sheepy928's picture
Training in progress, step 4000, checkpoint
94c3689 verified
raw
history blame
65.4 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9666505558240696,
"eval_steps": 500,
"global_step": 4000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"grad_norm": NaN,
"learning_rate": 2.0000000000000002e-07,
"loss": 2.0414,
"step": 10
},
{
"epoch": 0.0,
"grad_norm": 53.515953063964844,
"learning_rate": 6.5e-07,
"loss": 1.8355,
"step": 20
},
{
"epoch": 0.01,
"grad_norm": 27.920907974243164,
"learning_rate": 1.15e-06,
"loss": 1.5172,
"step": 30
},
{
"epoch": 0.01,
"grad_norm": 20.47172737121582,
"learning_rate": 1.65e-06,
"loss": 1.2654,
"step": 40
},
{
"epoch": 0.01,
"grad_norm": 8.903529167175293,
"learning_rate": 2.1499999999999997e-06,
"loss": 1.1095,
"step": 50
},
{
"epoch": 0.01,
"grad_norm": 5.902841567993164,
"learning_rate": 2.65e-06,
"loss": 1.0017,
"step": 60
},
{
"epoch": 0.02,
"grad_norm": 2.9340908527374268,
"learning_rate": 3.1500000000000003e-06,
"loss": 0.8846,
"step": 70
},
{
"epoch": 0.02,
"grad_norm": 8.701367378234863,
"learning_rate": 3.6499999999999998e-06,
"loss": 0.8572,
"step": 80
},
{
"epoch": 0.02,
"grad_norm": 1.3205362558364868,
"learning_rate": 4.15e-06,
"loss": 0.839,
"step": 90
},
{
"epoch": 0.02,
"grad_norm": 3.0334558486938477,
"learning_rate": 4.65e-06,
"loss": 0.7838,
"step": 100
},
{
"epoch": 0.03,
"grad_norm": 1.8086520433425903,
"learning_rate": 5.15e-06,
"loss": 0.7782,
"step": 110
},
{
"epoch": 0.03,
"grad_norm": 29.50135040283203,
"learning_rate": 5.65e-06,
"loss": 0.7883,
"step": 120
},
{
"epoch": 0.03,
"grad_norm": 3.2708683013916016,
"learning_rate": 6.15e-06,
"loss": 0.7961,
"step": 130
},
{
"epoch": 0.03,
"grad_norm": 33.43790817260742,
"learning_rate": 6.650000000000001e-06,
"loss": 0.7626,
"step": 140
},
{
"epoch": 0.04,
"grad_norm": 1.2334959506988525,
"learning_rate": 7.15e-06,
"loss": 0.7732,
"step": 150
},
{
"epoch": 0.04,
"grad_norm": 1.7489686012268066,
"learning_rate": 7.65e-06,
"loss": 0.7715,
"step": 160
},
{
"epoch": 0.04,
"grad_norm": 3.358823776245117,
"learning_rate": 8.15e-06,
"loss": 0.7778,
"step": 170
},
{
"epoch": 0.04,
"grad_norm": 4.348245620727539,
"learning_rate": 8.65e-06,
"loss": 0.7619,
"step": 180
},
{
"epoch": 0.05,
"grad_norm": 0.9023125767707825,
"learning_rate": 9.15e-06,
"loss": 0.7746,
"step": 190
},
{
"epoch": 0.05,
"grad_norm": 0.9158114790916443,
"learning_rate": 9.65e-06,
"loss": 0.7592,
"step": 200
},
{
"epoch": 0.05,
"grad_norm": 0.6256226301193237,
"learning_rate": 1.0150000000000001e-05,
"loss": 0.7787,
"step": 210
},
{
"epoch": 0.05,
"grad_norm": 0.3805778920650482,
"learning_rate": 1.065e-05,
"loss": 0.7811,
"step": 220
},
{
"epoch": 0.06,
"grad_norm": 0.6968041062355042,
"learning_rate": 1.115e-05,
"loss": 0.73,
"step": 230
},
{
"epoch": 0.06,
"grad_norm": 0.9035410284996033,
"learning_rate": 1.1650000000000002e-05,
"loss": 0.7518,
"step": 240
},
{
"epoch": 0.06,
"grad_norm": 1.6727488040924072,
"learning_rate": 1.215e-05,
"loss": 0.7805,
"step": 250
},
{
"epoch": 0.06,
"grad_norm": 0.4153461158275604,
"learning_rate": 1.2650000000000001e-05,
"loss": 0.7706,
"step": 260
},
{
"epoch": 0.07,
"grad_norm": 1.4846651554107666,
"learning_rate": 1.3150000000000001e-05,
"loss": 0.7492,
"step": 270
},
{
"epoch": 0.07,
"grad_norm": 2.6631388664245605,
"learning_rate": 1.3650000000000001e-05,
"loss": 0.7681,
"step": 280
},
{
"epoch": 0.07,
"grad_norm": 0.6325013041496277,
"learning_rate": 1.415e-05,
"loss": 0.7753,
"step": 290
},
{
"epoch": 0.07,
"grad_norm": 0.877907395362854,
"learning_rate": 1.465e-05,
"loss": 0.7188,
"step": 300
},
{
"epoch": 0.07,
"grad_norm": 0.37142279744148254,
"learning_rate": 1.515e-05,
"loss": 0.7204,
"step": 310
},
{
"epoch": 0.08,
"grad_norm": 0.7337246537208557,
"learning_rate": 1.565e-05,
"loss": 0.778,
"step": 320
},
{
"epoch": 0.08,
"grad_norm": 0.5847220420837402,
"learning_rate": 1.6150000000000003e-05,
"loss": 0.7288,
"step": 330
},
{
"epoch": 0.08,
"grad_norm": 3.745180606842041,
"learning_rate": 1.665e-05,
"loss": 0.7531,
"step": 340
},
{
"epoch": 0.08,
"grad_norm": 0.357301265001297,
"learning_rate": 1.7150000000000004e-05,
"loss": 0.7448,
"step": 350
},
{
"epoch": 0.09,
"grad_norm": 0.9032486081123352,
"learning_rate": 1.765e-05,
"loss": 0.7651,
"step": 360
},
{
"epoch": 0.09,
"grad_norm": 0.2864232361316681,
"learning_rate": 1.815e-05,
"loss": 0.7193,
"step": 370
},
{
"epoch": 0.09,
"grad_norm": 1.8560261726379395,
"learning_rate": 1.865e-05,
"loss": 0.7421,
"step": 380
},
{
"epoch": 0.09,
"grad_norm": 0.2971792221069336,
"learning_rate": 1.915e-05,
"loss": 0.7171,
"step": 390
},
{
"epoch": 0.1,
"grad_norm": 0.40850627422332764,
"learning_rate": 1.9650000000000003e-05,
"loss": 0.7459,
"step": 400
},
{
"epoch": 0.1,
"grad_norm": 0.3934139311313629,
"learning_rate": 2.0150000000000002e-05,
"loss": 0.7205,
"step": 410
},
{
"epoch": 0.1,
"grad_norm": 1.8674131631851196,
"learning_rate": 2.065e-05,
"loss": 0.752,
"step": 420
},
{
"epoch": 0.1,
"grad_norm": 0.36707818508148193,
"learning_rate": 2.115e-05,
"loss": 0.7494,
"step": 430
},
{
"epoch": 0.11,
"grad_norm": 0.5044310092926025,
"learning_rate": 2.165e-05,
"loss": 0.7595,
"step": 440
},
{
"epoch": 0.11,
"grad_norm": 1.9025150537490845,
"learning_rate": 2.215e-05,
"loss": 0.7379,
"step": 450
},
{
"epoch": 0.11,
"grad_norm": 0.5004140138626099,
"learning_rate": 2.265e-05,
"loss": 0.7886,
"step": 460
},
{
"epoch": 0.11,
"grad_norm": 3.544482707977295,
"learning_rate": 2.3150000000000004e-05,
"loss": 0.7259,
"step": 470
},
{
"epoch": 0.12,
"grad_norm": 0.33023321628570557,
"learning_rate": 2.365e-05,
"loss": 0.7333,
"step": 480
},
{
"epoch": 0.12,
"grad_norm": 0.3548080325126648,
"learning_rate": 2.415e-05,
"loss": 0.7527,
"step": 490
},
{
"epoch": 0.12,
"grad_norm": 0.42450150847435,
"learning_rate": 2.465e-05,
"loss": 0.7443,
"step": 500
},
{
"epoch": 0.12,
"eval_loss": 0.7428915500640869,
"eval_runtime": 68.3143,
"eval_samples_per_second": 29.276,
"eval_steps_per_second": 0.922,
"step": 500
},
{
"epoch": 0.12,
"grad_norm": 0.3299656808376312,
"learning_rate": 2.515e-05,
"loss": 0.7447,
"step": 510
},
{
"epoch": 0.13,
"grad_norm": 0.27258285880088806,
"learning_rate": 2.5650000000000003e-05,
"loss": 0.7278,
"step": 520
},
{
"epoch": 0.13,
"grad_norm": 0.23753009736537933,
"learning_rate": 2.6150000000000002e-05,
"loss": 0.7721,
"step": 530
},
{
"epoch": 0.13,
"grad_norm": 1.1074901819229126,
"learning_rate": 2.6650000000000004e-05,
"loss": 0.7383,
"step": 540
},
{
"epoch": 0.13,
"grad_norm": 0.428067147731781,
"learning_rate": 2.7150000000000003e-05,
"loss": 0.7508,
"step": 550
},
{
"epoch": 0.14,
"grad_norm": 0.35956478118896484,
"learning_rate": 2.7650000000000005e-05,
"loss": 0.7461,
"step": 560
},
{
"epoch": 0.14,
"grad_norm": 0.23017314076423645,
"learning_rate": 2.815e-05,
"loss": 0.7463,
"step": 570
},
{
"epoch": 0.14,
"grad_norm": 0.20370900630950928,
"learning_rate": 2.865e-05,
"loss": 0.7405,
"step": 580
},
{
"epoch": 0.14,
"grad_norm": 0.24586541950702667,
"learning_rate": 2.915e-05,
"loss": 0.6964,
"step": 590
},
{
"epoch": 0.14,
"grad_norm": 0.27857884764671326,
"learning_rate": 2.965e-05,
"loss": 0.7376,
"step": 600
},
{
"epoch": 0.15,
"grad_norm": 0.2655833065509796,
"learning_rate": 3.015e-05,
"loss": 0.7496,
"step": 610
},
{
"epoch": 0.15,
"grad_norm": 0.17174670100212097,
"learning_rate": 3.065e-05,
"loss": 0.7219,
"step": 620
},
{
"epoch": 0.15,
"grad_norm": 0.22892743349075317,
"learning_rate": 3.115e-05,
"loss": 0.7111,
"step": 630
},
{
"epoch": 0.15,
"grad_norm": 0.5913789868354797,
"learning_rate": 3.1650000000000004e-05,
"loss": 0.7097,
"step": 640
},
{
"epoch": 0.16,
"grad_norm": 0.25620004534721375,
"learning_rate": 3.215e-05,
"loss": 0.7306,
"step": 650
},
{
"epoch": 0.16,
"grad_norm": 0.3311476409435272,
"learning_rate": 3.265e-05,
"loss": 0.7406,
"step": 660
},
{
"epoch": 0.16,
"grad_norm": 0.4708893597126007,
"learning_rate": 3.3150000000000006e-05,
"loss": 0.7071,
"step": 670
},
{
"epoch": 0.16,
"grad_norm": 0.22469288110733032,
"learning_rate": 3.3650000000000005e-05,
"loss": 0.7116,
"step": 680
},
{
"epoch": 0.17,
"grad_norm": 0.2909330725669861,
"learning_rate": 3.415e-05,
"loss": 0.7592,
"step": 690
},
{
"epoch": 0.17,
"grad_norm": 0.21274766325950623,
"learning_rate": 3.465e-05,
"loss": 0.7144,
"step": 700
},
{
"epoch": 0.17,
"grad_norm": 0.23929230868816376,
"learning_rate": 3.515e-05,
"loss": 0.7382,
"step": 710
},
{
"epoch": 0.17,
"grad_norm": 0.3021218180656433,
"learning_rate": 3.565e-05,
"loss": 0.7287,
"step": 720
},
{
"epoch": 0.18,
"grad_norm": 0.23557321727275848,
"learning_rate": 3.615e-05,
"loss": 0.7475,
"step": 730
},
{
"epoch": 0.18,
"grad_norm": 0.22648084163665771,
"learning_rate": 3.665e-05,
"loss": 0.7457,
"step": 740
},
{
"epoch": 0.18,
"grad_norm": 0.2707761824131012,
"learning_rate": 3.715e-05,
"loss": 0.7322,
"step": 750
},
{
"epoch": 0.18,
"grad_norm": 0.2799661457538605,
"learning_rate": 3.765e-05,
"loss": 0.7682,
"step": 760
},
{
"epoch": 0.19,
"grad_norm": 0.326861172914505,
"learning_rate": 3.8150000000000006e-05,
"loss": 0.73,
"step": 770
},
{
"epoch": 0.19,
"grad_norm": 0.31985723972320557,
"learning_rate": 3.8650000000000004e-05,
"loss": 0.744,
"step": 780
},
{
"epoch": 0.19,
"grad_norm": 0.21637533605098724,
"learning_rate": 3.915e-05,
"loss": 0.7161,
"step": 790
},
{
"epoch": 0.19,
"grad_norm": 0.26102888584136963,
"learning_rate": 3.965e-05,
"loss": 0.7553,
"step": 800
},
{
"epoch": 0.2,
"grad_norm": 0.2469174712896347,
"learning_rate": 4.015000000000001e-05,
"loss": 0.7444,
"step": 810
},
{
"epoch": 0.2,
"grad_norm": 0.2022310495376587,
"learning_rate": 4.065e-05,
"loss": 0.7089,
"step": 820
},
{
"epoch": 0.2,
"grad_norm": 0.1729898750782013,
"learning_rate": 4.115e-05,
"loss": 0.7122,
"step": 830
},
{
"epoch": 0.2,
"grad_norm": 0.22450707852840424,
"learning_rate": 4.165e-05,
"loss": 0.7099,
"step": 840
},
{
"epoch": 0.21,
"grad_norm": 0.22389698028564453,
"learning_rate": 4.215e-05,
"loss": 0.7445,
"step": 850
},
{
"epoch": 0.21,
"grad_norm": 0.24962696433067322,
"learning_rate": 4.265e-05,
"loss": 0.7052,
"step": 860
},
{
"epoch": 0.21,
"grad_norm": 0.23780404031276703,
"learning_rate": 4.315e-05,
"loss": 0.7423,
"step": 870
},
{
"epoch": 0.21,
"grad_norm": 0.2434270679950714,
"learning_rate": 4.3650000000000004e-05,
"loss": 0.7291,
"step": 880
},
{
"epoch": 0.22,
"grad_norm": 0.3332350552082062,
"learning_rate": 4.415e-05,
"loss": 0.7306,
"step": 890
},
{
"epoch": 0.22,
"grad_norm": 0.5084495544433594,
"learning_rate": 4.465e-05,
"loss": 0.735,
"step": 900
},
{
"epoch": 0.22,
"grad_norm": 0.27552103996276855,
"learning_rate": 4.5150000000000006e-05,
"loss": 0.7082,
"step": 910
},
{
"epoch": 0.22,
"grad_norm": 0.2184416800737381,
"learning_rate": 4.5650000000000005e-05,
"loss": 0.7066,
"step": 920
},
{
"epoch": 0.22,
"grad_norm": 0.21738290786743164,
"learning_rate": 4.6150000000000004e-05,
"loss": 0.6977,
"step": 930
},
{
"epoch": 0.23,
"grad_norm": 0.21888123452663422,
"learning_rate": 4.665e-05,
"loss": 0.7083,
"step": 940
},
{
"epoch": 0.23,
"grad_norm": 0.2516074776649475,
"learning_rate": 4.715e-05,
"loss": 0.7451,
"step": 950
},
{
"epoch": 0.23,
"grad_norm": 0.39361557364463806,
"learning_rate": 4.765e-05,
"loss": 0.7013,
"step": 960
},
{
"epoch": 0.23,
"grad_norm": 0.2461744099855423,
"learning_rate": 4.815e-05,
"loss": 0.7325,
"step": 970
},
{
"epoch": 0.24,
"grad_norm": 0.2554089426994324,
"learning_rate": 4.8650000000000003e-05,
"loss": 0.7384,
"step": 980
},
{
"epoch": 0.24,
"grad_norm": 0.2717374861240387,
"learning_rate": 4.915e-05,
"loss": 0.7172,
"step": 990
},
{
"epoch": 0.24,
"grad_norm": 0.35919347405433655,
"learning_rate": 4.965e-05,
"loss": 0.6851,
"step": 1000
},
{
"epoch": 0.24,
"eval_loss": 0.7169972658157349,
"eval_runtime": 67.4497,
"eval_samples_per_second": 29.652,
"eval_steps_per_second": 0.934,
"step": 1000
},
{
"epoch": 0.24,
"grad_norm": 0.19895273447036743,
"learning_rate": 4.999999713608037e-05,
"loss": 0.7104,
"step": 1010
},
{
"epoch": 0.25,
"grad_norm": 0.21741873025894165,
"learning_rate": 4.999994622197174e-05,
"loss": 0.7217,
"step": 1020
},
{
"epoch": 0.25,
"grad_norm": 0.3415576219558716,
"learning_rate": 4.999983166535371e-05,
"loss": 0.716,
"step": 1030
},
{
"epoch": 0.25,
"grad_norm": 0.22503221035003662,
"learning_rate": 4.99996534665179e-05,
"loss": 0.7166,
"step": 1040
},
{
"epoch": 0.25,
"grad_norm": 0.22469323873519897,
"learning_rate": 4.999941162591795e-05,
"loss": 0.7173,
"step": 1050
},
{
"epoch": 0.26,
"grad_norm": 0.24848666787147522,
"learning_rate": 4.999910614416952e-05,
"loss": 0.7442,
"step": 1060
},
{
"epoch": 0.26,
"grad_norm": 0.23361551761627197,
"learning_rate": 4.999873702205027e-05,
"loss": 0.7142,
"step": 1070
},
{
"epoch": 0.26,
"grad_norm": 0.2513304054737091,
"learning_rate": 4.999830426049987e-05,
"loss": 0.6998,
"step": 1080
},
{
"epoch": 0.26,
"grad_norm": 0.2846020758152008,
"learning_rate": 4.999780786062003e-05,
"loss": 0.6967,
"step": 1090
},
{
"epoch": 0.27,
"grad_norm": 0.2192545235157013,
"learning_rate": 4.999724782367441e-05,
"loss": 0.7259,
"step": 1100
},
{
"epoch": 0.27,
"grad_norm": 0.37123605608940125,
"learning_rate": 4.999662415108872e-05,
"loss": 0.7228,
"step": 1110
},
{
"epoch": 0.27,
"grad_norm": 0.3963713049888611,
"learning_rate": 4.999593684445063e-05,
"loss": 0.6848,
"step": 1120
},
{
"epoch": 0.27,
"grad_norm": 0.24653227627277374,
"learning_rate": 4.9995185905509836e-05,
"loss": 0.7181,
"step": 1130
},
{
"epoch": 0.28,
"grad_norm": 0.25218069553375244,
"learning_rate": 4.999437133617799e-05,
"loss": 0.7263,
"step": 1140
},
{
"epoch": 0.28,
"grad_norm": 0.20412948727607727,
"learning_rate": 4.9993493138528765e-05,
"loss": 0.6965,
"step": 1150
},
{
"epoch": 0.28,
"grad_norm": 0.21635311841964722,
"learning_rate": 4.9992551314797775e-05,
"loss": 0.7042,
"step": 1160
},
{
"epoch": 0.28,
"grad_norm": 0.21901027858257294,
"learning_rate": 4.999154586738264e-05,
"loss": 0.6967,
"step": 1170
},
{
"epoch": 0.29,
"grad_norm": 0.24104245007038116,
"learning_rate": 4.9990476798842935e-05,
"loss": 0.7009,
"step": 1180
},
{
"epoch": 0.29,
"grad_norm": 0.2433982491493225,
"learning_rate": 4.998934411190018e-05,
"loss": 0.7052,
"step": 1190
},
{
"epoch": 0.29,
"grad_norm": 0.5261262655258179,
"learning_rate": 4.9988147809437876e-05,
"loss": 0.7103,
"step": 1200
},
{
"epoch": 0.29,
"grad_norm": 0.36266064643859863,
"learning_rate": 4.998688789450146e-05,
"loss": 0.7028,
"step": 1210
},
{
"epoch": 0.29,
"grad_norm": 0.2509474754333496,
"learning_rate": 4.9985564370298274e-05,
"loss": 0.7047,
"step": 1220
},
{
"epoch": 0.3,
"grad_norm": 0.1925598382949829,
"learning_rate": 4.9984177240197665e-05,
"loss": 0.6943,
"step": 1230
},
{
"epoch": 0.3,
"grad_norm": 0.22494031488895416,
"learning_rate": 4.998272650773083e-05,
"loss": 0.6953,
"step": 1240
},
{
"epoch": 0.3,
"grad_norm": 0.24152888357639313,
"learning_rate": 4.998121217659092e-05,
"loss": 0.7181,
"step": 1250
},
{
"epoch": 0.3,
"grad_norm": 0.24296504259109497,
"learning_rate": 4.997963425063297e-05,
"loss": 0.6934,
"step": 1260
},
{
"epoch": 0.31,
"grad_norm": 0.2438274323940277,
"learning_rate": 4.9977992733873906e-05,
"loss": 0.7088,
"step": 1270
},
{
"epoch": 0.31,
"grad_norm": 0.21081580221652985,
"learning_rate": 4.997628763049257e-05,
"loss": 0.6833,
"step": 1280
},
{
"epoch": 0.31,
"grad_norm": 0.2111068218946457,
"learning_rate": 4.9974518944829626e-05,
"loss": 0.6922,
"step": 1290
},
{
"epoch": 0.31,
"grad_norm": 0.309772253036499,
"learning_rate": 4.997268668138766e-05,
"loss": 0.7195,
"step": 1300
},
{
"epoch": 0.32,
"grad_norm": 0.2199670672416687,
"learning_rate": 4.997079084483105e-05,
"loss": 0.6924,
"step": 1310
},
{
"epoch": 0.32,
"grad_norm": 0.21781674027442932,
"learning_rate": 4.996883143998605e-05,
"loss": 0.6873,
"step": 1320
},
{
"epoch": 0.32,
"grad_norm": 0.2969052791595459,
"learning_rate": 4.996680847184072e-05,
"loss": 0.7135,
"step": 1330
},
{
"epoch": 0.32,
"grad_norm": 0.23092682659626007,
"learning_rate": 4.996472194554495e-05,
"loss": 0.6827,
"step": 1340
},
{
"epoch": 0.33,
"grad_norm": 0.29615628719329834,
"learning_rate": 4.996257186641042e-05,
"loss": 0.6738,
"step": 1350
},
{
"epoch": 0.33,
"grad_norm": 0.23491479456424713,
"learning_rate": 4.99603582399106e-05,
"loss": 0.7396,
"step": 1360
},
{
"epoch": 0.33,
"grad_norm": 0.1986425668001175,
"learning_rate": 4.9958081071680726e-05,
"loss": 0.6787,
"step": 1370
},
{
"epoch": 0.33,
"grad_norm": 0.19658410549163818,
"learning_rate": 4.99557403675178e-05,
"loss": 0.66,
"step": 1380
},
{
"epoch": 0.34,
"grad_norm": 0.29404616355895996,
"learning_rate": 4.995333613338057e-05,
"loss": 0.6824,
"step": 1390
},
{
"epoch": 0.34,
"grad_norm": 0.23689547181129456,
"learning_rate": 4.9950868375389514e-05,
"loss": 0.7008,
"step": 1400
},
{
"epoch": 0.34,
"grad_norm": 0.22758638858795166,
"learning_rate": 4.99483370998268e-05,
"loss": 0.6946,
"step": 1410
},
{
"epoch": 0.34,
"grad_norm": 0.20259805023670197,
"learning_rate": 4.994574231313634e-05,
"loss": 0.6734,
"step": 1420
},
{
"epoch": 0.35,
"grad_norm": 0.21935436129570007,
"learning_rate": 4.994308402192366e-05,
"loss": 0.7094,
"step": 1430
},
{
"epoch": 0.35,
"grad_norm": 0.23414015769958496,
"learning_rate": 4.9940362232956026e-05,
"loss": 0.7019,
"step": 1440
},
{
"epoch": 0.35,
"grad_norm": 0.2623121440410614,
"learning_rate": 4.993757695316228e-05,
"loss": 0.6793,
"step": 1450
},
{
"epoch": 0.35,
"grad_norm": 0.31137487292289734,
"learning_rate": 4.993472818963295e-05,
"loss": 0.6902,
"step": 1460
},
{
"epoch": 0.36,
"grad_norm": 0.22092680633068085,
"learning_rate": 4.993181594962013e-05,
"loss": 0.6912,
"step": 1470
},
{
"epoch": 0.36,
"grad_norm": 0.23723623156547546,
"learning_rate": 4.992884024053754e-05,
"loss": 0.652,
"step": 1480
},
{
"epoch": 0.36,
"grad_norm": 0.26028233766555786,
"learning_rate": 4.9925801069960454e-05,
"loss": 0.6969,
"step": 1490
},
{
"epoch": 0.36,
"grad_norm": 0.23856200277805328,
"learning_rate": 4.992269844562572e-05,
"loss": 0.6723,
"step": 1500
},
{
"epoch": 0.36,
"eval_loss": 0.6912096738815308,
"eval_runtime": 67.244,
"eval_samples_per_second": 29.742,
"eval_steps_per_second": 0.937,
"step": 1500
},
{
"epoch": 0.36,
"grad_norm": 0.167326420545578,
"learning_rate": 4.9919532375431677e-05,
"loss": 0.6897,
"step": 1510
},
{
"epoch": 0.37,
"grad_norm": 0.26183101534843445,
"learning_rate": 4.991630286743823e-05,
"loss": 0.6844,
"step": 1520
},
{
"epoch": 0.37,
"grad_norm": 0.25918036699295044,
"learning_rate": 4.991300992986676e-05,
"loss": 0.6645,
"step": 1530
},
{
"epoch": 0.37,
"grad_norm": 0.2354690134525299,
"learning_rate": 4.99096535711001e-05,
"loss": 0.6672,
"step": 1540
},
{
"epoch": 0.37,
"grad_norm": 0.24444998800754547,
"learning_rate": 4.990623379968257e-05,
"loss": 0.6749,
"step": 1550
},
{
"epoch": 0.38,
"grad_norm": 0.2562585771083832,
"learning_rate": 4.990275062431989e-05,
"loss": 0.6652,
"step": 1560
},
{
"epoch": 0.38,
"grad_norm": 0.22618699073791504,
"learning_rate": 4.98992040538792e-05,
"loss": 0.7163,
"step": 1570
},
{
"epoch": 0.38,
"grad_norm": 0.20957091450691223,
"learning_rate": 4.9895594097389044e-05,
"loss": 0.7053,
"step": 1580
},
{
"epoch": 0.38,
"grad_norm": 0.26973310112953186,
"learning_rate": 4.989192076403928e-05,
"loss": 0.6716,
"step": 1590
},
{
"epoch": 0.39,
"grad_norm": 0.24841250479221344,
"learning_rate": 4.9888184063181154e-05,
"loss": 0.6884,
"step": 1600
},
{
"epoch": 0.39,
"grad_norm": 0.25283706188201904,
"learning_rate": 4.98843840043272e-05,
"loss": 0.6632,
"step": 1610
},
{
"epoch": 0.39,
"grad_norm": 0.196313738822937,
"learning_rate": 4.988052059715126e-05,
"loss": 0.712,
"step": 1620
},
{
"epoch": 0.39,
"grad_norm": 0.21621187031269073,
"learning_rate": 4.987659385148842e-05,
"loss": 0.6732,
"step": 1630
},
{
"epoch": 0.4,
"grad_norm": 0.2180820256471634,
"learning_rate": 4.987260377733502e-05,
"loss": 0.7045,
"step": 1640
},
{
"epoch": 0.4,
"grad_norm": 0.20110243558883667,
"learning_rate": 4.986855038484862e-05,
"loss": 0.6604,
"step": 1650
},
{
"epoch": 0.4,
"grad_norm": 0.20813095569610596,
"learning_rate": 4.9864433684347964e-05,
"loss": 0.7051,
"step": 1660
},
{
"epoch": 0.4,
"grad_norm": 0.22026021778583527,
"learning_rate": 4.9860253686312964e-05,
"loss": 0.6752,
"step": 1670
},
{
"epoch": 0.41,
"grad_norm": 0.23409809172153473,
"learning_rate": 4.9856010401384654e-05,
"loss": 0.6805,
"step": 1680
},
{
"epoch": 0.41,
"grad_norm": 0.1982721984386444,
"learning_rate": 4.985170384036521e-05,
"loss": 0.6888,
"step": 1690
},
{
"epoch": 0.41,
"grad_norm": 0.24984286725521088,
"learning_rate": 4.984733401421785e-05,
"loss": 0.6741,
"step": 1700
},
{
"epoch": 0.41,
"grad_norm": 0.24465042352676392,
"learning_rate": 4.9842900934066874e-05,
"loss": 0.6781,
"step": 1710
},
{
"epoch": 0.42,
"grad_norm": 0.26656293869018555,
"learning_rate": 4.98384046111976e-05,
"loss": 0.7086,
"step": 1720
},
{
"epoch": 0.42,
"grad_norm": 0.23073184490203857,
"learning_rate": 4.9833845057056336e-05,
"loss": 0.6966,
"step": 1730
},
{
"epoch": 0.42,
"grad_norm": 0.23296333849430084,
"learning_rate": 4.982922228325037e-05,
"loss": 0.6769,
"step": 1740
},
{
"epoch": 0.42,
"grad_norm": 0.22282017767429352,
"learning_rate": 4.982453630154794e-05,
"loss": 0.6581,
"step": 1750
},
{
"epoch": 0.43,
"grad_norm": 0.22943562269210815,
"learning_rate": 4.981978712387815e-05,
"loss": 0.6803,
"step": 1760
},
{
"epoch": 0.43,
"grad_norm": 0.29295608401298523,
"learning_rate": 4.9814974762331034e-05,
"loss": 0.6514,
"step": 1770
},
{
"epoch": 0.43,
"grad_norm": 0.21845972537994385,
"learning_rate": 4.981009922915743e-05,
"loss": 0.691,
"step": 1780
},
{
"epoch": 0.43,
"grad_norm": 0.24616730213165283,
"learning_rate": 4.980516053676903e-05,
"loss": 0.6948,
"step": 1790
},
{
"epoch": 0.43,
"grad_norm": 0.2354935109615326,
"learning_rate": 4.9800158697738264e-05,
"loss": 0.6778,
"step": 1800
},
{
"epoch": 0.44,
"grad_norm": 0.3475668430328369,
"learning_rate": 4.979509372479837e-05,
"loss": 0.6626,
"step": 1810
},
{
"epoch": 0.44,
"grad_norm": 0.2817700207233429,
"learning_rate": 4.9789965630843265e-05,
"loss": 0.6724,
"step": 1820
},
{
"epoch": 0.44,
"grad_norm": 0.28306934237480164,
"learning_rate": 4.978477442892758e-05,
"loss": 0.6673,
"step": 1830
},
{
"epoch": 0.44,
"grad_norm": 0.2593139111995697,
"learning_rate": 4.9779520132266575e-05,
"loss": 0.6866,
"step": 1840
},
{
"epoch": 0.45,
"grad_norm": 0.2618165612220764,
"learning_rate": 4.9774202754236145e-05,
"loss": 0.6671,
"step": 1850
},
{
"epoch": 0.45,
"grad_norm": 0.22875259816646576,
"learning_rate": 4.9768822308372784e-05,
"loss": 0.6743,
"step": 1860
},
{
"epoch": 0.45,
"grad_norm": 0.20143887400627136,
"learning_rate": 4.976337880837351e-05,
"loss": 0.6866,
"step": 1870
},
{
"epoch": 0.45,
"grad_norm": 0.23163306713104248,
"learning_rate": 4.975787226809587e-05,
"loss": 0.6772,
"step": 1880
},
{
"epoch": 0.46,
"grad_norm": 0.21628862619400024,
"learning_rate": 4.975230270155791e-05,
"loss": 0.6706,
"step": 1890
},
{
"epoch": 0.46,
"grad_norm": 0.21055322885513306,
"learning_rate": 4.9746670122938105e-05,
"loss": 0.6536,
"step": 1900
},
{
"epoch": 0.46,
"grad_norm": 0.20045976340770721,
"learning_rate": 4.974097454657534e-05,
"loss": 0.6477,
"step": 1910
},
{
"epoch": 0.46,
"grad_norm": 0.20996223390102386,
"learning_rate": 4.9735215986968874e-05,
"loss": 0.6526,
"step": 1920
},
{
"epoch": 0.47,
"grad_norm": 0.1778746396303177,
"learning_rate": 4.972939445877831e-05,
"loss": 0.6951,
"step": 1930
},
{
"epoch": 0.47,
"grad_norm": 0.2416403591632843,
"learning_rate": 4.972350997682354e-05,
"loss": 0.6703,
"step": 1940
},
{
"epoch": 0.47,
"grad_norm": 0.23652783036231995,
"learning_rate": 4.9717562556084735e-05,
"loss": 0.6401,
"step": 1950
},
{
"epoch": 0.47,
"grad_norm": 0.23598232865333557,
"learning_rate": 4.9711552211702274e-05,
"loss": 0.6668,
"step": 1960
},
{
"epoch": 0.48,
"grad_norm": 0.22741574048995972,
"learning_rate": 4.970547895897672e-05,
"loss": 0.6765,
"step": 1970
},
{
"epoch": 0.48,
"grad_norm": 0.20418991148471832,
"learning_rate": 4.96993428133688e-05,
"loss": 0.6771,
"step": 1980
},
{
"epoch": 0.48,
"grad_norm": 0.24983160197734833,
"learning_rate": 4.969314379049932e-05,
"loss": 0.6656,
"step": 1990
},
{
"epoch": 0.48,
"grad_norm": 0.2511340379714966,
"learning_rate": 4.968688190614919e-05,
"loss": 0.6605,
"step": 2000
},
{
"epoch": 0.48,
"eval_loss": 0.6730201244354248,
"eval_runtime": 67.2807,
"eval_samples_per_second": 29.726,
"eval_steps_per_second": 0.936,
"step": 2000
},
{
"epoch": 0.49,
"grad_norm": 0.22646427154541016,
"learning_rate": 4.96805571762593e-05,
"loss": 0.6729,
"step": 2010
},
{
"epoch": 0.49,
"grad_norm": 0.2343759387731552,
"learning_rate": 4.9674169616930574e-05,
"loss": 0.6635,
"step": 2020
},
{
"epoch": 0.49,
"grad_norm": 0.23573775589466095,
"learning_rate": 4.966771924442385e-05,
"loss": 0.654,
"step": 2030
},
{
"epoch": 0.49,
"grad_norm": 0.2696162760257721,
"learning_rate": 4.966120607515987e-05,
"loss": 0.6728,
"step": 2040
},
{
"epoch": 0.5,
"grad_norm": 0.25032511353492737,
"learning_rate": 4.965463012571927e-05,
"loss": 0.6743,
"step": 2050
},
{
"epoch": 0.5,
"grad_norm": 0.24621817469596863,
"learning_rate": 4.964799141284247e-05,
"loss": 0.6596,
"step": 2060
},
{
"epoch": 0.5,
"grad_norm": 0.23296469449996948,
"learning_rate": 4.964128995342966e-05,
"loss": 0.6471,
"step": 2070
},
{
"epoch": 0.5,
"grad_norm": 0.22769325971603394,
"learning_rate": 4.963452576454082e-05,
"loss": 0.6499,
"step": 2080
},
{
"epoch": 0.51,
"grad_norm": 0.2572191655635834,
"learning_rate": 4.9627698863395564e-05,
"loss": 0.6499,
"step": 2090
},
{
"epoch": 0.51,
"grad_norm": 0.21782545745372772,
"learning_rate": 4.962080926737319e-05,
"loss": 0.6641,
"step": 2100
},
{
"epoch": 0.51,
"grad_norm": 0.22925855219364166,
"learning_rate": 4.9613856994012567e-05,
"loss": 0.6618,
"step": 2110
},
{
"epoch": 0.51,
"grad_norm": 0.2127605825662613,
"learning_rate": 4.960684206101214e-05,
"loss": 0.6821,
"step": 2120
},
{
"epoch": 0.51,
"grad_norm": 0.2520267069339752,
"learning_rate": 4.9599764486229865e-05,
"loss": 0.6748,
"step": 2130
},
{
"epoch": 0.52,
"grad_norm": 0.20282019674777985,
"learning_rate": 4.9592624287683176e-05,
"loss": 0.6674,
"step": 2140
},
{
"epoch": 0.52,
"grad_norm": 0.46226009726524353,
"learning_rate": 4.958542148354891e-05,
"loss": 0.6422,
"step": 2150
},
{
"epoch": 0.52,
"grad_norm": 0.2241724133491516,
"learning_rate": 4.957815609216329e-05,
"loss": 0.6612,
"step": 2160
},
{
"epoch": 0.52,
"grad_norm": 0.23772084712982178,
"learning_rate": 4.957082813202186e-05,
"loss": 0.6614,
"step": 2170
},
{
"epoch": 0.53,
"grad_norm": 0.2049807757139206,
"learning_rate": 4.9563437621779465e-05,
"loss": 0.6827,
"step": 2180
},
{
"epoch": 0.53,
"grad_norm": 0.22666208446025848,
"learning_rate": 4.955598458025015e-05,
"loss": 0.6522,
"step": 2190
},
{
"epoch": 0.53,
"grad_norm": 0.26957106590270996,
"learning_rate": 4.954846902640718e-05,
"loss": 0.6817,
"step": 2200
},
{
"epoch": 0.53,
"grad_norm": 0.22548554837703705,
"learning_rate": 4.954089097938294e-05,
"loss": 0.6444,
"step": 2210
},
{
"epoch": 0.54,
"grad_norm": 0.22197945415973663,
"learning_rate": 4.9533250458468914e-05,
"loss": 0.6779,
"step": 2220
},
{
"epoch": 0.54,
"grad_norm": 0.22711919248104095,
"learning_rate": 4.9525547483115617e-05,
"loss": 0.6416,
"step": 2230
},
{
"epoch": 0.54,
"grad_norm": 0.22576060891151428,
"learning_rate": 4.951778207293255e-05,
"loss": 0.6598,
"step": 2240
},
{
"epoch": 0.54,
"grad_norm": 0.22785042226314545,
"learning_rate": 4.950995424768818e-05,
"loss": 0.6499,
"step": 2250
},
{
"epoch": 0.55,
"grad_norm": 0.20107567310333252,
"learning_rate": 4.9502064027309836e-05,
"loss": 0.6509,
"step": 2260
},
{
"epoch": 0.55,
"grad_norm": 0.23379752039909363,
"learning_rate": 4.94941114318837e-05,
"loss": 0.6551,
"step": 2270
},
{
"epoch": 0.55,
"grad_norm": 0.22154928743839264,
"learning_rate": 4.948609648165475e-05,
"loss": 0.6428,
"step": 2280
},
{
"epoch": 0.55,
"grad_norm": 0.23805582523345947,
"learning_rate": 4.947801919702667e-05,
"loss": 0.6633,
"step": 2290
},
{
"epoch": 0.56,
"grad_norm": 0.24756357073783875,
"learning_rate": 4.946987959856188e-05,
"loss": 0.6561,
"step": 2300
},
{
"epoch": 0.56,
"grad_norm": 0.23767632246017456,
"learning_rate": 4.9461677706981374e-05,
"loss": 0.6552,
"step": 2310
},
{
"epoch": 0.56,
"grad_norm": 0.32177987694740295,
"learning_rate": 4.9453413543164775e-05,
"loss": 0.6705,
"step": 2320
},
{
"epoch": 0.56,
"grad_norm": 0.29961881041526794,
"learning_rate": 4.94450871281502e-05,
"loss": 0.6551,
"step": 2330
},
{
"epoch": 0.57,
"grad_norm": 0.2768830955028534,
"learning_rate": 4.943669848313427e-05,
"loss": 0.6468,
"step": 2340
},
{
"epoch": 0.57,
"grad_norm": 0.24728989601135254,
"learning_rate": 4.9428247629472e-05,
"loss": 0.6506,
"step": 2350
},
{
"epoch": 0.57,
"grad_norm": 0.2641989290714264,
"learning_rate": 4.941973458867677e-05,
"loss": 0.6517,
"step": 2360
},
{
"epoch": 0.57,
"grad_norm": 0.18787774443626404,
"learning_rate": 4.941115938242028e-05,
"loss": 0.6355,
"step": 2370
},
{
"epoch": 0.58,
"grad_norm": 0.26642099022865295,
"learning_rate": 4.940252203253248e-05,
"loss": 0.6811,
"step": 2380
},
{
"epoch": 0.58,
"grad_norm": 0.24736198782920837,
"learning_rate": 4.939382256100154e-05,
"loss": 0.6493,
"step": 2390
},
{
"epoch": 0.58,
"grad_norm": 0.3060832619667053,
"learning_rate": 4.938506098997374e-05,
"loss": 0.6707,
"step": 2400
},
{
"epoch": 0.58,
"grad_norm": 0.24164025485515594,
"learning_rate": 4.937623734175346e-05,
"loss": 0.6508,
"step": 2410
},
{
"epoch": 0.58,
"grad_norm": 0.26685982942581177,
"learning_rate": 4.936735163880313e-05,
"loss": 0.6441,
"step": 2420
},
{
"epoch": 0.59,
"grad_norm": 0.17739726603031158,
"learning_rate": 4.9358403903743124e-05,
"loss": 0.6499,
"step": 2430
},
{
"epoch": 0.59,
"grad_norm": 0.23560577630996704,
"learning_rate": 4.9349394159351735e-05,
"loss": 0.645,
"step": 2440
},
{
"epoch": 0.59,
"grad_norm": 0.23137669265270233,
"learning_rate": 4.9340322428565135e-05,
"loss": 0.6615,
"step": 2450
},
{
"epoch": 0.59,
"grad_norm": 0.3485463857650757,
"learning_rate": 4.933118873447728e-05,
"loss": 0.6283,
"step": 2460
},
{
"epoch": 0.6,
"grad_norm": 0.20602945983409882,
"learning_rate": 4.932199310033987e-05,
"loss": 0.6831,
"step": 2470
},
{
"epoch": 0.6,
"grad_norm": 0.21541930735111237,
"learning_rate": 4.931273554956227e-05,
"loss": 0.6722,
"step": 2480
},
{
"epoch": 0.6,
"grad_norm": 0.2092689573764801,
"learning_rate": 4.930341610571151e-05,
"loss": 0.6609,
"step": 2490
},
{
"epoch": 0.6,
"grad_norm": 0.23792122304439545,
"learning_rate": 4.9294034792512126e-05,
"loss": 0.6475,
"step": 2500
},
{
"epoch": 0.6,
"eval_loss": 0.6643247008323669,
"eval_runtime": 67.0911,
"eval_samples_per_second": 29.81,
"eval_steps_per_second": 0.939,
"step": 2500
},
{
"epoch": 0.61,
"grad_norm": 0.2101823091506958,
"learning_rate": 4.928459163384619e-05,
"loss": 0.6604,
"step": 2510
},
{
"epoch": 0.61,
"grad_norm": 0.1859540194272995,
"learning_rate": 4.927508665375321e-05,
"loss": 0.6252,
"step": 2520
},
{
"epoch": 0.61,
"grad_norm": 0.2371070235967636,
"learning_rate": 4.926551987643007e-05,
"loss": 0.6413,
"step": 2530
},
{
"epoch": 0.61,
"grad_norm": 0.2149161994457245,
"learning_rate": 4.9255891326230964e-05,
"loss": 0.6558,
"step": 2540
},
{
"epoch": 0.62,
"grad_norm": 0.21805475652217865,
"learning_rate": 4.9246201027667354e-05,
"loss": 0.631,
"step": 2550
},
{
"epoch": 0.62,
"grad_norm": 0.22705689072608948,
"learning_rate": 4.9236449005407895e-05,
"loss": 0.6785,
"step": 2560
},
{
"epoch": 0.62,
"grad_norm": 0.21902668476104736,
"learning_rate": 4.9226635284278355e-05,
"loss": 0.657,
"step": 2570
},
{
"epoch": 0.62,
"grad_norm": 0.21482336521148682,
"learning_rate": 4.9216759889261586e-05,
"loss": 0.6514,
"step": 2580
},
{
"epoch": 0.63,
"grad_norm": 0.20008622109889984,
"learning_rate": 4.9206822845497444e-05,
"loss": 0.6516,
"step": 2590
},
{
"epoch": 0.63,
"grad_norm": 0.205527663230896,
"learning_rate": 4.919682417828271e-05,
"loss": 0.6336,
"step": 2600
},
{
"epoch": 0.63,
"grad_norm": 0.263959139585495,
"learning_rate": 4.9186763913071065e-05,
"loss": 0.6449,
"step": 2610
},
{
"epoch": 0.63,
"grad_norm": 0.21950668096542358,
"learning_rate": 4.917664207547297e-05,
"loss": 0.6916,
"step": 2620
},
{
"epoch": 0.64,
"grad_norm": 0.215659499168396,
"learning_rate": 4.916645869125564e-05,
"loss": 0.6441,
"step": 2630
},
{
"epoch": 0.64,
"grad_norm": 0.22789780795574188,
"learning_rate": 4.915621378634301e-05,
"loss": 0.6576,
"step": 2640
},
{
"epoch": 0.64,
"grad_norm": 0.27431604266166687,
"learning_rate": 4.914590738681555e-05,
"loss": 0.6487,
"step": 2650
},
{
"epoch": 0.64,
"grad_norm": 0.21077977120876312,
"learning_rate": 4.913553951891036e-05,
"loss": 0.6688,
"step": 2660
},
{
"epoch": 0.65,
"grad_norm": 0.2798519730567932,
"learning_rate": 4.9125110209020954e-05,
"loss": 0.6618,
"step": 2670
},
{
"epoch": 0.65,
"grad_norm": 0.24790987372398376,
"learning_rate": 4.911461948369731e-05,
"loss": 0.6598,
"step": 2680
},
{
"epoch": 0.65,
"grad_norm": 0.2008027732372284,
"learning_rate": 4.91040673696457e-05,
"loss": 0.6388,
"step": 2690
},
{
"epoch": 0.65,
"grad_norm": 0.23889446258544922,
"learning_rate": 4.9093453893728733e-05,
"loss": 0.6526,
"step": 2700
},
{
"epoch": 0.65,
"grad_norm": 0.1997964233160019,
"learning_rate": 4.908277908296518e-05,
"loss": 0.6521,
"step": 2710
},
{
"epoch": 0.66,
"grad_norm": 0.24335375428199768,
"learning_rate": 4.907204296452997e-05,
"loss": 0.6636,
"step": 2720
},
{
"epoch": 0.66,
"grad_norm": 0.23431982100009918,
"learning_rate": 4.906124556575411e-05,
"loss": 0.6557,
"step": 2730
},
{
"epoch": 0.66,
"grad_norm": 0.21532365679740906,
"learning_rate": 4.90503869141246e-05,
"loss": 0.628,
"step": 2740
},
{
"epoch": 0.66,
"grad_norm": 0.22962962090969086,
"learning_rate": 4.903946703728436e-05,
"loss": 0.6422,
"step": 2750
},
{
"epoch": 0.67,
"grad_norm": 0.22807002067565918,
"learning_rate": 4.90284859630322e-05,
"loss": 0.6631,
"step": 2760
},
{
"epoch": 0.67,
"grad_norm": 0.27668192982673645,
"learning_rate": 4.90174437193227e-05,
"loss": 0.6778,
"step": 2770
},
{
"epoch": 0.67,
"grad_norm": 0.2503020763397217,
"learning_rate": 4.900634033426616e-05,
"loss": 0.6483,
"step": 2780
},
{
"epoch": 0.67,
"grad_norm": 0.3033277094364166,
"learning_rate": 4.8995175836128536e-05,
"loss": 0.6572,
"step": 2790
},
{
"epoch": 0.68,
"grad_norm": 0.2776722311973572,
"learning_rate": 4.898395025333136e-05,
"loss": 0.662,
"step": 2800
},
{
"epoch": 0.68,
"grad_norm": 0.31251537799835205,
"learning_rate": 4.897266361445165e-05,
"loss": 0.6617,
"step": 2810
},
{
"epoch": 0.68,
"grad_norm": 0.38570669293403625,
"learning_rate": 4.8961315948221884e-05,
"loss": 0.6419,
"step": 2820
},
{
"epoch": 0.68,
"grad_norm": 0.23277431726455688,
"learning_rate": 4.894990728352988e-05,
"loss": 0.6373,
"step": 2830
},
{
"epoch": 0.69,
"grad_norm": 0.2519891858100891,
"learning_rate": 4.893843764941874e-05,
"loss": 0.6759,
"step": 2840
},
{
"epoch": 0.69,
"grad_norm": 0.2358052134513855,
"learning_rate": 4.892690707508677e-05,
"loss": 0.6826,
"step": 2850
},
{
"epoch": 0.69,
"grad_norm": 0.23103272914886475,
"learning_rate": 4.8915315589887436e-05,
"loss": 0.624,
"step": 2860
},
{
"epoch": 0.69,
"grad_norm": 0.2128864973783493,
"learning_rate": 4.8903663223329245e-05,
"loss": 0.654,
"step": 2870
},
{
"epoch": 0.7,
"grad_norm": 0.22060807049274445,
"learning_rate": 4.889195000507568e-05,
"loss": 0.6376,
"step": 2880
},
{
"epoch": 0.7,
"grad_norm": 0.24493062496185303,
"learning_rate": 4.888017596494517e-05,
"loss": 0.6648,
"step": 2890
},
{
"epoch": 0.7,
"grad_norm": 0.2580796480178833,
"learning_rate": 4.886834113291094e-05,
"loss": 0.6607,
"step": 2900
},
{
"epoch": 0.7,
"grad_norm": 0.26906147599220276,
"learning_rate": 4.8856445539101016e-05,
"loss": 0.6441,
"step": 2910
},
{
"epoch": 0.71,
"grad_norm": 0.2145451009273529,
"learning_rate": 4.884448921379805e-05,
"loss": 0.6543,
"step": 2920
},
{
"epoch": 0.71,
"grad_norm": 0.2156788855791092,
"learning_rate": 4.8832472187439345e-05,
"loss": 0.6185,
"step": 2930
},
{
"epoch": 0.71,
"grad_norm": 0.25632360577583313,
"learning_rate": 4.882039449061673e-05,
"loss": 0.639,
"step": 2940
},
{
"epoch": 0.71,
"grad_norm": 0.2496800720691681,
"learning_rate": 4.8808256154076436e-05,
"loss": 0.6377,
"step": 2950
},
{
"epoch": 0.72,
"grad_norm": 0.39942485094070435,
"learning_rate": 4.8796057208719124e-05,
"loss": 0.6083,
"step": 2960
},
{
"epoch": 0.72,
"grad_norm": 0.27986374497413635,
"learning_rate": 4.8783797685599706e-05,
"loss": 0.6528,
"step": 2970
},
{
"epoch": 0.72,
"grad_norm": 0.22164517641067505,
"learning_rate": 4.877147761592733e-05,
"loss": 0.6456,
"step": 2980
},
{
"epoch": 0.72,
"grad_norm": 0.29858505725860596,
"learning_rate": 4.875909703106527e-05,
"loss": 0.643,
"step": 2990
},
{
"epoch": 0.72,
"grad_norm": 0.22595979273319244,
"learning_rate": 4.874665596253084e-05,
"loss": 0.6419,
"step": 3000
},
{
"epoch": 0.72,
"eval_loss": 0.6583871841430664,
"eval_runtime": 67.1961,
"eval_samples_per_second": 29.764,
"eval_steps_per_second": 0.938,
"step": 3000
},
{
"epoch": 0.73,
"grad_norm": 0.19871610403060913,
"learning_rate": 4.8734154441995364e-05,
"loss": 0.6556,
"step": 3010
},
{
"epoch": 0.73,
"grad_norm": 0.2072724997997284,
"learning_rate": 4.872159250128401e-05,
"loss": 0.6764,
"step": 3020
},
{
"epoch": 0.73,
"grad_norm": 0.22515198588371277,
"learning_rate": 4.87089701723758e-05,
"loss": 0.6703,
"step": 3030
},
{
"epoch": 0.73,
"grad_norm": 0.2292303442955017,
"learning_rate": 4.869628748740347e-05,
"loss": 0.6288,
"step": 3040
},
{
"epoch": 0.74,
"grad_norm": 0.25133493542671204,
"learning_rate": 4.8683544478653395e-05,
"loss": 0.6652,
"step": 3050
},
{
"epoch": 0.74,
"grad_norm": 0.29863354563713074,
"learning_rate": 4.867074117856555e-05,
"loss": 0.6352,
"step": 3060
},
{
"epoch": 0.74,
"grad_norm": 0.2168973684310913,
"learning_rate": 4.865787761973334e-05,
"loss": 0.6562,
"step": 3070
},
{
"epoch": 0.74,
"grad_norm": 0.2312694936990738,
"learning_rate": 4.864495383490363e-05,
"loss": 0.6463,
"step": 3080
},
{
"epoch": 0.75,
"grad_norm": 0.2333671748638153,
"learning_rate": 4.863196985697655e-05,
"loss": 0.6267,
"step": 3090
},
{
"epoch": 0.75,
"grad_norm": 0.23559433221817017,
"learning_rate": 4.86189257190055e-05,
"loss": 0.623,
"step": 3100
},
{
"epoch": 0.75,
"grad_norm": 0.22314316034317017,
"learning_rate": 4.860582145419703e-05,
"loss": 0.6298,
"step": 3110
},
{
"epoch": 0.75,
"grad_norm": 0.22701682150363922,
"learning_rate": 4.859265709591073e-05,
"loss": 0.6359,
"step": 3120
},
{
"epoch": 0.76,
"grad_norm": 0.21683742105960846,
"learning_rate": 4.857943267765919e-05,
"loss": 0.6666,
"step": 3130
},
{
"epoch": 0.76,
"grad_norm": 0.2519407272338867,
"learning_rate": 4.856614823310788e-05,
"loss": 0.6731,
"step": 3140
},
{
"epoch": 0.76,
"grad_norm": 0.23536355793476105,
"learning_rate": 4.855280379607509e-05,
"loss": 0.629,
"step": 3150
},
{
"epoch": 0.76,
"grad_norm": 0.19676193594932556,
"learning_rate": 4.853939940053184e-05,
"loss": 0.6329,
"step": 3160
},
{
"epoch": 0.77,
"grad_norm": 0.23497731983661652,
"learning_rate": 4.852593508060177e-05,
"loss": 0.6584,
"step": 3170
},
{
"epoch": 0.77,
"grad_norm": 0.2515724003314972,
"learning_rate": 4.8512410870561084e-05,
"loss": 0.6177,
"step": 3180
},
{
"epoch": 0.77,
"grad_norm": 0.20130635797977448,
"learning_rate": 4.8498826804838436e-05,
"loss": 0.6645,
"step": 3190
},
{
"epoch": 0.77,
"grad_norm": 0.23828361928462982,
"learning_rate": 4.8485182918014876e-05,
"loss": 0.6401,
"step": 3200
},
{
"epoch": 0.78,
"grad_norm": 0.2598424255847931,
"learning_rate": 4.847147924482371e-05,
"loss": 0.6193,
"step": 3210
},
{
"epoch": 0.78,
"grad_norm": 0.20697768032550812,
"learning_rate": 4.845771582015046e-05,
"loss": 0.626,
"step": 3220
},
{
"epoch": 0.78,
"grad_norm": 0.19661648571491241,
"learning_rate": 4.8443892679032775e-05,
"loss": 0.6283,
"step": 3230
},
{
"epoch": 0.78,
"grad_norm": 0.22154510021209717,
"learning_rate": 4.843000985666028e-05,
"loss": 0.6571,
"step": 3240
},
{
"epoch": 0.79,
"grad_norm": 0.3334987759590149,
"learning_rate": 4.841606738837458e-05,
"loss": 0.6553,
"step": 3250
},
{
"epoch": 0.79,
"grad_norm": 0.6604078412055969,
"learning_rate": 4.8402065309669085e-05,
"loss": 0.6239,
"step": 3260
},
{
"epoch": 0.79,
"grad_norm": 0.22238047420978546,
"learning_rate": 4.838800365618898e-05,
"loss": 0.6725,
"step": 3270
},
{
"epoch": 0.79,
"grad_norm": 0.21566419303417206,
"learning_rate": 4.837388246373108e-05,
"loss": 0.6597,
"step": 3280
},
{
"epoch": 0.8,
"grad_norm": 0.9337890148162842,
"learning_rate": 4.83597017682438e-05,
"loss": 0.6454,
"step": 3290
},
{
"epoch": 0.8,
"grad_norm": 0.2716673016548157,
"learning_rate": 4.8345461605827014e-05,
"loss": 0.6526,
"step": 3300
},
{
"epoch": 0.8,
"grad_norm": 0.23388756811618805,
"learning_rate": 4.8331162012732e-05,
"loss": 0.6642,
"step": 3310
},
{
"epoch": 0.8,
"grad_norm": 0.21243353188037872,
"learning_rate": 4.8316803025361304e-05,
"loss": 0.6233,
"step": 3320
},
{
"epoch": 0.8,
"grad_norm": 0.28600814938545227,
"learning_rate": 4.8302384680268684e-05,
"loss": 0.6559,
"step": 3330
},
{
"epoch": 0.81,
"grad_norm": 0.19718654453754425,
"learning_rate": 4.8287907014159004e-05,
"loss": 0.6357,
"step": 3340
},
{
"epoch": 0.81,
"grad_norm": 0.2330627292394638,
"learning_rate": 4.827337006388816e-05,
"loss": 0.6692,
"step": 3350
},
{
"epoch": 0.81,
"grad_norm": 0.18776822090148926,
"learning_rate": 4.8258773866462927e-05,
"loss": 0.6804,
"step": 3360
},
{
"epoch": 0.81,
"grad_norm": 0.24929189682006836,
"learning_rate": 4.8244118459040944e-05,
"loss": 0.6455,
"step": 3370
},
{
"epoch": 0.82,
"grad_norm": 0.23884710669517517,
"learning_rate": 4.8229403878930566e-05,
"loss": 0.6345,
"step": 3380
},
{
"epoch": 0.82,
"grad_norm": 0.2579725682735443,
"learning_rate": 4.821463016359078e-05,
"loss": 0.6584,
"step": 3390
},
{
"epoch": 0.82,
"grad_norm": 0.2741701900959015,
"learning_rate": 4.8199797350631136e-05,
"loss": 0.6463,
"step": 3400
},
{
"epoch": 0.82,
"grad_norm": 0.2254386991262436,
"learning_rate": 4.818490547781159e-05,
"loss": 0.6399,
"step": 3410
},
{
"epoch": 0.83,
"grad_norm": 0.2172977328300476,
"learning_rate": 4.816995458304249e-05,
"loss": 0.661,
"step": 3420
},
{
"epoch": 0.83,
"grad_norm": 0.23464016616344452,
"learning_rate": 4.815494470438441e-05,
"loss": 0.6516,
"step": 3430
},
{
"epoch": 0.83,
"grad_norm": 0.27163979411125183,
"learning_rate": 4.813987588004807e-05,
"loss": 0.6451,
"step": 3440
},
{
"epoch": 0.83,
"grad_norm": 0.20852161943912506,
"learning_rate": 4.8124748148394285e-05,
"loss": 0.6394,
"step": 3450
},
{
"epoch": 0.84,
"grad_norm": 0.2224467694759369,
"learning_rate": 4.81095615479338e-05,
"loss": 0.6427,
"step": 3460
},
{
"epoch": 0.84,
"grad_norm": 0.2205616980791092,
"learning_rate": 4.8094316117327245e-05,
"loss": 0.6491,
"step": 3470
},
{
"epoch": 0.84,
"grad_norm": 0.21984492242336273,
"learning_rate": 4.8079011895384985e-05,
"loss": 0.6271,
"step": 3480
},
{
"epoch": 0.84,
"grad_norm": 0.29770565032958984,
"learning_rate": 4.806364892106707e-05,
"loss": 0.6048,
"step": 3490
},
{
"epoch": 0.85,
"grad_norm": 0.2440354973077774,
"learning_rate": 4.8048227233483127e-05,
"loss": 0.6307,
"step": 3500
},
{
"epoch": 0.85,
"eval_loss": 0.6532164216041565,
"eval_runtime": 67.3224,
"eval_samples_per_second": 29.708,
"eval_steps_per_second": 0.936,
"step": 3500
},
{
"epoch": 0.85,
"grad_norm": 0.25720447301864624,
"learning_rate": 4.803274687189222e-05,
"loss": 0.6337,
"step": 3510
},
{
"epoch": 0.85,
"grad_norm": 0.21051880717277527,
"learning_rate": 4.8017207875702814e-05,
"loss": 0.6398,
"step": 3520
},
{
"epoch": 0.85,
"grad_norm": 0.2977493107318878,
"learning_rate": 4.800161028447261e-05,
"loss": 0.6346,
"step": 3530
},
{
"epoch": 0.86,
"grad_norm": 0.19749422371387482,
"learning_rate": 4.798595413790848e-05,
"loss": 0.6325,
"step": 3540
},
{
"epoch": 0.86,
"grad_norm": 0.21522672474384308,
"learning_rate": 4.7970239475866386e-05,
"loss": 0.6244,
"step": 3550
},
{
"epoch": 0.86,
"grad_norm": 0.24835069477558136,
"learning_rate": 4.7954466338351224e-05,
"loss": 0.612,
"step": 3560
},
{
"epoch": 0.86,
"grad_norm": 0.25459495186805725,
"learning_rate": 4.793863476551677e-05,
"loss": 0.6317,
"step": 3570
},
{
"epoch": 0.87,
"grad_norm": 0.28858670592308044,
"learning_rate": 4.7922744797665544e-05,
"loss": 0.6485,
"step": 3580
},
{
"epoch": 0.87,
"grad_norm": 0.30466488003730774,
"learning_rate": 4.790679647524873e-05,
"loss": 0.6307,
"step": 3590
},
{
"epoch": 0.87,
"grad_norm": 0.36087316274642944,
"learning_rate": 4.789078983886607e-05,
"loss": 0.6749,
"step": 3600
},
{
"epoch": 0.87,
"grad_norm": 0.2436164766550064,
"learning_rate": 4.787472492926575e-05,
"loss": 0.6239,
"step": 3610
},
{
"epoch": 0.87,
"grad_norm": 0.2527044415473938,
"learning_rate": 4.78586017873443e-05,
"loss": 0.6584,
"step": 3620
},
{
"epoch": 0.88,
"grad_norm": 0.25492456555366516,
"learning_rate": 4.784242045414651e-05,
"loss": 0.6487,
"step": 3630
},
{
"epoch": 0.88,
"grad_norm": 0.2293190211057663,
"learning_rate": 4.782618097086528e-05,
"loss": 0.646,
"step": 3640
},
{
"epoch": 0.88,
"grad_norm": 0.7955949902534485,
"learning_rate": 4.780988337884157e-05,
"loss": 0.6391,
"step": 3650
},
{
"epoch": 0.88,
"grad_norm": 0.2331811785697937,
"learning_rate": 4.779352771956425e-05,
"loss": 0.6267,
"step": 3660
},
{
"epoch": 0.89,
"grad_norm": 0.23316140472888947,
"learning_rate": 4.7777114034670006e-05,
"loss": 0.6405,
"step": 3670
},
{
"epoch": 0.89,
"grad_norm": 0.2253628969192505,
"learning_rate": 4.776064236594327e-05,
"loss": 0.6415,
"step": 3680
},
{
"epoch": 0.89,
"grad_norm": 0.2378271222114563,
"learning_rate": 4.774411275531606e-05,
"loss": 0.6362,
"step": 3690
},
{
"epoch": 0.89,
"grad_norm": 0.22514688968658447,
"learning_rate": 4.7727525244867896e-05,
"loss": 0.6375,
"step": 3700
},
{
"epoch": 0.9,
"grad_norm": 0.20843249559402466,
"learning_rate": 4.771087987682571e-05,
"loss": 0.6166,
"step": 3710
},
{
"epoch": 0.9,
"grad_norm": 0.2776215076446533,
"learning_rate": 4.7694176693563705e-05,
"loss": 0.6441,
"step": 3720
},
{
"epoch": 0.9,
"grad_norm": 0.21339628100395203,
"learning_rate": 4.767741573760327e-05,
"loss": 0.6388,
"step": 3730
},
{
"epoch": 0.9,
"grad_norm": 0.24429263174533844,
"learning_rate": 4.766059705161288e-05,
"loss": 0.6263,
"step": 3740
},
{
"epoch": 0.91,
"grad_norm": 0.2099929004907608,
"learning_rate": 4.764372067840795e-05,
"loss": 0.6383,
"step": 3750
},
{
"epoch": 0.91,
"grad_norm": 0.22066868841648102,
"learning_rate": 4.7626786660950784e-05,
"loss": 0.6313,
"step": 3760
},
{
"epoch": 0.91,
"grad_norm": 0.22953028976917267,
"learning_rate": 4.760979504235038e-05,
"loss": 0.6264,
"step": 3770
},
{
"epoch": 0.91,
"grad_norm": 0.3549204170703888,
"learning_rate": 4.759274586586242e-05,
"loss": 0.6368,
"step": 3780
},
{
"epoch": 0.92,
"grad_norm": 0.24681401252746582,
"learning_rate": 4.757563917488909e-05,
"loss": 0.6422,
"step": 3790
},
{
"epoch": 0.92,
"grad_norm": 0.2417946755886078,
"learning_rate": 4.755847501297898e-05,
"loss": 0.6438,
"step": 3800
},
{
"epoch": 0.92,
"grad_norm": 0.2778252959251404,
"learning_rate": 4.7541253423827006e-05,
"loss": 0.6312,
"step": 3810
},
{
"epoch": 0.92,
"grad_norm": 0.20280030369758606,
"learning_rate": 4.7523974451274275e-05,
"loss": 0.6349,
"step": 3820
},
{
"epoch": 0.93,
"grad_norm": 0.24433667957782745,
"learning_rate": 4.7506638139307966e-05,
"loss": 0.6648,
"step": 3830
},
{
"epoch": 0.93,
"grad_norm": 0.3273336887359619,
"learning_rate": 4.7489244532061225e-05,
"loss": 0.6498,
"step": 3840
},
{
"epoch": 0.93,
"grad_norm": 0.23943009972572327,
"learning_rate": 4.747179367381307e-05,
"loss": 0.6358,
"step": 3850
},
{
"epoch": 0.93,
"grad_norm": 0.26670822501182556,
"learning_rate": 4.745428560898824e-05,
"loss": 0.6563,
"step": 3860
},
{
"epoch": 0.94,
"grad_norm": 0.18882083892822266,
"learning_rate": 4.7436720382157116e-05,
"loss": 0.6358,
"step": 3870
},
{
"epoch": 0.94,
"grad_norm": 0.23588411509990692,
"learning_rate": 4.741909803803562e-05,
"loss": 0.6396,
"step": 3880
},
{
"epoch": 0.94,
"grad_norm": 0.24063749611377716,
"learning_rate": 4.740141862148503e-05,
"loss": 0.6412,
"step": 3890
},
{
"epoch": 0.94,
"grad_norm": 0.19155845046043396,
"learning_rate": 4.738368217751196e-05,
"loss": 0.62,
"step": 3900
},
{
"epoch": 0.94,
"grad_norm": 0.20259159803390503,
"learning_rate": 4.736588875126816e-05,
"loss": 0.6172,
"step": 3910
},
{
"epoch": 0.95,
"grad_norm": 0.22824090719223022,
"learning_rate": 4.734803838805048e-05,
"loss": 0.6286,
"step": 3920
},
{
"epoch": 0.95,
"grad_norm": 0.2356242537498474,
"learning_rate": 4.7330131133300686e-05,
"loss": 0.6121,
"step": 3930
},
{
"epoch": 0.95,
"grad_norm": 0.22052697837352753,
"learning_rate": 4.731216703260538e-05,
"loss": 0.6402,
"step": 3940
},
{
"epoch": 0.95,
"grad_norm": 0.2817922532558441,
"learning_rate": 4.7294146131695874e-05,
"loss": 0.6256,
"step": 3950
},
{
"epoch": 0.96,
"grad_norm": 0.23476284742355347,
"learning_rate": 4.7276068476448097e-05,
"loss": 0.644,
"step": 3960
},
{
"epoch": 0.96,
"grad_norm": 0.25303345918655396,
"learning_rate": 4.725793411288242e-05,
"loss": 0.649,
"step": 3970
},
{
"epoch": 0.96,
"grad_norm": 0.2615244686603546,
"learning_rate": 4.723974308716361e-05,
"loss": 0.6158,
"step": 3980
},
{
"epoch": 0.96,
"grad_norm": 0.2798680365085602,
"learning_rate": 4.722149544560067e-05,
"loss": 0.6149,
"step": 3990
},
{
"epoch": 0.97,
"grad_norm": 0.19780561327934265,
"learning_rate": 4.720319123464672e-05,
"loss": 0.6167,
"step": 4000
},
{
"epoch": 0.97,
"eval_loss": 0.6494756937026978,
"eval_runtime": 67.2363,
"eval_samples_per_second": 29.746,
"eval_steps_per_second": 0.937,
"step": 4000
}
],
"logging_steps": 10,
"max_steps": 20690,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 8.648479843115074e+18,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}