842e-06,
"loss": 0.2221,
"step": 1110
},
{
"epoch": 3.081360946745562,
"grad_norm": 0.06217151880264282,
"learning_rate": 9.539473684210528e-06,
"loss": 0.2991,
"step": 1120
},
{
"epoch": 3.088757396449704,
"grad_norm": 0.07419371604919434,
"learning_rate": 9.128289473684211e-06,
"loss": 0.6161,
"step": 1130
},
{
"epoch": 3.0961538461538463,
"grad_norm": 0.09845346957445145,
"learning_rate": 8.717105263157894e-06,
"loss": 0.204,
"step": 1140
},
{
"epoch": 3.103550295857988,
"grad_norm": 0.020927123725414276,
"learning_rate": 8.30592105263158e-06,
"loss": 0.5403,
"step": 1150
},
{
"epoch": 3.11094674556213,
"grad_norm": 0.07396041601896286,
"learning_rate": 7.894736842105263e-06,
"loss": 0.4637,
"step": 1160
},
{
"epoch": 3.1183431952662723,
"grad_norm": 0.11606968194246292,
"learning_rate": 7.483552631578948e-06,
"loss": 1.3375,
"step": 1170
},
{
"epoch": 3.1257396449704142,
"grad_norm": 0.08246757835149765,
"learning_rate": 7.072368421052632e-06,
"loss": 0.4418,
"step": 1180
},
{
"epoch": 3.133136094674556,
"grad_norm": 0.17207257449626923,
"learning_rate": 6.661184210526317e-06,
"loss": 0.2426,
"step": 1190
},
{
"epoch": 3.140532544378698,
"grad_norm": 0.49036145210266113,
"learning_rate": 6.25e-06,
"loss": 0.2545,
"step": 1200
},
{
"epoch": 3.1479289940828403,
"grad_norm": 68.61939239501953,
"learning_rate": 5.838815789473685e-06,
"loss": 0.5084,
"step": 1210
},
{
"epoch": 3.155325443786982,
"grad_norm": 0.08831863105297089,
"learning_rate": 5.4276315789473686e-06,
"loss": 0.0049,
"step": 1220
},
{
"epoch": 3.162721893491124,
"grad_norm": 0.10537251830101013,
"learning_rate": 5.016447368421053e-06,
"loss": 0.1406,
"step": 1230
},
{
"epoch": 3.1701183431952664,
"grad_norm": 17.6816349029541,
"learning_rate": 4.605263157894737e-06,
"loss": 0.486,
"step": 1240
},
{
"epoch": 3.1775147928994083,
"grad_norm": 0.461302250623703,
"learning_rate": 4.194078947368421e-06,
"loss": 0.0042,
"step": 1250
},
{
"epoch": 3.18491124260355,
"grad_norm": 0.04792853444814682,
"learning_rate": 3.7828947368421055e-06,
"loss": 0.719,
"step": 1260
},
{
"epoch": 3.1923076923076925,
"grad_norm": 0.058049630373716354,
"learning_rate": 3.3717105263157897e-06,
"loss": 0.2919,
"step": 1270
},
{
"epoch": 3.1997041420118344,
"grad_norm": 13.063404083251953,
"learning_rate": 2.960526315789474e-06,
"loss": 0.2856,
"step": 1280
},
{
"epoch": 3.2071005917159763,
"grad_norm": 0.4628206193447113,
"learning_rate": 2.549342105263158e-06,
"loss": 0.0601,
"step": 1290
},
{
"epoch": 3.214497041420118,
"grad_norm": 0.045592982321977615,
"learning_rate": 2.138157894736842e-06,
"loss": 0.53,
"step": 1300
},
{
"epoch": 3.2218934911242605,
"grad_norm": 0.04378387704491615,
"learning_rate": 1.7269736842105266e-06,
"loss": 0.0033,
"step": 1310
},
{
"epoch": 3.2292899408284024,
"grad_norm": 0.052677396684885025,
"learning_rate": 1.3157894736842106e-06,
"loss": 0.003,
"step": 1320
},
{
"epoch": 3.2366863905325443,
"grad_norm": 0.07824493199586868,
"learning_rate": 9.046052631578948e-07,
"loss": 0.3084,
"step": 1330
},
{
"epoch": 3.2440828402366866,
"grad_norm": 0.05082060024142265,
"learning_rate": 4.934210526315789e-07,
"loss": 0.0037,
"step": 1340
},
{
"epoch": 3.247041420118343,
"eval_accuracy": 0.8978675645342312,
"eval_loss": 0.40774551033973694,
"eval_runtime": 1030.7199,
"eval_samples_per_second": 0.864,
"eval_steps_per_second": 0.433,
"step": 1344
},
{
"epoch": 4.004437869822485,
"grad_norm": 0.045703090727329254,
"learning_rate": 8.223684210526316e-08,
"loss": 0.2113,
"step": 1350
},
{
"epoch": 4.005917159763314,
"eval_accuracy": 0.8978675645342312,
"eval_loss": 0.40793702006340027,
"eval_runtime": 1018.6267,
"eval_samples_per_second": 0.875,
"eval_steps_per_second": 0.438,
"step": 1352
},
{
"epoch": 4.005917159763314,
"step": 1352,
"total_flos": 3.3668665207526523e+18,
"train_loss": 0.5293050300165978,
"train_runtime": 11070.5286,
"train_samples_per_second": 0.244,
"train_steps_per_second": 0.122
},
{
"epoch": 4.005917159763314,
"eval_accuracy": 0.890927624872579,
"eval_loss": 0.4576520323753357,
"eval_runtime": 1130.3017,
"eval_samples_per_second": 0.868,
"eval_steps_per_second": 0.434,
"step": 1352
},
{
"epoch": 4.005917159763314,
"eval_accuracy": 0.890927624872579,
"eval_loss": 0.4576520621776581,
"eval_runtime": 1124.6542,
"eval_samples_per_second": 0.872,
"eval_steps_per_second": 0.437,
"step": 1352
}
],
"logging_steps": 10,
"max_steps": 1352,
"num_input_tokens_seen": 0,
"num_train_epochs": 9223372036854775807,
"save_steps": 500,
"total_flos": 3.3668665207526523e+18,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}