tsec_vit_model / trainer_state.json
suredream's picture
End of training
45a67ae verified
{
"best_metric": 0.8894595801697186,
"best_model_checkpoint": "tsec_vit_model/checkpoint-2520",
"epoch": 10.0,
"eval_steps": 500,
"global_step": 2800,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03571428571428571,
"grad_norm": 0.632759690284729,
"learning_rate": 1.7857142857142857e-06,
"loss": 0.7005,
"step": 10
},
{
"epoch": 0.07142857142857142,
"grad_norm": 0.6090130805969238,
"learning_rate": 3.5714285714285714e-06,
"loss": 0.6888,
"step": 20
},
{
"epoch": 0.10714285714285714,
"grad_norm": 0.5027034878730774,
"learning_rate": 5.357142857142857e-06,
"loss": 0.6841,
"step": 30
},
{
"epoch": 0.14285714285714285,
"grad_norm": 0.7158761620521545,
"learning_rate": 7.142857142857143e-06,
"loss": 0.6711,
"step": 40
},
{
"epoch": 0.17857142857142858,
"grad_norm": 1.0677605867385864,
"learning_rate": 8.92857142857143e-06,
"loss": 0.6439,
"step": 50
},
{
"epoch": 0.21428571428571427,
"grad_norm": 0.716444194316864,
"learning_rate": 1.0714285714285714e-05,
"loss": 0.6388,
"step": 60
},
{
"epoch": 0.25,
"grad_norm": 0.9646293520927429,
"learning_rate": 1.25e-05,
"loss": 0.5942,
"step": 70
},
{
"epoch": 0.2857142857142857,
"grad_norm": 1.5171552896499634,
"learning_rate": 1.4285714285714285e-05,
"loss": 0.587,
"step": 80
},
{
"epoch": 0.32142857142857145,
"grad_norm": 1.2230862379074097,
"learning_rate": 1.6071428571428572e-05,
"loss": 0.5882,
"step": 90
},
{
"epoch": 0.35714285714285715,
"grad_norm": 1.0726110935211182,
"learning_rate": 1.785714285714286e-05,
"loss": 0.5506,
"step": 100
},
{
"epoch": 0.39285714285714285,
"grad_norm": 2.488288640975952,
"learning_rate": 1.9642857142857145e-05,
"loss": 0.5487,
"step": 110
},
{
"epoch": 0.42857142857142855,
"grad_norm": 2.186500310897827,
"learning_rate": 2.1428571428571428e-05,
"loss": 0.5138,
"step": 120
},
{
"epoch": 0.4642857142857143,
"grad_norm": 2.18605637550354,
"learning_rate": 2.3214285714285715e-05,
"loss": 0.5203,
"step": 130
},
{
"epoch": 0.5,
"grad_norm": 1.5495938062667847,
"learning_rate": 2.5e-05,
"loss": 0.5006,
"step": 140
},
{
"epoch": 0.5357142857142857,
"grad_norm": 1.3939014673233032,
"learning_rate": 2.6785714285714288e-05,
"loss": 0.5171,
"step": 150
},
{
"epoch": 0.5714285714285714,
"grad_norm": 1.4516574144363403,
"learning_rate": 2.857142857142857e-05,
"loss": 0.4882,
"step": 160
},
{
"epoch": 0.6071428571428571,
"grad_norm": 1.5352245569229126,
"learning_rate": 3.0357142857142857e-05,
"loss": 0.4809,
"step": 170
},
{
"epoch": 0.6428571428571429,
"grad_norm": 2.6514480113983154,
"learning_rate": 3.2142857142857144e-05,
"loss": 0.458,
"step": 180
},
{
"epoch": 0.6785714285714286,
"grad_norm": 2.1452901363372803,
"learning_rate": 3.392857142857143e-05,
"loss": 0.478,
"step": 190
},
{
"epoch": 0.7142857142857143,
"grad_norm": 1.5536608695983887,
"learning_rate": 3.571428571428572e-05,
"loss": 0.4851,
"step": 200
},
{
"epoch": 0.75,
"grad_norm": 1.7489731311798096,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.4608,
"step": 210
},
{
"epoch": 0.7857142857142857,
"grad_norm": 1.6109980344772339,
"learning_rate": 3.928571428571429e-05,
"loss": 0.4611,
"step": 220
},
{
"epoch": 0.8214285714285714,
"grad_norm": 2.391719102859497,
"learning_rate": 4.107142857142857e-05,
"loss": 0.3998,
"step": 230
},
{
"epoch": 0.8571428571428571,
"grad_norm": 3.6933391094207764,
"learning_rate": 4.2857142857142856e-05,
"loss": 0.4624,
"step": 240
},
{
"epoch": 0.8928571428571429,
"grad_norm": 2.0588369369506836,
"learning_rate": 4.464285714285715e-05,
"loss": 0.4437,
"step": 250
},
{
"epoch": 0.9285714285714286,
"grad_norm": 2.831156015396118,
"learning_rate": 4.642857142857143e-05,
"loss": 0.4652,
"step": 260
},
{
"epoch": 0.9642857142857143,
"grad_norm": 2.0229790210723877,
"learning_rate": 4.8214285714285716e-05,
"loss": 0.4037,
"step": 270
},
{
"epoch": 1.0,
"grad_norm": 2.1834700107574463,
"learning_rate": 5e-05,
"loss": 0.4387,
"step": 280
},
{
"epoch": 1.0,
"eval_accuracy": 0.8150960250111657,
"eval_loss": 0.4179099500179291,
"eval_runtime": 110.4006,
"eval_samples_per_second": 40.561,
"eval_steps_per_second": 2.536,
"step": 280
},
{
"epoch": 1.0357142857142858,
"grad_norm": 1.7746831178665161,
"learning_rate": 4.9801587301587306e-05,
"loss": 0.4069,
"step": 290
},
{
"epoch": 1.0714285714285714,
"grad_norm": 2.6947922706604004,
"learning_rate": 4.960317460317461e-05,
"loss": 0.4334,
"step": 300
},
{
"epoch": 1.1071428571428572,
"grad_norm": 1.4933383464813232,
"learning_rate": 4.940476190476191e-05,
"loss": 0.4409,
"step": 310
},
{
"epoch": 1.1428571428571428,
"grad_norm": 1.8830089569091797,
"learning_rate": 4.9206349206349204e-05,
"loss": 0.4065,
"step": 320
},
{
"epoch": 1.1785714285714286,
"grad_norm": 1.978843092918396,
"learning_rate": 4.900793650793651e-05,
"loss": 0.4273,
"step": 330
},
{
"epoch": 1.2142857142857142,
"grad_norm": 1.2928149700164795,
"learning_rate": 4.880952380952381e-05,
"loss": 0.4376,
"step": 340
},
{
"epoch": 1.25,
"grad_norm": 1.5401108264923096,
"learning_rate": 4.8611111111111115e-05,
"loss": 0.4061,
"step": 350
},
{
"epoch": 1.2857142857142856,
"grad_norm": 2.296520709991455,
"learning_rate": 4.841269841269841e-05,
"loss": 0.433,
"step": 360
},
{
"epoch": 1.3214285714285714,
"grad_norm": 1.2640879154205322,
"learning_rate": 4.8214285714285716e-05,
"loss": 0.4142,
"step": 370
},
{
"epoch": 1.3571428571428572,
"grad_norm": 1.3309462070465088,
"learning_rate": 4.801587301587302e-05,
"loss": 0.4222,
"step": 380
},
{
"epoch": 1.3928571428571428,
"grad_norm": 1.8462568521499634,
"learning_rate": 4.781746031746032e-05,
"loss": 0.4483,
"step": 390
},
{
"epoch": 1.4285714285714286,
"grad_norm": 1.074401617050171,
"learning_rate": 4.761904761904762e-05,
"loss": 0.3762,
"step": 400
},
{
"epoch": 1.4642857142857144,
"grad_norm": 1.5691473484039307,
"learning_rate": 4.7420634920634924e-05,
"loss": 0.415,
"step": 410
},
{
"epoch": 1.5,
"grad_norm": 1.2222189903259277,
"learning_rate": 4.722222222222222e-05,
"loss": 0.3862,
"step": 420
},
{
"epoch": 1.5357142857142856,
"grad_norm": 1.4210410118103027,
"learning_rate": 4.7023809523809525e-05,
"loss": 0.352,
"step": 430
},
{
"epoch": 1.5714285714285714,
"grad_norm": 1.632934331893921,
"learning_rate": 4.682539682539683e-05,
"loss": 0.3873,
"step": 440
},
{
"epoch": 1.6071428571428572,
"grad_norm": 1.3977652788162231,
"learning_rate": 4.662698412698413e-05,
"loss": 0.3763,
"step": 450
},
{
"epoch": 1.6428571428571428,
"grad_norm": 2.74967622756958,
"learning_rate": 4.642857142857143e-05,
"loss": 0.3823,
"step": 460
},
{
"epoch": 1.6785714285714286,
"grad_norm": 1.1579408645629883,
"learning_rate": 4.623015873015873e-05,
"loss": 0.3967,
"step": 470
},
{
"epoch": 1.7142857142857144,
"grad_norm": 2.7291440963745117,
"learning_rate": 4.603174603174603e-05,
"loss": 0.4201,
"step": 480
},
{
"epoch": 1.75,
"grad_norm": 1.0126214027404785,
"learning_rate": 4.5833333333333334e-05,
"loss": 0.3598,
"step": 490
},
{
"epoch": 1.7857142857142856,
"grad_norm": 1.6705009937286377,
"learning_rate": 4.563492063492064e-05,
"loss": 0.366,
"step": 500
},
{
"epoch": 1.8214285714285714,
"grad_norm": 1.6749546527862549,
"learning_rate": 4.543650793650794e-05,
"loss": 0.3854,
"step": 510
},
{
"epoch": 1.8571428571428572,
"grad_norm": 1.0892480611801147,
"learning_rate": 4.523809523809524e-05,
"loss": 0.3563,
"step": 520
},
{
"epoch": 1.8928571428571428,
"grad_norm": 1.4169703722000122,
"learning_rate": 4.503968253968254e-05,
"loss": 0.3903,
"step": 530
},
{
"epoch": 1.9285714285714286,
"grad_norm": 1.393864631652832,
"learning_rate": 4.4841269841269846e-05,
"loss": 0.3915,
"step": 540
},
{
"epoch": 1.9642857142857144,
"grad_norm": 1.6008644104003906,
"learning_rate": 4.464285714285715e-05,
"loss": 0.3665,
"step": 550
},
{
"epoch": 2.0,
"grad_norm": 1.0977710485458374,
"learning_rate": 4.4444444444444447e-05,
"loss": 0.4239,
"step": 560
},
{
"epoch": 2.0,
"eval_accuracy": 0.8398838767306833,
"eval_loss": 0.3610800504684448,
"eval_runtime": 114.7581,
"eval_samples_per_second": 39.021,
"eval_steps_per_second": 2.44,
"step": 560
},
{
"epoch": 2.0357142857142856,
"grad_norm": 1.4717609882354736,
"learning_rate": 4.4246031746031744e-05,
"loss": 0.3609,
"step": 570
},
{
"epoch": 2.0714285714285716,
"grad_norm": 1.0943901538848877,
"learning_rate": 4.404761904761905e-05,
"loss": 0.3379,
"step": 580
},
{
"epoch": 2.107142857142857,
"grad_norm": 1.8704146146774292,
"learning_rate": 4.384920634920635e-05,
"loss": 0.364,
"step": 590
},
{
"epoch": 2.142857142857143,
"grad_norm": 1.2977815866470337,
"learning_rate": 4.3650793650793655e-05,
"loss": 0.3834,
"step": 600
},
{
"epoch": 2.1785714285714284,
"grad_norm": 1.6529649496078491,
"learning_rate": 4.345238095238096e-05,
"loss": 0.3688,
"step": 610
},
{
"epoch": 2.2142857142857144,
"grad_norm": 1.1914763450622559,
"learning_rate": 4.3253968253968256e-05,
"loss": 0.3548,
"step": 620
},
{
"epoch": 2.25,
"grad_norm": 1.70374596118927,
"learning_rate": 4.305555555555556e-05,
"loss": 0.3323,
"step": 630
},
{
"epoch": 2.2857142857142856,
"grad_norm": 1.1331793069839478,
"learning_rate": 4.2857142857142856e-05,
"loss": 0.3494,
"step": 640
},
{
"epoch": 2.3214285714285716,
"grad_norm": 1.3151099681854248,
"learning_rate": 4.265873015873016e-05,
"loss": 0.3177,
"step": 650
},
{
"epoch": 2.357142857142857,
"grad_norm": 1.7670910358428955,
"learning_rate": 4.2460317460317464e-05,
"loss": 0.3145,
"step": 660
},
{
"epoch": 2.392857142857143,
"grad_norm": 1.441829800605774,
"learning_rate": 4.226190476190476e-05,
"loss": 0.2968,
"step": 670
},
{
"epoch": 2.4285714285714284,
"grad_norm": 1.5992521047592163,
"learning_rate": 4.2063492063492065e-05,
"loss": 0.353,
"step": 680
},
{
"epoch": 2.4642857142857144,
"grad_norm": 1.076711654663086,
"learning_rate": 4.186507936507937e-05,
"loss": 0.3461,
"step": 690
},
{
"epoch": 2.5,
"grad_norm": 1.0764169692993164,
"learning_rate": 4.166666666666667e-05,
"loss": 0.3345,
"step": 700
},
{
"epoch": 2.5357142857142856,
"grad_norm": 1.3735431432724,
"learning_rate": 4.1468253968253976e-05,
"loss": 0.3612,
"step": 710
},
{
"epoch": 2.571428571428571,
"grad_norm": 1.4703452587127686,
"learning_rate": 4.126984126984127e-05,
"loss": 0.3444,
"step": 720
},
{
"epoch": 2.607142857142857,
"grad_norm": 1.451257348060608,
"learning_rate": 4.107142857142857e-05,
"loss": 0.3287,
"step": 730
},
{
"epoch": 2.642857142857143,
"grad_norm": 1.8615336418151855,
"learning_rate": 4.0873015873015874e-05,
"loss": 0.3419,
"step": 740
},
{
"epoch": 2.678571428571429,
"grad_norm": 2.925323963165283,
"learning_rate": 4.067460317460318e-05,
"loss": 0.3217,
"step": 750
},
{
"epoch": 2.7142857142857144,
"grad_norm": 1.7260726690292358,
"learning_rate": 4.047619047619048e-05,
"loss": 0.3584,
"step": 760
},
{
"epoch": 2.75,
"grad_norm": 1.4659258127212524,
"learning_rate": 4.027777777777778e-05,
"loss": 0.3821,
"step": 770
},
{
"epoch": 2.7857142857142856,
"grad_norm": 1.7859466075897217,
"learning_rate": 4.007936507936508e-05,
"loss": 0.3595,
"step": 780
},
{
"epoch": 2.821428571428571,
"grad_norm": 1.2320784330368042,
"learning_rate": 3.9880952380952386e-05,
"loss": 0.3764,
"step": 790
},
{
"epoch": 2.857142857142857,
"grad_norm": 1.4097161293029785,
"learning_rate": 3.968253968253968e-05,
"loss": 0.3282,
"step": 800
},
{
"epoch": 2.892857142857143,
"grad_norm": 1.185285210609436,
"learning_rate": 3.9484126984126986e-05,
"loss": 0.3214,
"step": 810
},
{
"epoch": 2.928571428571429,
"grad_norm": 1.3706032037734985,
"learning_rate": 3.928571428571429e-05,
"loss": 0.3092,
"step": 820
},
{
"epoch": 2.9642857142857144,
"grad_norm": 2.1342084407806396,
"learning_rate": 3.908730158730159e-05,
"loss": 0.3274,
"step": 830
},
{
"epoch": 3.0,
"grad_norm": 1.438025712966919,
"learning_rate": 3.888888888888889e-05,
"loss": 0.3148,
"step": 840
},
{
"epoch": 3.0,
"eval_accuracy": 0.8599821348816435,
"eval_loss": 0.31560027599334717,
"eval_runtime": 116.0276,
"eval_samples_per_second": 38.594,
"eval_steps_per_second": 2.413,
"step": 840
},
{
"epoch": 3.0357142857142856,
"grad_norm": 1.6471174955368042,
"learning_rate": 3.8690476190476195e-05,
"loss": 0.3198,
"step": 850
},
{
"epoch": 3.0714285714285716,
"grad_norm": 1.5575629472732544,
"learning_rate": 3.84920634920635e-05,
"loss": 0.3127,
"step": 860
},
{
"epoch": 3.107142857142857,
"grad_norm": 1.1578933000564575,
"learning_rate": 3.8293650793650795e-05,
"loss": 0.3416,
"step": 870
},
{
"epoch": 3.142857142857143,
"grad_norm": 1.9167553186416626,
"learning_rate": 3.809523809523809e-05,
"loss": 0.3626,
"step": 880
},
{
"epoch": 3.1785714285714284,
"grad_norm": 1.038451910018921,
"learning_rate": 3.7896825396825396e-05,
"loss": 0.336,
"step": 890
},
{
"epoch": 3.2142857142857144,
"grad_norm": 1.2739081382751465,
"learning_rate": 3.76984126984127e-05,
"loss": 0.2838,
"step": 900
},
{
"epoch": 3.25,
"grad_norm": 1.2061238288879395,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.3099,
"step": 910
},
{
"epoch": 3.2857142857142856,
"grad_norm": 1.5365681648254395,
"learning_rate": 3.730158730158731e-05,
"loss": 0.3163,
"step": 920
},
{
"epoch": 3.3214285714285716,
"grad_norm": 2.33838152885437,
"learning_rate": 3.7103174603174604e-05,
"loss": 0.3109,
"step": 930
},
{
"epoch": 3.357142857142857,
"grad_norm": 1.2280935049057007,
"learning_rate": 3.690476190476191e-05,
"loss": 0.3104,
"step": 940
},
{
"epoch": 3.392857142857143,
"grad_norm": 1.2506486177444458,
"learning_rate": 3.6706349206349205e-05,
"loss": 0.339,
"step": 950
},
{
"epoch": 3.4285714285714284,
"grad_norm": 1.0430885553359985,
"learning_rate": 3.650793650793651e-05,
"loss": 0.3015,
"step": 960
},
{
"epoch": 3.4642857142857144,
"grad_norm": 1.5323641300201416,
"learning_rate": 3.630952380952381e-05,
"loss": 0.3408,
"step": 970
},
{
"epoch": 3.5,
"grad_norm": 1.0826184749603271,
"learning_rate": 3.611111111111111e-05,
"loss": 0.3061,
"step": 980
},
{
"epoch": 3.5357142857142856,
"grad_norm": 2.019437313079834,
"learning_rate": 3.591269841269841e-05,
"loss": 0.3465,
"step": 990
},
{
"epoch": 3.571428571428571,
"grad_norm": 1.5877951383590698,
"learning_rate": 3.571428571428572e-05,
"loss": 0.3023,
"step": 1000
},
{
"epoch": 3.607142857142857,
"grad_norm": 1.5832830667495728,
"learning_rate": 3.551587301587302e-05,
"loss": 0.262,
"step": 1010
},
{
"epoch": 3.642857142857143,
"grad_norm": 1.5758150815963745,
"learning_rate": 3.5317460317460324e-05,
"loss": 0.3337,
"step": 1020
},
{
"epoch": 3.678571428571429,
"grad_norm": 1.255638837814331,
"learning_rate": 3.511904761904762e-05,
"loss": 0.2857,
"step": 1030
},
{
"epoch": 3.7142857142857144,
"grad_norm": 1.5360593795776367,
"learning_rate": 3.492063492063492e-05,
"loss": 0.2964,
"step": 1040
},
{
"epoch": 3.75,
"grad_norm": 1.0826270580291748,
"learning_rate": 3.472222222222222e-05,
"loss": 0.2872,
"step": 1050
},
{
"epoch": 3.7857142857142856,
"grad_norm": 1.6883878707885742,
"learning_rate": 3.4523809523809526e-05,
"loss": 0.3016,
"step": 1060
},
{
"epoch": 3.821428571428571,
"grad_norm": 1.8018616437911987,
"learning_rate": 3.432539682539683e-05,
"loss": 0.3115,
"step": 1070
},
{
"epoch": 3.857142857142857,
"grad_norm": 1.9364327192306519,
"learning_rate": 3.412698412698413e-05,
"loss": 0.2986,
"step": 1080
},
{
"epoch": 3.892857142857143,
"grad_norm": 1.6606508493423462,
"learning_rate": 3.392857142857143e-05,
"loss": 0.3017,
"step": 1090
},
{
"epoch": 3.928571428571429,
"grad_norm": 2.1849498748779297,
"learning_rate": 3.3730158730158734e-05,
"loss": 0.2779,
"step": 1100
},
{
"epoch": 3.9642857142857144,
"grad_norm": 1.5579859018325806,
"learning_rate": 3.353174603174603e-05,
"loss": 0.3199,
"step": 1110
},
{
"epoch": 4.0,
"grad_norm": 2.6287128925323486,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.2988,
"step": 1120
},
{
"epoch": 4.0,
"eval_accuracy": 0.8729343456900402,
"eval_loss": 0.30018478631973267,
"eval_runtime": 115.4434,
"eval_samples_per_second": 38.79,
"eval_steps_per_second": 2.425,
"step": 1120
},
{
"epoch": 4.035714285714286,
"grad_norm": 1.5949360132217407,
"learning_rate": 3.313492063492064e-05,
"loss": 0.305,
"step": 1130
},
{
"epoch": 4.071428571428571,
"grad_norm": 1.678801417350769,
"learning_rate": 3.2936507936507936e-05,
"loss": 0.2677,
"step": 1140
},
{
"epoch": 4.107142857142857,
"grad_norm": 1.6015031337738037,
"learning_rate": 3.273809523809524e-05,
"loss": 0.3167,
"step": 1150
},
{
"epoch": 4.142857142857143,
"grad_norm": 1.525894045829773,
"learning_rate": 3.253968253968254e-05,
"loss": 0.2745,
"step": 1160
},
{
"epoch": 4.178571428571429,
"grad_norm": 1.2955286502838135,
"learning_rate": 3.234126984126985e-05,
"loss": 0.2925,
"step": 1170
},
{
"epoch": 4.214285714285714,
"grad_norm": 2.1041815280914307,
"learning_rate": 3.2142857142857144e-05,
"loss": 0.2782,
"step": 1180
},
{
"epoch": 4.25,
"grad_norm": 1.7325941324234009,
"learning_rate": 3.194444444444444e-05,
"loss": 0.302,
"step": 1190
},
{
"epoch": 4.285714285714286,
"grad_norm": 1.9786794185638428,
"learning_rate": 3.1746031746031745e-05,
"loss": 0.2864,
"step": 1200
},
{
"epoch": 4.321428571428571,
"grad_norm": 1.3869291543960571,
"learning_rate": 3.154761904761905e-05,
"loss": 0.2834,
"step": 1210
},
{
"epoch": 4.357142857142857,
"grad_norm": 2.1556620597839355,
"learning_rate": 3.134920634920635e-05,
"loss": 0.3011,
"step": 1220
},
{
"epoch": 4.392857142857143,
"grad_norm": 1.6058976650238037,
"learning_rate": 3.1150793650793656e-05,
"loss": 0.2893,
"step": 1230
},
{
"epoch": 4.428571428571429,
"grad_norm": 1.9528658390045166,
"learning_rate": 3.095238095238095e-05,
"loss": 0.2725,
"step": 1240
},
{
"epoch": 4.464285714285714,
"grad_norm": 1.5343618392944336,
"learning_rate": 3.075396825396826e-05,
"loss": 0.2578,
"step": 1250
},
{
"epoch": 4.5,
"grad_norm": 1.0154541730880737,
"learning_rate": 3.055555555555556e-05,
"loss": 0.2897,
"step": 1260
},
{
"epoch": 4.535714285714286,
"grad_norm": 2.355865001678467,
"learning_rate": 3.0357142857142857e-05,
"loss": 0.2658,
"step": 1270
},
{
"epoch": 4.571428571428571,
"grad_norm": 1.24091374874115,
"learning_rate": 3.0158730158730158e-05,
"loss": 0.3198,
"step": 1280
},
{
"epoch": 4.607142857142857,
"grad_norm": 1.0385922193527222,
"learning_rate": 2.996031746031746e-05,
"loss": 0.285,
"step": 1290
},
{
"epoch": 4.642857142857143,
"grad_norm": 1.032619833946228,
"learning_rate": 2.9761904761904762e-05,
"loss": 0.2741,
"step": 1300
},
{
"epoch": 4.678571428571429,
"grad_norm": 1.587849736213684,
"learning_rate": 2.9563492063492066e-05,
"loss": 0.3093,
"step": 1310
},
{
"epoch": 4.714285714285714,
"grad_norm": 1.2873854637145996,
"learning_rate": 2.9365079365079366e-05,
"loss": 0.299,
"step": 1320
},
{
"epoch": 4.75,
"grad_norm": 1.6315929889678955,
"learning_rate": 2.916666666666667e-05,
"loss": 0.2776,
"step": 1330
},
{
"epoch": 4.785714285714286,
"grad_norm": 1.8490331172943115,
"learning_rate": 2.8968253968253974e-05,
"loss": 0.2613,
"step": 1340
},
{
"epoch": 4.821428571428571,
"grad_norm": 1.8352344036102295,
"learning_rate": 2.876984126984127e-05,
"loss": 0.281,
"step": 1350
},
{
"epoch": 4.857142857142857,
"grad_norm": 1.0744656324386597,
"learning_rate": 2.857142857142857e-05,
"loss": 0.2622,
"step": 1360
},
{
"epoch": 4.892857142857143,
"grad_norm": 1.2477718591690063,
"learning_rate": 2.8373015873015875e-05,
"loss": 0.2997,
"step": 1370
},
{
"epoch": 4.928571428571429,
"grad_norm": 1.5114529132843018,
"learning_rate": 2.8174603174603175e-05,
"loss": 0.2827,
"step": 1380
},
{
"epoch": 4.964285714285714,
"grad_norm": 1.3266760110855103,
"learning_rate": 2.797619047619048e-05,
"loss": 0.283,
"step": 1390
},
{
"epoch": 5.0,
"grad_norm": 2.2647006511688232,
"learning_rate": 2.777777777777778e-05,
"loss": 0.2498,
"step": 1400
},
{
"epoch": 5.0,
"eval_accuracy": 0.8693613220187584,
"eval_loss": 0.30874186754226685,
"eval_runtime": 116.4377,
"eval_samples_per_second": 38.458,
"eval_steps_per_second": 2.405,
"step": 1400
},
{
"epoch": 5.035714285714286,
"grad_norm": 1.6704837083816528,
"learning_rate": 2.7579365079365083e-05,
"loss": 0.2973,
"step": 1410
},
{
"epoch": 5.071428571428571,
"grad_norm": 1.9291003942489624,
"learning_rate": 2.7380952380952383e-05,
"loss": 0.2756,
"step": 1420
},
{
"epoch": 5.107142857142857,
"grad_norm": 2.44765043258667,
"learning_rate": 2.718253968253968e-05,
"loss": 0.2575,
"step": 1430
},
{
"epoch": 5.142857142857143,
"grad_norm": 2.0583574771881104,
"learning_rate": 2.6984126984126984e-05,
"loss": 0.256,
"step": 1440
},
{
"epoch": 5.178571428571429,
"grad_norm": 1.8882899284362793,
"learning_rate": 2.6785714285714288e-05,
"loss": 0.299,
"step": 1450
},
{
"epoch": 5.214285714285714,
"grad_norm": 1.5178027153015137,
"learning_rate": 2.6587301587301588e-05,
"loss": 0.3209,
"step": 1460
},
{
"epoch": 5.25,
"grad_norm": 1.4342873096466064,
"learning_rate": 2.6388888888888892e-05,
"loss": 0.2828,
"step": 1470
},
{
"epoch": 5.285714285714286,
"grad_norm": 1.3157316446304321,
"learning_rate": 2.6190476190476192e-05,
"loss": 0.2655,
"step": 1480
},
{
"epoch": 5.321428571428571,
"grad_norm": 1.0939053297042847,
"learning_rate": 2.5992063492063496e-05,
"loss": 0.278,
"step": 1490
},
{
"epoch": 5.357142857142857,
"grad_norm": 1.9465833902359009,
"learning_rate": 2.5793650793650796e-05,
"loss": 0.2652,
"step": 1500
},
{
"epoch": 5.392857142857143,
"grad_norm": 1.5688917636871338,
"learning_rate": 2.5595238095238093e-05,
"loss": 0.2622,
"step": 1510
},
{
"epoch": 5.428571428571429,
"grad_norm": 1.3724976778030396,
"learning_rate": 2.5396825396825397e-05,
"loss": 0.2524,
"step": 1520
},
{
"epoch": 5.464285714285714,
"grad_norm": 1.160733699798584,
"learning_rate": 2.5198412698412697e-05,
"loss": 0.2868,
"step": 1530
},
{
"epoch": 5.5,
"grad_norm": 1.2799668312072754,
"learning_rate": 2.5e-05,
"loss": 0.2713,
"step": 1540
},
{
"epoch": 5.535714285714286,
"grad_norm": 1.9062435626983643,
"learning_rate": 2.4801587301587305e-05,
"loss": 0.2527,
"step": 1550
},
{
"epoch": 5.571428571428571,
"grad_norm": 2.5998220443725586,
"learning_rate": 2.4603174603174602e-05,
"loss": 0.2668,
"step": 1560
},
{
"epoch": 5.607142857142857,
"grad_norm": 2.346576452255249,
"learning_rate": 2.4404761904761906e-05,
"loss": 0.2557,
"step": 1570
},
{
"epoch": 5.642857142857143,
"grad_norm": 1.9609774351119995,
"learning_rate": 2.4206349206349206e-05,
"loss": 0.2706,
"step": 1580
},
{
"epoch": 5.678571428571429,
"grad_norm": 1.8994617462158203,
"learning_rate": 2.400793650793651e-05,
"loss": 0.2878,
"step": 1590
},
{
"epoch": 5.714285714285714,
"grad_norm": 1.677849531173706,
"learning_rate": 2.380952380952381e-05,
"loss": 0.2761,
"step": 1600
},
{
"epoch": 5.75,
"grad_norm": 1.6345635652542114,
"learning_rate": 2.361111111111111e-05,
"loss": 0.2231,
"step": 1610
},
{
"epoch": 5.785714285714286,
"grad_norm": 1.6198554039001465,
"learning_rate": 2.3412698412698414e-05,
"loss": 0.2612,
"step": 1620
},
{
"epoch": 5.821428571428571,
"grad_norm": 2.052764415740967,
"learning_rate": 2.3214285714285715e-05,
"loss": 0.2738,
"step": 1630
},
{
"epoch": 5.857142857142857,
"grad_norm": 1.0596450567245483,
"learning_rate": 2.3015873015873015e-05,
"loss": 0.2597,
"step": 1640
},
{
"epoch": 5.892857142857143,
"grad_norm": 1.9113609790802002,
"learning_rate": 2.281746031746032e-05,
"loss": 0.2894,
"step": 1650
},
{
"epoch": 5.928571428571429,
"grad_norm": 1.7055829763412476,
"learning_rate": 2.261904761904762e-05,
"loss": 0.226,
"step": 1660
},
{
"epoch": 5.964285714285714,
"grad_norm": 0.969464898109436,
"learning_rate": 2.2420634920634923e-05,
"loss": 0.2977,
"step": 1670
},
{
"epoch": 6.0,
"grad_norm": 3.1485939025878906,
"learning_rate": 2.2222222222222223e-05,
"loss": 0.3028,
"step": 1680
},
{
"epoch": 6.0,
"eval_accuracy": 0.8715944618133095,
"eval_loss": 0.29658079147338867,
"eval_runtime": 113.6842,
"eval_samples_per_second": 39.39,
"eval_steps_per_second": 2.463,
"step": 1680
},
{
"epoch": 6.035714285714286,
"grad_norm": 1.4179354906082153,
"learning_rate": 2.2023809523809524e-05,
"loss": 0.2681,
"step": 1690
},
{
"epoch": 6.071428571428571,
"grad_norm": 2.099681854248047,
"learning_rate": 2.1825396825396827e-05,
"loss": 0.2441,
"step": 1700
},
{
"epoch": 6.107142857142857,
"grad_norm": 1.5406619310379028,
"learning_rate": 2.1626984126984128e-05,
"loss": 0.2707,
"step": 1710
},
{
"epoch": 6.142857142857143,
"grad_norm": 2.4518625736236572,
"learning_rate": 2.1428571428571428e-05,
"loss": 0.2942,
"step": 1720
},
{
"epoch": 6.178571428571429,
"grad_norm": 1.1450200080871582,
"learning_rate": 2.1230158730158732e-05,
"loss": 0.2336,
"step": 1730
},
{
"epoch": 6.214285714285714,
"grad_norm": 1.4170438051223755,
"learning_rate": 2.1031746031746032e-05,
"loss": 0.2857,
"step": 1740
},
{
"epoch": 6.25,
"grad_norm": 1.2231560945510864,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.2507,
"step": 1750
},
{
"epoch": 6.285714285714286,
"grad_norm": 1.4597039222717285,
"learning_rate": 2.0634920634920636e-05,
"loss": 0.2688,
"step": 1760
},
{
"epoch": 6.321428571428571,
"grad_norm": 1.9395854473114014,
"learning_rate": 2.0436507936507937e-05,
"loss": 0.2715,
"step": 1770
},
{
"epoch": 6.357142857142857,
"grad_norm": 0.9649907946586609,
"learning_rate": 2.023809523809524e-05,
"loss": 0.2508,
"step": 1780
},
{
"epoch": 6.392857142857143,
"grad_norm": 1.3842933177947998,
"learning_rate": 2.003968253968254e-05,
"loss": 0.2818,
"step": 1790
},
{
"epoch": 6.428571428571429,
"grad_norm": 1.7550239562988281,
"learning_rate": 1.984126984126984e-05,
"loss": 0.2553,
"step": 1800
},
{
"epoch": 6.464285714285714,
"grad_norm": 1.04232919216156,
"learning_rate": 1.9642857142857145e-05,
"loss": 0.2303,
"step": 1810
},
{
"epoch": 6.5,
"grad_norm": 2.306325674057007,
"learning_rate": 1.9444444444444445e-05,
"loss": 0.257,
"step": 1820
},
{
"epoch": 6.535714285714286,
"grad_norm": 1.704186201095581,
"learning_rate": 1.924603174603175e-05,
"loss": 0.2844,
"step": 1830
},
{
"epoch": 6.571428571428571,
"grad_norm": 1.3930236101150513,
"learning_rate": 1.9047619047619046e-05,
"loss": 0.2784,
"step": 1840
},
{
"epoch": 6.607142857142857,
"grad_norm": 1.5494309663772583,
"learning_rate": 1.884920634920635e-05,
"loss": 0.2735,
"step": 1850
},
{
"epoch": 6.642857142857143,
"grad_norm": 1.5971505641937256,
"learning_rate": 1.8650793650793654e-05,
"loss": 0.2466,
"step": 1860
},
{
"epoch": 6.678571428571429,
"grad_norm": 1.8419290781021118,
"learning_rate": 1.8452380952380954e-05,
"loss": 0.2476,
"step": 1870
},
{
"epoch": 6.714285714285714,
"grad_norm": 1.3679064512252808,
"learning_rate": 1.8253968253968254e-05,
"loss": 0.2277,
"step": 1880
},
{
"epoch": 6.75,
"grad_norm": 1.1586477756500244,
"learning_rate": 1.8055555555555555e-05,
"loss": 0.2266,
"step": 1890
},
{
"epoch": 6.785714285714286,
"grad_norm": 1.9979654550552368,
"learning_rate": 1.785714285714286e-05,
"loss": 0.2747,
"step": 1900
},
{
"epoch": 6.821428571428571,
"grad_norm": 1.4875764846801758,
"learning_rate": 1.7658730158730162e-05,
"loss": 0.2666,
"step": 1910
},
{
"epoch": 6.857142857142857,
"grad_norm": 1.6196482181549072,
"learning_rate": 1.746031746031746e-05,
"loss": 0.2532,
"step": 1920
},
{
"epoch": 6.892857142857143,
"grad_norm": 2.7135918140411377,
"learning_rate": 1.7261904761904763e-05,
"loss": 0.2265,
"step": 1930
},
{
"epoch": 6.928571428571429,
"grad_norm": 1.423257827758789,
"learning_rate": 1.7063492063492063e-05,
"loss": 0.2621,
"step": 1940
},
{
"epoch": 6.964285714285714,
"grad_norm": 2.1509084701538086,
"learning_rate": 1.6865079365079367e-05,
"loss": 0.2636,
"step": 1950
},
{
"epoch": 7.0,
"grad_norm": 1.562657117843628,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.2179,
"step": 1960
},
{
"epoch": 7.0,
"eval_accuracy": 0.8807503349709692,
"eval_loss": 0.2742190361022949,
"eval_runtime": 113.8381,
"eval_samples_per_second": 39.337,
"eval_steps_per_second": 2.46,
"step": 1960
},
{
"epoch": 7.035714285714286,
"grad_norm": 1.4138425588607788,
"learning_rate": 1.6468253968253968e-05,
"loss": 0.2251,
"step": 1970
},
{
"epoch": 7.071428571428571,
"grad_norm": 2.6958255767822266,
"learning_rate": 1.626984126984127e-05,
"loss": 0.2521,
"step": 1980
},
{
"epoch": 7.107142857142857,
"grad_norm": 2.014803647994995,
"learning_rate": 1.6071428571428572e-05,
"loss": 0.216,
"step": 1990
},
{
"epoch": 7.142857142857143,
"grad_norm": 2.1656410694122314,
"learning_rate": 1.5873015873015872e-05,
"loss": 0.2085,
"step": 2000
},
{
"epoch": 7.178571428571429,
"grad_norm": 2.207980155944824,
"learning_rate": 1.5674603174603176e-05,
"loss": 0.263,
"step": 2010
},
{
"epoch": 7.214285714285714,
"grad_norm": 1.978162169456482,
"learning_rate": 1.5476190476190476e-05,
"loss": 0.2478,
"step": 2020
},
{
"epoch": 7.25,
"grad_norm": 1.2705848217010498,
"learning_rate": 1.527777777777778e-05,
"loss": 0.2338,
"step": 2030
},
{
"epoch": 7.285714285714286,
"grad_norm": 1.6287261247634888,
"learning_rate": 1.5079365079365079e-05,
"loss": 0.2063,
"step": 2040
},
{
"epoch": 7.321428571428571,
"grad_norm": 2.0518875122070312,
"learning_rate": 1.4880952380952381e-05,
"loss": 0.3032,
"step": 2050
},
{
"epoch": 7.357142857142857,
"grad_norm": 2.7869536876678467,
"learning_rate": 1.4682539682539683e-05,
"loss": 0.2242,
"step": 2060
},
{
"epoch": 7.392857142857143,
"grad_norm": 1.007391095161438,
"learning_rate": 1.4484126984126987e-05,
"loss": 0.2589,
"step": 2070
},
{
"epoch": 7.428571428571429,
"grad_norm": 1.2812851667404175,
"learning_rate": 1.4285714285714285e-05,
"loss": 0.2383,
"step": 2080
},
{
"epoch": 7.464285714285714,
"grad_norm": 1.1465330123901367,
"learning_rate": 1.4087301587301587e-05,
"loss": 0.2284,
"step": 2090
},
{
"epoch": 7.5,
"grad_norm": 2.2567813396453857,
"learning_rate": 1.388888888888889e-05,
"loss": 0.213,
"step": 2100
},
{
"epoch": 7.535714285714286,
"grad_norm": 1.7949641942977905,
"learning_rate": 1.3690476190476192e-05,
"loss": 0.2368,
"step": 2110
},
{
"epoch": 7.571428571428571,
"grad_norm": 2.443598985671997,
"learning_rate": 1.3492063492063492e-05,
"loss": 0.2665,
"step": 2120
},
{
"epoch": 7.607142857142857,
"grad_norm": 1.9699336290359497,
"learning_rate": 1.3293650793650794e-05,
"loss": 0.2346,
"step": 2130
},
{
"epoch": 7.642857142857143,
"grad_norm": 1.543039083480835,
"learning_rate": 1.3095238095238096e-05,
"loss": 0.2432,
"step": 2140
},
{
"epoch": 7.678571428571429,
"grad_norm": 1.9814691543579102,
"learning_rate": 1.2896825396825398e-05,
"loss": 0.2575,
"step": 2150
},
{
"epoch": 7.714285714285714,
"grad_norm": 2.1088602542877197,
"learning_rate": 1.2698412698412699e-05,
"loss": 0.2346,
"step": 2160
},
{
"epoch": 7.75,
"grad_norm": 1.5648256540298462,
"learning_rate": 1.25e-05,
"loss": 0.2403,
"step": 2170
},
{
"epoch": 7.785714285714286,
"grad_norm": 1.6079583168029785,
"learning_rate": 1.2301587301587301e-05,
"loss": 0.2414,
"step": 2180
},
{
"epoch": 7.821428571428571,
"grad_norm": 1.2859593629837036,
"learning_rate": 1.2103174603174603e-05,
"loss": 0.2756,
"step": 2190
},
{
"epoch": 7.857142857142857,
"grad_norm": 2.072089672088623,
"learning_rate": 1.1904761904761905e-05,
"loss": 0.2484,
"step": 2200
},
{
"epoch": 7.892857142857143,
"grad_norm": 1.623353362083435,
"learning_rate": 1.1706349206349207e-05,
"loss": 0.274,
"step": 2210
},
{
"epoch": 7.928571428571429,
"grad_norm": 1.70241379737854,
"learning_rate": 1.1507936507936508e-05,
"loss": 0.2307,
"step": 2220
},
{
"epoch": 7.964285714285714,
"grad_norm": 1.7186700105667114,
"learning_rate": 1.130952380952381e-05,
"loss": 0.2223,
"step": 2230
},
{
"epoch": 8.0,
"grad_norm": 2.0665862560272217,
"learning_rate": 1.1111111111111112e-05,
"loss": 0.2274,
"step": 2240
},
{
"epoch": 8.0,
"eval_accuracy": 0.8814202769093346,
"eval_loss": 0.2860513925552368,
"eval_runtime": 115.9278,
"eval_samples_per_second": 38.627,
"eval_steps_per_second": 2.415,
"step": 2240
},
{
"epoch": 8.035714285714286,
"grad_norm": 1.3822689056396484,
"learning_rate": 1.0912698412698414e-05,
"loss": 0.238,
"step": 2250
},
{
"epoch": 8.071428571428571,
"grad_norm": 1.4226853847503662,
"learning_rate": 1.0714285714285714e-05,
"loss": 0.266,
"step": 2260
},
{
"epoch": 8.107142857142858,
"grad_norm": 1.9675803184509277,
"learning_rate": 1.0515873015873016e-05,
"loss": 0.2814,
"step": 2270
},
{
"epoch": 8.142857142857142,
"grad_norm": 1.8133440017700195,
"learning_rate": 1.0317460317460318e-05,
"loss": 0.2486,
"step": 2280
},
{
"epoch": 8.178571428571429,
"grad_norm": 1.2323780059814453,
"learning_rate": 1.011904761904762e-05,
"loss": 0.2175,
"step": 2290
},
{
"epoch": 8.214285714285714,
"grad_norm": 1.7063086032867432,
"learning_rate": 9.92063492063492e-06,
"loss": 0.2125,
"step": 2300
},
{
"epoch": 8.25,
"grad_norm": 1.532769799232483,
"learning_rate": 9.722222222222223e-06,
"loss": 0.1813,
"step": 2310
},
{
"epoch": 8.285714285714286,
"grad_norm": 1.5620160102844238,
"learning_rate": 9.523809523809523e-06,
"loss": 0.2355,
"step": 2320
},
{
"epoch": 8.321428571428571,
"grad_norm": 1.6230847835540771,
"learning_rate": 9.325396825396827e-06,
"loss": 0.2199,
"step": 2330
},
{
"epoch": 8.357142857142858,
"grad_norm": 1.552085041999817,
"learning_rate": 9.126984126984127e-06,
"loss": 0.2479,
"step": 2340
},
{
"epoch": 8.392857142857142,
"grad_norm": 1.8006685972213745,
"learning_rate": 8.92857142857143e-06,
"loss": 0.2677,
"step": 2350
},
{
"epoch": 8.428571428571429,
"grad_norm": 1.2057029008865356,
"learning_rate": 8.73015873015873e-06,
"loss": 0.2022,
"step": 2360
},
{
"epoch": 8.464285714285714,
"grad_norm": 1.4805638790130615,
"learning_rate": 8.531746031746032e-06,
"loss": 0.2158,
"step": 2370
},
{
"epoch": 8.5,
"grad_norm": 1.441603422164917,
"learning_rate": 8.333333333333334e-06,
"loss": 0.2166,
"step": 2380
},
{
"epoch": 8.535714285714286,
"grad_norm": 2.171687602996826,
"learning_rate": 8.134920634920636e-06,
"loss": 0.2332,
"step": 2390
},
{
"epoch": 8.571428571428571,
"grad_norm": 1.5523641109466553,
"learning_rate": 7.936507936507936e-06,
"loss": 0.2099,
"step": 2400
},
{
"epoch": 8.607142857142858,
"grad_norm": 1.429527759552002,
"learning_rate": 7.738095238095238e-06,
"loss": 0.1934,
"step": 2410
},
{
"epoch": 8.642857142857142,
"grad_norm": 3.8555209636688232,
"learning_rate": 7.5396825396825394e-06,
"loss": 0.2813,
"step": 2420
},
{
"epoch": 8.678571428571429,
"grad_norm": 1.118416666984558,
"learning_rate": 7.3412698412698415e-06,
"loss": 0.2358,
"step": 2430
},
{
"epoch": 8.714285714285714,
"grad_norm": 1.9037988185882568,
"learning_rate": 7.142857142857143e-06,
"loss": 0.2091,
"step": 2440
},
{
"epoch": 8.75,
"grad_norm": 1.8768919706344604,
"learning_rate": 6.944444444444445e-06,
"loss": 0.2176,
"step": 2450
},
{
"epoch": 8.785714285714286,
"grad_norm": 1.606123685836792,
"learning_rate": 6.746031746031746e-06,
"loss": 0.2388,
"step": 2460
},
{
"epoch": 8.821428571428571,
"grad_norm": 2.6942107677459717,
"learning_rate": 6.547619047619048e-06,
"loss": 0.2182,
"step": 2470
},
{
"epoch": 8.857142857142858,
"grad_norm": 1.449601173400879,
"learning_rate": 6.349206349206349e-06,
"loss": 0.2102,
"step": 2480
},
{
"epoch": 8.892857142857142,
"grad_norm": 2.651686668395996,
"learning_rate": 6.1507936507936505e-06,
"loss": 0.2166,
"step": 2490
},
{
"epoch": 8.928571428571429,
"grad_norm": 2.6743762493133545,
"learning_rate": 5.9523809523809525e-06,
"loss": 0.2468,
"step": 2500
},
{
"epoch": 8.964285714285714,
"grad_norm": 2.29903244972229,
"learning_rate": 5.753968253968254e-06,
"loss": 0.2111,
"step": 2510
},
{
"epoch": 9.0,
"grad_norm": 1.8269622325897217,
"learning_rate": 5.555555555555556e-06,
"loss": 0.2195,
"step": 2520
},
{
"epoch": 9.0,
"eval_accuracy": 0.8894595801697186,
"eval_loss": 0.26261791586875916,
"eval_runtime": 114.2956,
"eval_samples_per_second": 39.179,
"eval_steps_per_second": 2.45,
"step": 2520
},
{
"epoch": 9.035714285714286,
"grad_norm": 3.011329174041748,
"learning_rate": 5.357142857142857e-06,
"loss": 0.1827,
"step": 2530
},
{
"epoch": 9.071428571428571,
"grad_norm": 2.3060262203216553,
"learning_rate": 5.158730158730159e-06,
"loss": 0.1862,
"step": 2540
},
{
"epoch": 9.107142857142858,
"grad_norm": 1.7220100164413452,
"learning_rate": 4.96031746031746e-06,
"loss": 0.2215,
"step": 2550
},
{
"epoch": 9.142857142857142,
"grad_norm": 2.463092803955078,
"learning_rate": 4.7619047619047615e-06,
"loss": 0.2228,
"step": 2560
},
{
"epoch": 9.178571428571429,
"grad_norm": 1.204136848449707,
"learning_rate": 4.563492063492064e-06,
"loss": 0.2439,
"step": 2570
},
{
"epoch": 9.214285714285714,
"grad_norm": 2.263396978378296,
"learning_rate": 4.365079365079365e-06,
"loss": 0.2266,
"step": 2580
},
{
"epoch": 9.25,
"grad_norm": 2.7832555770874023,
"learning_rate": 4.166666666666667e-06,
"loss": 0.2089,
"step": 2590
},
{
"epoch": 9.285714285714286,
"grad_norm": 2.0564024448394775,
"learning_rate": 3.968253968253968e-06,
"loss": 0.1839,
"step": 2600
},
{
"epoch": 9.321428571428571,
"grad_norm": 2.0316998958587646,
"learning_rate": 3.7698412698412697e-06,
"loss": 0.2424,
"step": 2610
},
{
"epoch": 9.357142857142858,
"grad_norm": 2.229687213897705,
"learning_rate": 3.5714285714285714e-06,
"loss": 0.2489,
"step": 2620
},
{
"epoch": 9.392857142857142,
"grad_norm": 1.7529199123382568,
"learning_rate": 3.373015873015873e-06,
"loss": 0.2273,
"step": 2630
},
{
"epoch": 9.428571428571429,
"grad_norm": 1.5242239236831665,
"learning_rate": 3.1746031746031746e-06,
"loss": 0.2224,
"step": 2640
},
{
"epoch": 9.464285714285714,
"grad_norm": 1.5499508380889893,
"learning_rate": 2.9761904761904763e-06,
"loss": 0.2338,
"step": 2650
},
{
"epoch": 9.5,
"grad_norm": 2.5902230739593506,
"learning_rate": 2.777777777777778e-06,
"loss": 0.1909,
"step": 2660
},
{
"epoch": 9.535714285714286,
"grad_norm": 1.3243242502212524,
"learning_rate": 2.5793650793650795e-06,
"loss": 0.2199,
"step": 2670
},
{
"epoch": 9.571428571428571,
"grad_norm": 1.9745112657546997,
"learning_rate": 2.3809523809523808e-06,
"loss": 0.2172,
"step": 2680
},
{
"epoch": 9.607142857142858,
"grad_norm": 2.10951828956604,
"learning_rate": 2.1825396825396824e-06,
"loss": 0.2331,
"step": 2690
},
{
"epoch": 9.642857142857142,
"grad_norm": 2.113539457321167,
"learning_rate": 1.984126984126984e-06,
"loss": 0.2373,
"step": 2700
},
{
"epoch": 9.678571428571429,
"grad_norm": 1.548854112625122,
"learning_rate": 1.7857142857142857e-06,
"loss": 0.2035,
"step": 2710
},
{
"epoch": 9.714285714285714,
"grad_norm": 1.797196626663208,
"learning_rate": 1.5873015873015873e-06,
"loss": 0.2015,
"step": 2720
},
{
"epoch": 9.75,
"grad_norm": 1.9279841184616089,
"learning_rate": 1.388888888888889e-06,
"loss": 0.2208,
"step": 2730
},
{
"epoch": 9.785714285714286,
"grad_norm": 0.950290858745575,
"learning_rate": 1.1904761904761904e-06,
"loss": 0.2394,
"step": 2740
},
{
"epoch": 9.821428571428571,
"grad_norm": 1.2573094367980957,
"learning_rate": 9.92063492063492e-07,
"loss": 0.2491,
"step": 2750
},
{
"epoch": 9.857142857142858,
"grad_norm": 1.8162927627563477,
"learning_rate": 7.936507936507937e-07,
"loss": 0.2229,
"step": 2760
},
{
"epoch": 9.892857142857142,
"grad_norm": 2.0845260620117188,
"learning_rate": 5.952380952380952e-07,
"loss": 0.1984,
"step": 2770
},
{
"epoch": 9.928571428571429,
"grad_norm": 1.8501282930374146,
"learning_rate": 3.9682539682539683e-07,
"loss": 0.2287,
"step": 2780
},
{
"epoch": 9.964285714285714,
"grad_norm": 2.941807746887207,
"learning_rate": 1.9841269841269841e-07,
"loss": 0.2233,
"step": 2790
},
{
"epoch": 10.0,
"grad_norm": 1.2508630752563477,
"learning_rate": 0.0,
"loss": 0.1886,
"step": 2800
},
{
"epoch": 10.0,
"eval_accuracy": 0.8865564984368022,
"eval_loss": 0.2717145085334778,
"eval_runtime": 113.959,
"eval_samples_per_second": 39.295,
"eval_steps_per_second": 2.457,
"step": 2800
},
{
"epoch": 10.0,
"step": 2800,
"total_flos": 1.3877265500181135e+19,
"train_loss": 0.3087472263830049,
"train_runtime": 6534.1253,
"train_samples_per_second": 27.407,
"train_steps_per_second": 0.429
}
],
"logging_steps": 10,
"max_steps": 2800,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.3877265500181135e+19,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}