v51 / trainer_state.json
LogicCrafters's picture
Upload 9 files
1d3a92f verified
raw
history blame
203 kB
{
"best_metric": 0.9952178768165258,
"best_model_checkpoint": "output_dev/v51-albert-xxlarge-v2-f1-warmup-greater-patience-subword-len-500-maxlen-512-6epoch/checkpoint-10400",
"epoch": 6.0,
"eval_steps": 200,
"global_step": 14346,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 6.963788300835655e-07,
"loss": 1.4647,
"step": 10
},
{
"epoch": 0.01,
"learning_rate": 1.392757660167131e-06,
"loss": 1.4624,
"step": 20
},
{
"epoch": 0.01,
"learning_rate": 2.0891364902506964e-06,
"loss": 1.4653,
"step": 30
},
{
"epoch": 0.02,
"learning_rate": 2.785515320334262e-06,
"loss": 1.4554,
"step": 40
},
{
"epoch": 0.02,
"learning_rate": 3.4818941504178276e-06,
"loss": 1.4516,
"step": 50
},
{
"epoch": 0.03,
"learning_rate": 4.178272980501393e-06,
"loss": 1.4369,
"step": 60
},
{
"epoch": 0.03,
"learning_rate": 4.874651810584958e-06,
"loss": 1.425,
"step": 70
},
{
"epoch": 0.03,
"learning_rate": 5.571030640668524e-06,
"loss": 1.3959,
"step": 80
},
{
"epoch": 0.04,
"learning_rate": 6.267409470752089e-06,
"loss": 1.3708,
"step": 90
},
{
"epoch": 0.04,
"learning_rate": 6.963788300835655e-06,
"loss": 1.3367,
"step": 100
},
{
"epoch": 0.05,
"learning_rate": 7.66016713091922e-06,
"loss": 1.2928,
"step": 110
},
{
"epoch": 0.05,
"learning_rate": 8.356545961002786e-06,
"loss": 1.2327,
"step": 120
},
{
"epoch": 0.05,
"learning_rate": 9.05292479108635e-06,
"loss": 1.1415,
"step": 130
},
{
"epoch": 0.06,
"learning_rate": 9.749303621169916e-06,
"loss": 1.0479,
"step": 140
},
{
"epoch": 0.06,
"learning_rate": 1.0445682451253484e-05,
"loss": 0.9095,
"step": 150
},
{
"epoch": 0.07,
"learning_rate": 1.1142061281337048e-05,
"loss": 0.8537,
"step": 160
},
{
"epoch": 0.07,
"learning_rate": 1.1838440111420614e-05,
"loss": 0.7396,
"step": 170
},
{
"epoch": 0.08,
"learning_rate": 1.2534818941504178e-05,
"loss": 0.7262,
"step": 180
},
{
"epoch": 0.08,
"learning_rate": 1.3231197771587744e-05,
"loss": 0.7011,
"step": 190
},
{
"epoch": 0.08,
"learning_rate": 1.392757660167131e-05,
"loss": 0.5065,
"step": 200
},
{
"epoch": 0.08,
"eval_f1": 0.4947420073867864,
"eval_f1_0": 0.907618966173514,
"eval_f1_1": 0.2064158062680717,
"eval_f1_2": 2.3092021706500406e-05,
"eval_f1_3": 0.8649101650838534,
"eval_loss": 0.5066748857498169,
"eval_runtime": 32.3718,
"eval_samples_per_second": 262.636,
"eval_steps_per_second": 1.05,
"step": 200
},
{
"epoch": 0.09,
"learning_rate": 1.4623955431754876e-05,
"loss": 0.4888,
"step": 210
},
{
"epoch": 0.09,
"learning_rate": 1.532033426183844e-05,
"loss": 0.4292,
"step": 220
},
{
"epoch": 0.1,
"learning_rate": 1.6016713091922005e-05,
"loss": 0.4197,
"step": 230
},
{
"epoch": 0.1,
"learning_rate": 1.671309192200557e-05,
"loss": 0.366,
"step": 240
},
{
"epoch": 0.1,
"learning_rate": 1.7409470752089137e-05,
"loss": 0.3432,
"step": 250
},
{
"epoch": 0.11,
"learning_rate": 1.81058495821727e-05,
"loss": 0.3301,
"step": 260
},
{
"epoch": 0.11,
"learning_rate": 1.8802228412256266e-05,
"loss": 0.3191,
"step": 270
},
{
"epoch": 0.12,
"learning_rate": 1.9498607242339832e-05,
"loss": 0.3272,
"step": 280
},
{
"epoch": 0.12,
"learning_rate": 2.01949860724234e-05,
"loss": 0.2712,
"step": 290
},
{
"epoch": 0.13,
"learning_rate": 2.0891364902506967e-05,
"loss": 0.2079,
"step": 300
},
{
"epoch": 0.13,
"learning_rate": 2.158774373259053e-05,
"loss": 0.2738,
"step": 310
},
{
"epoch": 0.13,
"learning_rate": 2.2284122562674096e-05,
"loss": 0.2826,
"step": 320
},
{
"epoch": 0.14,
"learning_rate": 2.2980501392757662e-05,
"loss": 0.2076,
"step": 330
},
{
"epoch": 0.14,
"learning_rate": 2.3676880222841228e-05,
"loss": 0.1961,
"step": 340
},
{
"epoch": 0.15,
"learning_rate": 2.4373259052924794e-05,
"loss": 0.1935,
"step": 350
},
{
"epoch": 0.15,
"learning_rate": 2.5069637883008357e-05,
"loss": 0.1943,
"step": 360
},
{
"epoch": 0.15,
"learning_rate": 2.5766016713091923e-05,
"loss": 0.1781,
"step": 370
},
{
"epoch": 0.16,
"learning_rate": 2.646239554317549e-05,
"loss": 0.1982,
"step": 380
},
{
"epoch": 0.16,
"learning_rate": 2.7158774373259055e-05,
"loss": 0.1658,
"step": 390
},
{
"epoch": 0.17,
"learning_rate": 2.785515320334262e-05,
"loss": 0.1647,
"step": 400
},
{
"epoch": 0.17,
"eval_f1": 0.86330015860688,
"eval_f1_0": 0.9694002757472255,
"eval_f1_1": 0.9603298128589958,
"eval_f1_2": 0.5580695616873426,
"eval_f1_3": 0.9654009841339561,
"eval_loss": 0.16295476257801056,
"eval_runtime": 32.4551,
"eval_samples_per_second": 261.962,
"eval_steps_per_second": 1.048,
"step": 400
},
{
"epoch": 0.17,
"learning_rate": 2.8551532033426187e-05,
"loss": 0.1659,
"step": 410
},
{
"epoch": 0.18,
"learning_rate": 2.9247910863509753e-05,
"loss": 0.1693,
"step": 420
},
{
"epoch": 0.18,
"learning_rate": 2.9944289693593312e-05,
"loss": 0.1594,
"step": 430
},
{
"epoch": 0.18,
"learning_rate": 3.064066852367688e-05,
"loss": 0.136,
"step": 440
},
{
"epoch": 0.19,
"learning_rate": 3.133704735376045e-05,
"loss": 0.128,
"step": 450
},
{
"epoch": 0.19,
"learning_rate": 3.203342618384401e-05,
"loss": 0.1643,
"step": 460
},
{
"epoch": 0.2,
"learning_rate": 3.272980501392758e-05,
"loss": 0.1686,
"step": 470
},
{
"epoch": 0.2,
"learning_rate": 3.342618384401114e-05,
"loss": 0.127,
"step": 480
},
{
"epoch": 0.2,
"learning_rate": 3.412256267409471e-05,
"loss": 0.1007,
"step": 490
},
{
"epoch": 0.21,
"learning_rate": 3.4818941504178274e-05,
"loss": 0.1653,
"step": 500
},
{
"epoch": 0.21,
"learning_rate": 3.5515320334261844e-05,
"loss": 0.2064,
"step": 510
},
{
"epoch": 0.22,
"learning_rate": 3.62116991643454e-05,
"loss": 0.1251,
"step": 520
},
{
"epoch": 0.22,
"learning_rate": 3.690807799442897e-05,
"loss": 0.1596,
"step": 530
},
{
"epoch": 0.23,
"learning_rate": 3.760445682451253e-05,
"loss": 0.1209,
"step": 540
},
{
"epoch": 0.23,
"learning_rate": 3.83008356545961e-05,
"loss": 0.123,
"step": 550
},
{
"epoch": 0.23,
"learning_rate": 3.8997214484679664e-05,
"loss": 0.1157,
"step": 560
},
{
"epoch": 0.24,
"learning_rate": 3.969359331476323e-05,
"loss": 0.1054,
"step": 570
},
{
"epoch": 0.24,
"learning_rate": 4.03899721448468e-05,
"loss": 0.1422,
"step": 580
},
{
"epoch": 0.25,
"learning_rate": 4.1086350974930365e-05,
"loss": 0.0972,
"step": 590
},
{
"epoch": 0.25,
"learning_rate": 4.1782729805013935e-05,
"loss": 0.098,
"step": 600
},
{
"epoch": 0.25,
"eval_f1": 0.9404645037956795,
"eval_f1_0": 0.9818690721149516,
"eval_f1_1": 0.9774477675396847,
"eval_f1_2": 0.8171029502056425,
"eval_f1_3": 0.9854382253224393,
"eval_loss": 0.09070613980293274,
"eval_runtime": 32.3186,
"eval_samples_per_second": 263.069,
"eval_steps_per_second": 1.052,
"step": 600
},
{
"epoch": 0.26,
"learning_rate": 4.24791086350975e-05,
"loss": 0.1296,
"step": 610
},
{
"epoch": 0.26,
"learning_rate": 4.317548746518106e-05,
"loss": 0.1035,
"step": 620
},
{
"epoch": 0.26,
"learning_rate": 4.387186629526462e-05,
"loss": 0.1238,
"step": 630
},
{
"epoch": 0.27,
"learning_rate": 4.456824512534819e-05,
"loss": 0.111,
"step": 640
},
{
"epoch": 0.27,
"learning_rate": 4.5264623955431755e-05,
"loss": 0.0967,
"step": 650
},
{
"epoch": 0.28,
"learning_rate": 4.5961002785515324e-05,
"loss": 0.0875,
"step": 660
},
{
"epoch": 0.28,
"learning_rate": 4.665738161559889e-05,
"loss": 0.0859,
"step": 670
},
{
"epoch": 0.28,
"learning_rate": 4.7353760445682456e-05,
"loss": 0.0822,
"step": 680
},
{
"epoch": 0.29,
"learning_rate": 4.805013927576602e-05,
"loss": 0.0947,
"step": 690
},
{
"epoch": 0.29,
"learning_rate": 4.874651810584959e-05,
"loss": 0.097,
"step": 700
},
{
"epoch": 0.3,
"learning_rate": 4.9442896935933144e-05,
"loss": 0.1086,
"step": 710
},
{
"epoch": 0.3,
"learning_rate": 4.999266216612856e-05,
"loss": 0.0841,
"step": 720
},
{
"epoch": 0.31,
"learning_rate": 4.9955972996771355e-05,
"loss": 0.0966,
"step": 730
},
{
"epoch": 0.31,
"learning_rate": 4.991928382741415e-05,
"loss": 0.0714,
"step": 740
},
{
"epoch": 0.31,
"learning_rate": 4.988259465805694e-05,
"loss": 0.1077,
"step": 750
},
{
"epoch": 0.32,
"learning_rate": 4.984590548869974e-05,
"loss": 0.0761,
"step": 760
},
{
"epoch": 0.32,
"learning_rate": 4.9809216319342535e-05,
"loss": 0.0783,
"step": 770
},
{
"epoch": 0.33,
"learning_rate": 4.977252714998533e-05,
"loss": 0.0731,
"step": 780
},
{
"epoch": 0.33,
"learning_rate": 4.973583798062812e-05,
"loss": 0.0694,
"step": 790
},
{
"epoch": 0.33,
"learning_rate": 4.969914881127091e-05,
"loss": 0.0373,
"step": 800
},
{
"epoch": 0.33,
"eval_f1": 0.9654745267462422,
"eval_f1_0": 0.9892180356896579,
"eval_f1_1": 0.9879130190277954,
"eval_f1_2": 0.8930634422007959,
"eval_f1_3": 0.9917036100667196,
"eval_loss": 0.05413857102394104,
"eval_runtime": 32.4375,
"eval_samples_per_second": 262.104,
"eval_steps_per_second": 1.048,
"step": 800
},
{
"epoch": 0.34,
"learning_rate": 4.966245964191371e-05,
"loss": 0.0682,
"step": 810
},
{
"epoch": 0.34,
"learning_rate": 4.96257704725565e-05,
"loss": 0.0405,
"step": 820
},
{
"epoch": 0.35,
"learning_rate": 4.9589081303199296e-05,
"loss": 0.05,
"step": 830
},
{
"epoch": 0.35,
"learning_rate": 4.955239213384209e-05,
"loss": 0.0348,
"step": 840
},
{
"epoch": 0.36,
"learning_rate": 4.951570296448489e-05,
"loss": 0.079,
"step": 850
},
{
"epoch": 0.36,
"learning_rate": 4.9479013795127684e-05,
"loss": 0.0522,
"step": 860
},
{
"epoch": 0.36,
"learning_rate": 4.944232462577048e-05,
"loss": 0.0309,
"step": 870
},
{
"epoch": 0.37,
"learning_rate": 4.940563545641327e-05,
"loss": 0.043,
"step": 880
},
{
"epoch": 0.37,
"learning_rate": 4.9368946287056064e-05,
"loss": 0.0267,
"step": 890
},
{
"epoch": 0.38,
"learning_rate": 4.933225711769886e-05,
"loss": 0.0533,
"step": 900
},
{
"epoch": 0.38,
"learning_rate": 4.929556794834165e-05,
"loss": 0.0576,
"step": 910
},
{
"epoch": 0.38,
"learning_rate": 4.9258878778984445e-05,
"loss": 0.0232,
"step": 920
},
{
"epoch": 0.39,
"learning_rate": 4.922218960962724e-05,
"loss": 0.0439,
"step": 930
},
{
"epoch": 0.39,
"learning_rate": 4.918550044027003e-05,
"loss": 0.0313,
"step": 940
},
{
"epoch": 0.4,
"learning_rate": 4.914881127091283e-05,
"loss": 0.0549,
"step": 950
},
{
"epoch": 0.4,
"learning_rate": 4.9112122101555626e-05,
"loss": 0.0263,
"step": 960
},
{
"epoch": 0.41,
"learning_rate": 4.907543293219842e-05,
"loss": 0.0342,
"step": 970
},
{
"epoch": 0.41,
"learning_rate": 4.903874376284121e-05,
"loss": 0.0686,
"step": 980
},
{
"epoch": 0.41,
"learning_rate": 4.9002054593484006e-05,
"loss": 0.023,
"step": 990
},
{
"epoch": 0.42,
"learning_rate": 4.89653654241268e-05,
"loss": 0.0288,
"step": 1000
},
{
"epoch": 0.42,
"eval_f1": 0.9863366482064234,
"eval_f1_0": 0.9958609934536309,
"eval_f1_1": 0.9927170125738525,
"eval_f1_2": 0.9601819591865082,
"eval_f1_3": 0.9965866276117024,
"eval_loss": 0.025030702352523804,
"eval_runtime": 32.3462,
"eval_samples_per_second": 262.844,
"eval_steps_per_second": 1.051,
"step": 1000
},
{
"epoch": 0.42,
"learning_rate": 4.892867625476959e-05,
"loss": 0.0179,
"step": 1010
},
{
"epoch": 0.43,
"learning_rate": 4.8891987085412387e-05,
"loss": 0.0352,
"step": 1020
},
{
"epoch": 0.43,
"learning_rate": 4.885529791605518e-05,
"loss": 0.0311,
"step": 1030
},
{
"epoch": 0.43,
"learning_rate": 4.8818608746697974e-05,
"loss": 0.0202,
"step": 1040
},
{
"epoch": 0.44,
"learning_rate": 4.8781919577340774e-05,
"loss": 0.0189,
"step": 1050
},
{
"epoch": 0.44,
"learning_rate": 4.874523040798357e-05,
"loss": 0.0166,
"step": 1060
},
{
"epoch": 0.45,
"learning_rate": 4.870854123862636e-05,
"loss": 0.0193,
"step": 1070
},
{
"epoch": 0.45,
"learning_rate": 4.8671852069269154e-05,
"loss": 0.0575,
"step": 1080
},
{
"epoch": 0.46,
"learning_rate": 4.863516289991195e-05,
"loss": 0.0294,
"step": 1090
},
{
"epoch": 0.46,
"learning_rate": 4.859847373055474e-05,
"loss": 0.0437,
"step": 1100
},
{
"epoch": 0.46,
"learning_rate": 4.8561784561197535e-05,
"loss": 0.0408,
"step": 1110
},
{
"epoch": 0.47,
"learning_rate": 4.852509539184033e-05,
"loss": 0.0207,
"step": 1120
},
{
"epoch": 0.47,
"learning_rate": 4.848840622248312e-05,
"loss": 0.0285,
"step": 1130
},
{
"epoch": 0.48,
"learning_rate": 4.8451717053125915e-05,
"loss": 0.0175,
"step": 1140
},
{
"epoch": 0.48,
"learning_rate": 4.8415027883768716e-05,
"loss": 0.025,
"step": 1150
},
{
"epoch": 0.49,
"learning_rate": 4.837833871441151e-05,
"loss": 0.0286,
"step": 1160
},
{
"epoch": 0.49,
"learning_rate": 4.83416495450543e-05,
"loss": 0.0422,
"step": 1170
},
{
"epoch": 0.49,
"learning_rate": 4.8304960375697096e-05,
"loss": 0.0591,
"step": 1180
},
{
"epoch": 0.5,
"learning_rate": 4.8268271206339896e-05,
"loss": 0.037,
"step": 1190
},
{
"epoch": 0.5,
"learning_rate": 4.823158203698268e-05,
"loss": 0.0144,
"step": 1200
},
{
"epoch": 0.5,
"eval_f1": 0.9720761324865212,
"eval_f1_0": 0.9919368850035413,
"eval_f1_1": 0.9884808336322845,
"eval_f1_2": 0.9129289565887235,
"eval_f1_3": 0.9949578547215354,
"eval_loss": 0.04629531130194664,
"eval_runtime": 32.4102,
"eval_samples_per_second": 262.325,
"eval_steps_per_second": 1.049,
"step": 1200
},
{
"epoch": 0.51,
"learning_rate": 4.819489286762548e-05,
"loss": 0.0199,
"step": 1210
},
{
"epoch": 0.51,
"learning_rate": 4.815820369826827e-05,
"loss": 0.0465,
"step": 1220
},
{
"epoch": 0.51,
"learning_rate": 4.8121514528911064e-05,
"loss": 0.0252,
"step": 1230
},
{
"epoch": 0.52,
"learning_rate": 4.8084825359553864e-05,
"loss": 0.0225,
"step": 1240
},
{
"epoch": 0.52,
"learning_rate": 4.804813619019666e-05,
"loss": 0.0133,
"step": 1250
},
{
"epoch": 0.53,
"learning_rate": 4.801144702083945e-05,
"loss": 0.053,
"step": 1260
},
{
"epoch": 0.53,
"learning_rate": 4.7974757851482245e-05,
"loss": 0.0192,
"step": 1270
},
{
"epoch": 0.54,
"learning_rate": 4.793806868212504e-05,
"loss": 0.0141,
"step": 1280
},
{
"epoch": 0.54,
"learning_rate": 4.790137951276784e-05,
"loss": 0.033,
"step": 1290
},
{
"epoch": 0.54,
"learning_rate": 4.786469034341063e-05,
"loss": 0.0396,
"step": 1300
},
{
"epoch": 0.55,
"learning_rate": 4.782800117405342e-05,
"loss": 0.0229,
"step": 1310
},
{
"epoch": 0.55,
"learning_rate": 4.779131200469621e-05,
"loss": 0.0316,
"step": 1320
},
{
"epoch": 0.56,
"learning_rate": 4.7754622835339006e-05,
"loss": 0.0301,
"step": 1330
},
{
"epoch": 0.56,
"learning_rate": 4.7717933665981806e-05,
"loss": 0.0455,
"step": 1340
},
{
"epoch": 0.56,
"learning_rate": 4.76812444966246e-05,
"loss": 0.0147,
"step": 1350
},
{
"epoch": 0.57,
"learning_rate": 4.764455532726739e-05,
"loss": 0.0761,
"step": 1360
},
{
"epoch": 0.57,
"learning_rate": 4.7607866157910186e-05,
"loss": 0.007,
"step": 1370
},
{
"epoch": 0.58,
"learning_rate": 4.757117698855298e-05,
"loss": 0.0662,
"step": 1380
},
{
"epoch": 0.58,
"learning_rate": 4.753448781919578e-05,
"loss": 0.0136,
"step": 1390
},
{
"epoch": 0.59,
"learning_rate": 4.7497798649838574e-05,
"loss": 0.0085,
"step": 1400
},
{
"epoch": 0.59,
"eval_f1": 0.9695984266051492,
"eval_f1_0": 0.9916777683074244,
"eval_f1_1": 0.9946781089850502,
"eval_f1_2": 0.894025197251209,
"eval_f1_3": 0.9980126318769129,
"eval_loss": 0.04921557009220123,
"eval_runtime": 32.43,
"eval_samples_per_second": 262.165,
"eval_steps_per_second": 1.048,
"step": 1400
},
{
"epoch": 0.59,
"learning_rate": 4.746110948048137e-05,
"loss": 0.0229,
"step": 1410
},
{
"epoch": 0.59,
"learning_rate": 4.7424420311124154e-05,
"loss": 0.0285,
"step": 1420
},
{
"epoch": 0.6,
"learning_rate": 4.738773114176695e-05,
"loss": 0.0336,
"step": 1430
},
{
"epoch": 0.6,
"learning_rate": 4.735104197240975e-05,
"loss": 0.0334,
"step": 1440
},
{
"epoch": 0.61,
"learning_rate": 4.731435280305254e-05,
"loss": 0.0133,
"step": 1450
},
{
"epoch": 0.61,
"learning_rate": 4.7277663633695335e-05,
"loss": 0.0254,
"step": 1460
},
{
"epoch": 0.61,
"learning_rate": 4.724097446433813e-05,
"loss": 0.0198,
"step": 1470
},
{
"epoch": 0.62,
"learning_rate": 4.720428529498093e-05,
"loss": 0.0309,
"step": 1480
},
{
"epoch": 0.62,
"learning_rate": 4.716759612562372e-05,
"loss": 0.0253,
"step": 1490
},
{
"epoch": 0.63,
"learning_rate": 4.7130906956266515e-05,
"loss": 0.0236,
"step": 1500
},
{
"epoch": 0.63,
"learning_rate": 4.709421778690931e-05,
"loss": 0.0113,
"step": 1510
},
{
"epoch": 0.64,
"learning_rate": 4.70575286175521e-05,
"loss": 0.0246,
"step": 1520
},
{
"epoch": 0.64,
"learning_rate": 4.7020839448194896e-05,
"loss": 0.0076,
"step": 1530
},
{
"epoch": 0.64,
"learning_rate": 4.698415027883769e-05,
"loss": 0.0114,
"step": 1540
},
{
"epoch": 0.65,
"learning_rate": 4.694746110948048e-05,
"loss": 0.0198,
"step": 1550
},
{
"epoch": 0.65,
"learning_rate": 4.6910771940123276e-05,
"loss": 0.0428,
"step": 1560
},
{
"epoch": 0.66,
"learning_rate": 4.687408277076607e-05,
"loss": 0.0251,
"step": 1570
},
{
"epoch": 0.66,
"learning_rate": 4.683739360140887e-05,
"loss": 0.0157,
"step": 1580
},
{
"epoch": 0.66,
"learning_rate": 4.6800704432051664e-05,
"loss": 0.0072,
"step": 1590
},
{
"epoch": 0.67,
"learning_rate": 4.676401526269446e-05,
"loss": 0.0094,
"step": 1600
},
{
"epoch": 0.67,
"eval_f1": 0.9883447204504365,
"eval_f1_0": 0.9964754968694822,
"eval_f1_1": 0.9945256375675747,
"eval_f1_2": 0.970150567415005,
"eval_f1_3": 0.9922271799496843,
"eval_loss": 0.02080400660634041,
"eval_runtime": 32.5031,
"eval_samples_per_second": 261.575,
"eval_steps_per_second": 1.046,
"step": 1600
},
{
"epoch": 0.67,
"learning_rate": 4.672732609333725e-05,
"loss": 0.1042,
"step": 1610
},
{
"epoch": 0.68,
"learning_rate": 4.6690636923980044e-05,
"loss": 0.0216,
"step": 1620
},
{
"epoch": 0.68,
"learning_rate": 4.665394775462284e-05,
"loss": 0.0269,
"step": 1630
},
{
"epoch": 0.69,
"learning_rate": 4.661725858526563e-05,
"loss": 0.0092,
"step": 1640
},
{
"epoch": 0.69,
"learning_rate": 4.6580569415908425e-05,
"loss": 0.0208,
"step": 1650
},
{
"epoch": 0.69,
"learning_rate": 4.654388024655122e-05,
"loss": 0.0416,
"step": 1660
},
{
"epoch": 0.7,
"learning_rate": 4.650719107719401e-05,
"loss": 0.021,
"step": 1670
},
{
"epoch": 0.7,
"learning_rate": 4.647050190783681e-05,
"loss": 0.0123,
"step": 1680
},
{
"epoch": 0.71,
"learning_rate": 4.6433812738479606e-05,
"loss": 0.033,
"step": 1690
},
{
"epoch": 0.71,
"learning_rate": 4.63971235691224e-05,
"loss": 0.0353,
"step": 1700
},
{
"epoch": 0.72,
"learning_rate": 4.636043439976519e-05,
"loss": 0.0184,
"step": 1710
},
{
"epoch": 0.72,
"learning_rate": 4.6323745230407986e-05,
"loss": 0.0068,
"step": 1720
},
{
"epoch": 0.72,
"learning_rate": 4.628705606105078e-05,
"loss": 0.0113,
"step": 1730
},
{
"epoch": 0.73,
"learning_rate": 4.625036689169357e-05,
"loss": 0.0292,
"step": 1740
},
{
"epoch": 0.73,
"learning_rate": 4.6213677722336367e-05,
"loss": 0.0139,
"step": 1750
},
{
"epoch": 0.74,
"learning_rate": 4.617698855297916e-05,
"loss": 0.0238,
"step": 1760
},
{
"epoch": 0.74,
"learning_rate": 4.6140299383621954e-05,
"loss": 0.0054,
"step": 1770
},
{
"epoch": 0.74,
"learning_rate": 4.6103610214264754e-05,
"loss": 0.0049,
"step": 1780
},
{
"epoch": 0.75,
"learning_rate": 4.606692104490755e-05,
"loss": 0.0125,
"step": 1790
},
{
"epoch": 0.75,
"learning_rate": 4.603023187555034e-05,
"loss": 0.0288,
"step": 1800
},
{
"epoch": 0.75,
"eval_f1": 0.9926857697067606,
"eval_f1_0": 0.9977940926048485,
"eval_f1_1": 0.9963591980001175,
"eval_f1_2": 0.9781227618962013,
"eval_f1_3": 0.9984670263258748,
"eval_loss": 0.014699622988700867,
"eval_runtime": 32.4881,
"eval_samples_per_second": 261.696,
"eval_steps_per_second": 1.047,
"step": 1800
},
{
"epoch": 0.76,
"learning_rate": 4.5993542706193134e-05,
"loss": 0.0097,
"step": 1810
},
{
"epoch": 0.76,
"learning_rate": 4.595685353683593e-05,
"loss": 0.0336,
"step": 1820
},
{
"epoch": 0.77,
"learning_rate": 4.592016436747872e-05,
"loss": 0.0103,
"step": 1830
},
{
"epoch": 0.77,
"learning_rate": 4.5883475198121515e-05,
"loss": 0.018,
"step": 1840
},
{
"epoch": 0.77,
"learning_rate": 4.584678602876431e-05,
"loss": 0.0384,
"step": 1850
},
{
"epoch": 0.78,
"learning_rate": 4.58100968594071e-05,
"loss": 0.0132,
"step": 1860
},
{
"epoch": 0.78,
"learning_rate": 4.57734076900499e-05,
"loss": 0.0121,
"step": 1870
},
{
"epoch": 0.79,
"learning_rate": 4.5736718520692696e-05,
"loss": 0.0324,
"step": 1880
},
{
"epoch": 0.79,
"learning_rate": 4.570002935133549e-05,
"loss": 0.0103,
"step": 1890
},
{
"epoch": 0.79,
"learning_rate": 4.566334018197828e-05,
"loss": 0.0298,
"step": 1900
},
{
"epoch": 0.8,
"learning_rate": 4.5626651012621076e-05,
"loss": 0.0058,
"step": 1910
},
{
"epoch": 0.8,
"learning_rate": 4.558996184326387e-05,
"loss": 0.0143,
"step": 1920
},
{
"epoch": 0.81,
"learning_rate": 4.555327267390666e-05,
"loss": 0.0296,
"step": 1930
},
{
"epoch": 0.81,
"learning_rate": 4.551658350454946e-05,
"loss": 0.0261,
"step": 1940
},
{
"epoch": 0.82,
"learning_rate": 4.547989433519225e-05,
"loss": 0.0161,
"step": 1950
},
{
"epoch": 0.82,
"learning_rate": 4.5443205165835044e-05,
"loss": 0.0386,
"step": 1960
},
{
"epoch": 0.82,
"learning_rate": 4.5406515996477844e-05,
"loss": 0.0119,
"step": 1970
},
{
"epoch": 0.83,
"learning_rate": 4.536982682712064e-05,
"loss": 0.0064,
"step": 1980
},
{
"epoch": 0.83,
"learning_rate": 4.533313765776343e-05,
"loss": 0.0327,
"step": 1990
},
{
"epoch": 0.84,
"learning_rate": 4.5296448488406225e-05,
"loss": 0.0167,
"step": 2000
},
{
"epoch": 0.84,
"eval_f1": 0.9910545937397852,
"eval_f1_0": 0.9972999588222994,
"eval_f1_1": 0.9961797867726301,
"eval_f1_2": 0.9735686874672773,
"eval_f1_3": 0.9971699418969341,
"eval_loss": 0.016176139935851097,
"eval_runtime": 32.5916,
"eval_samples_per_second": 260.865,
"eval_steps_per_second": 1.043,
"step": 2000
},
{
"epoch": 0.84,
"learning_rate": 4.525975931904902e-05,
"loss": 0.0134,
"step": 2010
},
{
"epoch": 0.84,
"learning_rate": 4.522307014969182e-05,
"loss": 0.0105,
"step": 2020
},
{
"epoch": 0.85,
"learning_rate": 4.5186380980334605e-05,
"loss": 0.0172,
"step": 2030
},
{
"epoch": 0.85,
"learning_rate": 4.51496918109774e-05,
"loss": 0.0255,
"step": 2040
},
{
"epoch": 0.86,
"learning_rate": 4.511300264162019e-05,
"loss": 0.0291,
"step": 2050
},
{
"epoch": 0.86,
"learning_rate": 4.5076313472262986e-05,
"loss": 0.0057,
"step": 2060
},
{
"epoch": 0.87,
"learning_rate": 4.5039624302905786e-05,
"loss": 0.0143,
"step": 2070
},
{
"epoch": 0.87,
"learning_rate": 4.500293513354858e-05,
"loss": 0.0128,
"step": 2080
},
{
"epoch": 0.87,
"learning_rate": 4.496624596419137e-05,
"loss": 0.0144,
"step": 2090
},
{
"epoch": 0.88,
"learning_rate": 4.4929556794834166e-05,
"loss": 0.0361,
"step": 2100
},
{
"epoch": 0.88,
"learning_rate": 4.489286762547696e-05,
"loss": 0.0085,
"step": 2110
},
{
"epoch": 0.89,
"learning_rate": 4.485617845611976e-05,
"loss": 0.0265,
"step": 2120
},
{
"epoch": 0.89,
"learning_rate": 4.4819489286762554e-05,
"loss": 0.0178,
"step": 2130
},
{
"epoch": 0.9,
"learning_rate": 4.478280011740534e-05,
"loss": 0.0355,
"step": 2140
},
{
"epoch": 0.9,
"learning_rate": 4.4746110948048134e-05,
"loss": 0.0227,
"step": 2150
},
{
"epoch": 0.9,
"learning_rate": 4.4709421778690934e-05,
"loss": 0.0238,
"step": 2160
},
{
"epoch": 0.91,
"learning_rate": 4.467273260933373e-05,
"loss": 0.0085,
"step": 2170
},
{
"epoch": 0.91,
"learning_rate": 4.463604343997652e-05,
"loss": 0.0039,
"step": 2180
},
{
"epoch": 0.92,
"learning_rate": 4.4599354270619315e-05,
"loss": 0.0326,
"step": 2190
},
{
"epoch": 0.92,
"learning_rate": 4.456266510126211e-05,
"loss": 0.01,
"step": 2200
},
{
"epoch": 0.92,
"eval_f1": 0.9908739561084989,
"eval_f1_0": 0.9972950573865325,
"eval_f1_1": 0.9961871658563921,
"eval_f1_2": 0.9742525085994586,
"eval_f1_3": 0.9957610925916127,
"eval_loss": 0.0161944180727005,
"eval_runtime": 32.5605,
"eval_samples_per_second": 261.114,
"eval_steps_per_second": 1.044,
"step": 2200
},
{
"epoch": 0.92,
"learning_rate": 4.452597593190491e-05,
"loss": 0.0145,
"step": 2210
},
{
"epoch": 0.93,
"learning_rate": 4.44892867625477e-05,
"loss": 0.0097,
"step": 2220
},
{
"epoch": 0.93,
"learning_rate": 4.4452597593190495e-05,
"loss": 0.0152,
"step": 2230
},
{
"epoch": 0.94,
"learning_rate": 4.441590842383329e-05,
"loss": 0.0174,
"step": 2240
},
{
"epoch": 0.94,
"learning_rate": 4.4379219254476076e-05,
"loss": 0.0321,
"step": 2250
},
{
"epoch": 0.95,
"learning_rate": 4.4342530085118876e-05,
"loss": 0.008,
"step": 2260
},
{
"epoch": 0.95,
"learning_rate": 4.430584091576167e-05,
"loss": 0.0071,
"step": 2270
},
{
"epoch": 0.95,
"learning_rate": 4.426915174640446e-05,
"loss": 0.017,
"step": 2280
},
{
"epoch": 0.96,
"learning_rate": 4.4232462577047257e-05,
"loss": 0.0225,
"step": 2290
},
{
"epoch": 0.96,
"learning_rate": 4.419577340769005e-05,
"loss": 0.0261,
"step": 2300
},
{
"epoch": 0.97,
"learning_rate": 4.415908423833285e-05,
"loss": 0.0236,
"step": 2310
},
{
"epoch": 0.97,
"learning_rate": 4.4122395068975644e-05,
"loss": 0.0082,
"step": 2320
},
{
"epoch": 0.97,
"learning_rate": 4.408570589961844e-05,
"loss": 0.0162,
"step": 2330
},
{
"epoch": 0.98,
"learning_rate": 4.404901673026123e-05,
"loss": 0.028,
"step": 2340
},
{
"epoch": 0.98,
"learning_rate": 4.401232756090402e-05,
"loss": 0.0268,
"step": 2350
},
{
"epoch": 0.99,
"learning_rate": 4.397563839154682e-05,
"loss": 0.033,
"step": 2360
},
{
"epoch": 0.99,
"learning_rate": 4.393894922218961e-05,
"loss": 0.0262,
"step": 2370
},
{
"epoch": 1.0,
"learning_rate": 4.3902260052832405e-05,
"loss": 0.0286,
"step": 2380
},
{
"epoch": 1.0,
"learning_rate": 4.38655708834752e-05,
"loss": 0.0074,
"step": 2390
},
{
"epoch": 1.0,
"learning_rate": 4.382888171411799e-05,
"loss": 0.0051,
"step": 2400
},
{
"epoch": 1.0,
"eval_f1": 0.9880051721937252,
"eval_f1_0": 0.9964674585868089,
"eval_f1_1": 0.9961339401451428,
"eval_f1_2": 0.9608874677337239,
"eval_f1_3": 0.9985318223092253,
"eval_loss": 0.02136407606303692,
"eval_runtime": 32.4029,
"eval_samples_per_second": 262.384,
"eval_steps_per_second": 1.049,
"step": 2400
},
{
"epoch": 1.01,
"learning_rate": 4.379219254476079e-05,
"loss": 0.0086,
"step": 2410
},
{
"epoch": 1.01,
"learning_rate": 4.3755503375403586e-05,
"loss": 0.0149,
"step": 2420
},
{
"epoch": 1.02,
"learning_rate": 4.371881420604638e-05,
"loss": 0.0089,
"step": 2430
},
{
"epoch": 1.02,
"learning_rate": 4.368212503668917e-05,
"loss": 0.0125,
"step": 2440
},
{
"epoch": 1.02,
"learning_rate": 4.3645435867331966e-05,
"loss": 0.0087,
"step": 2450
},
{
"epoch": 1.03,
"learning_rate": 4.360874669797476e-05,
"loss": 0.0174,
"step": 2460
},
{
"epoch": 1.03,
"learning_rate": 4.357205752861755e-05,
"loss": 0.0235,
"step": 2470
},
{
"epoch": 1.04,
"learning_rate": 4.353536835926035e-05,
"loss": 0.0208,
"step": 2480
},
{
"epoch": 1.04,
"learning_rate": 4.349867918990314e-05,
"loss": 0.0323,
"step": 2490
},
{
"epoch": 1.05,
"learning_rate": 4.346199002054594e-05,
"loss": 0.0152,
"step": 2500
},
{
"epoch": 1.05,
"learning_rate": 4.3425300851188734e-05,
"loss": 0.0206,
"step": 2510
},
{
"epoch": 1.05,
"learning_rate": 4.338861168183153e-05,
"loss": 0.0115,
"step": 2520
},
{
"epoch": 1.06,
"learning_rate": 4.335192251247432e-05,
"loss": 0.0059,
"step": 2530
},
{
"epoch": 1.06,
"learning_rate": 4.3315233343117114e-05,
"loss": 0.0073,
"step": 2540
},
{
"epoch": 1.07,
"learning_rate": 4.327854417375991e-05,
"loss": 0.0041,
"step": 2550
},
{
"epoch": 1.07,
"learning_rate": 4.32418550044027e-05,
"loss": 0.0035,
"step": 2560
},
{
"epoch": 1.07,
"learning_rate": 4.3205165835045495e-05,
"loss": 0.0168,
"step": 2570
},
{
"epoch": 1.08,
"learning_rate": 4.316847666568829e-05,
"loss": 0.0069,
"step": 2580
},
{
"epoch": 1.08,
"learning_rate": 4.313178749633108e-05,
"loss": 0.0108,
"step": 2590
},
{
"epoch": 1.09,
"learning_rate": 4.309509832697388e-05,
"loss": 0.0062,
"step": 2600
},
{
"epoch": 1.09,
"eval_f1": 0.992929983091436,
"eval_f1_0": 0.997889318599733,
"eval_f1_1": 0.9965863061785774,
"eval_f1_2": 0.9788712248150722,
"eval_f1_3": 0.9983730827723613,
"eval_loss": 0.014554007910192013,
"eval_runtime": 32.4876,
"eval_samples_per_second": 261.7,
"eval_steps_per_second": 1.047,
"step": 2600
},
{
"epoch": 1.09,
"learning_rate": 4.3058409157616676e-05,
"loss": 0.0089,
"step": 2610
},
{
"epoch": 1.1,
"learning_rate": 4.302171998825947e-05,
"loss": 0.005,
"step": 2620
},
{
"epoch": 1.1,
"learning_rate": 4.298503081890226e-05,
"loss": 0.0033,
"step": 2630
},
{
"epoch": 1.1,
"learning_rate": 4.2948341649545056e-05,
"loss": 0.0162,
"step": 2640
},
{
"epoch": 1.11,
"learning_rate": 4.291165248018785e-05,
"loss": 0.0036,
"step": 2650
},
{
"epoch": 1.11,
"learning_rate": 4.287496331083064e-05,
"loss": 0.0407,
"step": 2660
},
{
"epoch": 1.12,
"learning_rate": 4.283827414147344e-05,
"loss": 0.0058,
"step": 2670
},
{
"epoch": 1.12,
"learning_rate": 4.280158497211623e-05,
"loss": 0.0089,
"step": 2680
},
{
"epoch": 1.13,
"learning_rate": 4.2764895802759024e-05,
"loss": 0.0046,
"step": 2690
},
{
"epoch": 1.13,
"learning_rate": 4.2728206633401824e-05,
"loss": 0.0059,
"step": 2700
},
{
"epoch": 1.13,
"learning_rate": 4.269151746404462e-05,
"loss": 0.0089,
"step": 2710
},
{
"epoch": 1.14,
"learning_rate": 4.265482829468741e-05,
"loss": 0.0124,
"step": 2720
},
{
"epoch": 1.14,
"learning_rate": 4.2618139125330205e-05,
"loss": 0.0251,
"step": 2730
},
{
"epoch": 1.15,
"learning_rate": 4.2581449955973e-05,
"loss": 0.0057,
"step": 2740
},
{
"epoch": 1.15,
"learning_rate": 4.254476078661579e-05,
"loss": 0.0068,
"step": 2750
},
{
"epoch": 1.15,
"learning_rate": 4.2508071617258585e-05,
"loss": 0.0217,
"step": 2760
},
{
"epoch": 1.16,
"learning_rate": 4.247138244790138e-05,
"loss": 0.0053,
"step": 2770
},
{
"epoch": 1.16,
"learning_rate": 4.243469327854417e-05,
"loss": 0.0059,
"step": 2780
},
{
"epoch": 1.17,
"learning_rate": 4.2398004109186966e-05,
"loss": 0.0034,
"step": 2790
},
{
"epoch": 1.17,
"learning_rate": 4.2361314939829766e-05,
"loss": 0.0126,
"step": 2800
},
{
"epoch": 1.17,
"eval_f1": 0.9924767659206969,
"eval_f1_0": 0.9977185341990826,
"eval_f1_1": 0.996399014227777,
"eval_f1_2": 0.9773225070072529,
"eval_f1_3": 0.9984670082486748,
"eval_loss": 0.01500015240162611,
"eval_runtime": 32.5538,
"eval_samples_per_second": 261.167,
"eval_steps_per_second": 1.044,
"step": 2800
},
{
"epoch": 1.18,
"learning_rate": 4.232462577047256e-05,
"loss": 0.0036,
"step": 2810
},
{
"epoch": 1.18,
"learning_rate": 4.228793660111535e-05,
"loss": 0.0053,
"step": 2820
},
{
"epoch": 1.18,
"learning_rate": 4.2251247431758146e-05,
"loss": 0.0273,
"step": 2830
},
{
"epoch": 1.19,
"learning_rate": 4.221455826240095e-05,
"loss": 0.0038,
"step": 2840
},
{
"epoch": 1.19,
"learning_rate": 4.217786909304374e-05,
"loss": 0.0087,
"step": 2850
},
{
"epoch": 1.2,
"learning_rate": 4.214117992368653e-05,
"loss": 0.0076,
"step": 2860
},
{
"epoch": 1.2,
"learning_rate": 4.210449075432932e-05,
"loss": 0.0442,
"step": 2870
},
{
"epoch": 1.2,
"learning_rate": 4.2067801584972114e-05,
"loss": 0.0278,
"step": 2880
},
{
"epoch": 1.21,
"learning_rate": 4.2031112415614914e-05,
"loss": 0.0341,
"step": 2890
},
{
"epoch": 1.21,
"learning_rate": 4.199442324625771e-05,
"loss": 0.0076,
"step": 2900
},
{
"epoch": 1.22,
"learning_rate": 4.19577340769005e-05,
"loss": 0.0051,
"step": 2910
},
{
"epoch": 1.22,
"learning_rate": 4.1921044907543295e-05,
"loss": 0.0105,
"step": 2920
},
{
"epoch": 1.23,
"learning_rate": 4.188435573818609e-05,
"loss": 0.0067,
"step": 2930
},
{
"epoch": 1.23,
"learning_rate": 4.184766656882889e-05,
"loss": 0.0052,
"step": 2940
},
{
"epoch": 1.23,
"learning_rate": 4.181097739947168e-05,
"loss": 0.0446,
"step": 2950
},
{
"epoch": 1.24,
"learning_rate": 4.1774288230114476e-05,
"loss": 0.0113,
"step": 2960
},
{
"epoch": 1.24,
"learning_rate": 4.173759906075726e-05,
"loss": 0.0047,
"step": 2970
},
{
"epoch": 1.25,
"learning_rate": 4.1700909891400056e-05,
"loss": 0.005,
"step": 2980
},
{
"epoch": 1.25,
"learning_rate": 4.1664220722042856e-05,
"loss": 0.0159,
"step": 2990
},
{
"epoch": 1.25,
"learning_rate": 4.162753155268565e-05,
"loss": 0.0047,
"step": 3000
},
{
"epoch": 1.25,
"eval_f1": 0.9924093763184394,
"eval_f1_0": 0.9977303468144987,
"eval_f1_1": 0.9967006575956847,
"eval_f1_2": 0.9765950171557934,
"eval_f1_3": 0.9986114837077807,
"eval_loss": 0.01629544049501419,
"eval_runtime": 32.5008,
"eval_samples_per_second": 261.594,
"eval_steps_per_second": 1.046,
"step": 3000
},
{
"epoch": 1.26,
"learning_rate": 4.159084238332844e-05,
"loss": 0.012,
"step": 3010
},
{
"epoch": 1.26,
"learning_rate": 4.1554153213971237e-05,
"loss": 0.0108,
"step": 3020
},
{
"epoch": 1.27,
"learning_rate": 4.151746404461403e-05,
"loss": 0.007,
"step": 3030
},
{
"epoch": 1.27,
"learning_rate": 4.148077487525683e-05,
"loss": 0.0089,
"step": 3040
},
{
"epoch": 1.28,
"learning_rate": 4.1444085705899624e-05,
"loss": 0.015,
"step": 3050
},
{
"epoch": 1.28,
"learning_rate": 4.140739653654242e-05,
"loss": 0.0185,
"step": 3060
},
{
"epoch": 1.28,
"learning_rate": 4.137070736718521e-05,
"loss": 0.0094,
"step": 3070
},
{
"epoch": 1.29,
"learning_rate": 4.1334018197828e-05,
"loss": 0.0116,
"step": 3080
},
{
"epoch": 1.29,
"learning_rate": 4.12973290284708e-05,
"loss": 0.0072,
"step": 3090
},
{
"epoch": 1.3,
"learning_rate": 4.126063985911359e-05,
"loss": 0.0075,
"step": 3100
},
{
"epoch": 1.3,
"learning_rate": 4.1223950689756385e-05,
"loss": 0.0134,
"step": 3110
},
{
"epoch": 1.3,
"learning_rate": 4.118726152039918e-05,
"loss": 0.0223,
"step": 3120
},
{
"epoch": 1.31,
"learning_rate": 4.115057235104197e-05,
"loss": 0.01,
"step": 3130
},
{
"epoch": 1.31,
"learning_rate": 4.111388318168477e-05,
"loss": 0.0075,
"step": 3140
},
{
"epoch": 1.32,
"learning_rate": 4.1077194012327566e-05,
"loss": 0.0081,
"step": 3150
},
{
"epoch": 1.32,
"learning_rate": 4.104050484297036e-05,
"loss": 0.0055,
"step": 3160
},
{
"epoch": 1.33,
"learning_rate": 4.100381567361315e-05,
"loss": 0.0089,
"step": 3170
},
{
"epoch": 1.33,
"learning_rate": 4.0967126504255946e-05,
"loss": 0.0081,
"step": 3180
},
{
"epoch": 1.33,
"learning_rate": 4.093043733489874e-05,
"loss": 0.0054,
"step": 3190
},
{
"epoch": 1.34,
"learning_rate": 4.089374816554153e-05,
"loss": 0.0057,
"step": 3200
},
{
"epoch": 1.34,
"eval_f1": 0.9936108701366878,
"eval_f1_0": 0.9980681267132205,
"eval_f1_1": 0.9966263673435388,
"eval_f1_2": 0.9811435221508926,
"eval_f1_3": 0.9986054643390991,
"eval_loss": 0.012877003289759159,
"eval_runtime": 32.5902,
"eval_samples_per_second": 260.876,
"eval_steps_per_second": 1.043,
"step": 3200
},
{
"epoch": 1.34,
"learning_rate": 4.085705899618433e-05,
"loss": 0.0056,
"step": 3210
},
{
"epoch": 1.35,
"learning_rate": 4.082036982682712e-05,
"loss": 0.0074,
"step": 3220
},
{
"epoch": 1.35,
"learning_rate": 4.078368065746992e-05,
"loss": 0.031,
"step": 3230
},
{
"epoch": 1.36,
"learning_rate": 4.0746991488112714e-05,
"loss": 0.0066,
"step": 3240
},
{
"epoch": 1.36,
"learning_rate": 4.071030231875551e-05,
"loss": 0.0111,
"step": 3250
},
{
"epoch": 1.36,
"learning_rate": 4.06736131493983e-05,
"loss": 0.0111,
"step": 3260
},
{
"epoch": 1.37,
"learning_rate": 4.0636923980041094e-05,
"loss": 0.0226,
"step": 3270
},
{
"epoch": 1.37,
"learning_rate": 4.060023481068389e-05,
"loss": 0.0021,
"step": 3280
},
{
"epoch": 1.38,
"learning_rate": 4.056354564132668e-05,
"loss": 0.004,
"step": 3290
},
{
"epoch": 1.38,
"learning_rate": 4.0526856471969475e-05,
"loss": 0.0031,
"step": 3300
},
{
"epoch": 1.38,
"learning_rate": 4.049016730261227e-05,
"loss": 0.0068,
"step": 3310
},
{
"epoch": 1.39,
"learning_rate": 4.045347813325506e-05,
"loss": 0.0178,
"step": 3320
},
{
"epoch": 1.39,
"learning_rate": 4.041678896389786e-05,
"loss": 0.006,
"step": 3330
},
{
"epoch": 1.4,
"learning_rate": 4.0380099794540656e-05,
"loss": 0.0052,
"step": 3340
},
{
"epoch": 1.4,
"learning_rate": 4.034341062518345e-05,
"loss": 0.0247,
"step": 3350
},
{
"epoch": 1.41,
"learning_rate": 4.030672145582624e-05,
"loss": 0.0068,
"step": 3360
},
{
"epoch": 1.41,
"learning_rate": 4.0270032286469036e-05,
"loss": 0.0159,
"step": 3370
},
{
"epoch": 1.41,
"learning_rate": 4.023334311711183e-05,
"loss": 0.0035,
"step": 3380
},
{
"epoch": 1.42,
"learning_rate": 4.019665394775462e-05,
"loss": 0.022,
"step": 3390
},
{
"epoch": 1.42,
"learning_rate": 4.015996477839742e-05,
"loss": 0.0054,
"step": 3400
},
{
"epoch": 1.42,
"eval_f1": 0.9920029780078923,
"eval_f1_0": 0.9976414763783462,
"eval_f1_1": 0.9969571440759103,
"eval_f1_2": 0.9748046205322835,
"eval_f1_3": 0.9986086710450295,
"eval_loss": 0.016329294070601463,
"eval_runtime": 32.6588,
"eval_samples_per_second": 260.328,
"eval_steps_per_second": 1.041,
"step": 3400
},
{
"epoch": 1.43,
"learning_rate": 4.012327560904021e-05,
"loss": 0.0062,
"step": 3410
},
{
"epoch": 1.43,
"learning_rate": 4.0086586439683004e-05,
"loss": 0.0088,
"step": 3420
},
{
"epoch": 1.43,
"learning_rate": 4.0049897270325804e-05,
"loss": 0.01,
"step": 3430
},
{
"epoch": 1.44,
"learning_rate": 4.00132081009686e-05,
"loss": 0.0066,
"step": 3440
},
{
"epoch": 1.44,
"learning_rate": 3.997651893161139e-05,
"loss": 0.0052,
"step": 3450
},
{
"epoch": 1.45,
"learning_rate": 3.9939829762254185e-05,
"loss": 0.0043,
"step": 3460
},
{
"epoch": 1.45,
"learning_rate": 3.990314059289698e-05,
"loss": 0.0087,
"step": 3470
},
{
"epoch": 1.46,
"learning_rate": 3.986645142353977e-05,
"loss": 0.0139,
"step": 3480
},
{
"epoch": 1.46,
"learning_rate": 3.9829762254182565e-05,
"loss": 0.009,
"step": 3490
},
{
"epoch": 1.46,
"learning_rate": 3.979307308482536e-05,
"loss": 0.0058,
"step": 3500
},
{
"epoch": 1.47,
"learning_rate": 3.975638391546815e-05,
"loss": 0.0046,
"step": 3510
},
{
"epoch": 1.47,
"learning_rate": 3.971969474611095e-05,
"loss": 0.0052,
"step": 3520
},
{
"epoch": 1.48,
"learning_rate": 3.9683005576753746e-05,
"loss": 0.0062,
"step": 3530
},
{
"epoch": 1.48,
"learning_rate": 3.964631640739654e-05,
"loss": 0.0101,
"step": 3540
},
{
"epoch": 1.48,
"learning_rate": 3.960962723803933e-05,
"loss": 0.011,
"step": 3550
},
{
"epoch": 1.49,
"learning_rate": 3.9572938068682126e-05,
"loss": 0.0071,
"step": 3560
},
{
"epoch": 1.49,
"learning_rate": 3.953624889932493e-05,
"loss": 0.0121,
"step": 3570
},
{
"epoch": 1.5,
"learning_rate": 3.9499559729967713e-05,
"loss": 0.0095,
"step": 3580
},
{
"epoch": 1.5,
"learning_rate": 3.946287056061051e-05,
"loss": 0.0039,
"step": 3590
},
{
"epoch": 1.51,
"learning_rate": 3.94261813912533e-05,
"loss": 0.004,
"step": 3600
},
{
"epoch": 1.51,
"eval_f1": 0.9940251012499961,
"eval_f1_0": 0.998198324154023,
"eval_f1_1": 0.996981634010334,
"eval_f1_2": 0.9821706606103885,
"eval_f1_3": 0.998749786225239,
"eval_loss": 0.012289030477404594,
"eval_runtime": 32.6576,
"eval_samples_per_second": 260.337,
"eval_steps_per_second": 1.041,
"step": 3600
},
{
"epoch": 1.51,
"learning_rate": 3.9389492221896094e-05,
"loss": 0.0225,
"step": 3610
},
{
"epoch": 1.51,
"learning_rate": 3.9352803052538894e-05,
"loss": 0.0079,
"step": 3620
},
{
"epoch": 1.52,
"learning_rate": 3.931611388318169e-05,
"loss": 0.0037,
"step": 3630
},
{
"epoch": 1.52,
"learning_rate": 3.927942471382448e-05,
"loss": 0.009,
"step": 3640
},
{
"epoch": 1.53,
"learning_rate": 3.9242735544467275e-05,
"loss": 0.0031,
"step": 3650
},
{
"epoch": 1.53,
"learning_rate": 3.920604637511007e-05,
"loss": 0.0081,
"step": 3660
},
{
"epoch": 1.53,
"learning_rate": 3.916935720575287e-05,
"loss": 0.009,
"step": 3670
},
{
"epoch": 1.54,
"learning_rate": 3.913266803639566e-05,
"loss": 0.0216,
"step": 3680
},
{
"epoch": 1.54,
"learning_rate": 3.909597886703845e-05,
"loss": 0.004,
"step": 3690
},
{
"epoch": 1.55,
"learning_rate": 3.905928969768124e-05,
"loss": 0.0292,
"step": 3700
},
{
"epoch": 1.55,
"learning_rate": 3.9022600528324036e-05,
"loss": 0.0062,
"step": 3710
},
{
"epoch": 1.56,
"learning_rate": 3.8985911358966836e-05,
"loss": 0.0065,
"step": 3720
},
{
"epoch": 1.56,
"learning_rate": 3.894922218960963e-05,
"loss": 0.0026,
"step": 3730
},
{
"epoch": 1.56,
"learning_rate": 3.891253302025242e-05,
"loss": 0.0126,
"step": 3740
},
{
"epoch": 1.57,
"learning_rate": 3.8875843850895217e-05,
"loss": 0.0049,
"step": 3750
},
{
"epoch": 1.57,
"learning_rate": 3.883915468153801e-05,
"loss": 0.0237,
"step": 3760
},
{
"epoch": 1.58,
"learning_rate": 3.880246551218081e-05,
"loss": 0.0038,
"step": 3770
},
{
"epoch": 1.58,
"learning_rate": 3.8765776342823604e-05,
"loss": 0.0085,
"step": 3780
},
{
"epoch": 1.59,
"learning_rate": 3.87290871734664e-05,
"loss": 0.0033,
"step": 3790
},
{
"epoch": 1.59,
"learning_rate": 3.8692398004109184e-05,
"loss": 0.0299,
"step": 3800
},
{
"epoch": 1.59,
"eval_f1": 0.9932776161516476,
"eval_f1_0": 0.9979713691118977,
"eval_f1_1": 0.9971163963485729,
"eval_f1_2": 0.9792434123492829,
"eval_f1_3": 0.9987792867968368,
"eval_loss": 0.012060901150107384,
"eval_runtime": 32.7616,
"eval_samples_per_second": 259.511,
"eval_steps_per_second": 1.038,
"step": 3800
},
{
"epoch": 1.59,
"learning_rate": 3.8655708834751984e-05,
"loss": 0.0158,
"step": 3810
},
{
"epoch": 1.6,
"learning_rate": 3.861901966539478e-05,
"loss": 0.0309,
"step": 3820
},
{
"epoch": 1.6,
"learning_rate": 3.858233049603757e-05,
"loss": 0.0032,
"step": 3830
},
{
"epoch": 1.61,
"learning_rate": 3.8545641326680365e-05,
"loss": 0.0257,
"step": 3840
},
{
"epoch": 1.61,
"learning_rate": 3.850895215732316e-05,
"loss": 0.0038,
"step": 3850
},
{
"epoch": 1.61,
"learning_rate": 3.847226298796596e-05,
"loss": 0.011,
"step": 3860
},
{
"epoch": 1.62,
"learning_rate": 3.843557381860875e-05,
"loss": 0.004,
"step": 3870
},
{
"epoch": 1.62,
"learning_rate": 3.8398884649251546e-05,
"loss": 0.0176,
"step": 3880
},
{
"epoch": 1.63,
"learning_rate": 3.836219547989434e-05,
"loss": 0.0047,
"step": 3890
},
{
"epoch": 1.63,
"learning_rate": 3.8325506310537126e-05,
"loss": 0.0139,
"step": 3900
},
{
"epoch": 1.64,
"learning_rate": 3.8288817141179926e-05,
"loss": 0.0037,
"step": 3910
},
{
"epoch": 1.64,
"learning_rate": 3.825212797182272e-05,
"loss": 0.0176,
"step": 3920
},
{
"epoch": 1.64,
"learning_rate": 3.821543880246551e-05,
"loss": 0.0054,
"step": 3930
},
{
"epoch": 1.65,
"learning_rate": 3.817874963310831e-05,
"loss": 0.0061,
"step": 3940
},
{
"epoch": 1.65,
"learning_rate": 3.81420604637511e-05,
"loss": 0.0273,
"step": 3950
},
{
"epoch": 1.66,
"learning_rate": 3.81053712943939e-05,
"loss": 0.0045,
"step": 3960
},
{
"epoch": 1.66,
"learning_rate": 3.8068682125036694e-05,
"loss": 0.0301,
"step": 3970
},
{
"epoch": 1.66,
"learning_rate": 3.803199295567949e-05,
"loss": 0.0108,
"step": 3980
},
{
"epoch": 1.67,
"learning_rate": 3.799530378632228e-05,
"loss": 0.0091,
"step": 3990
},
{
"epoch": 1.67,
"learning_rate": 3.7958614616965075e-05,
"loss": 0.0052,
"step": 4000
},
{
"epoch": 1.67,
"eval_f1": 0.9934967532249497,
"eval_f1_0": 0.9980441662159518,
"eval_f1_1": 0.9970681600667103,
"eval_f1_2": 0.9801039664713919,
"eval_f1_3": 0.9987707201457449,
"eval_loss": 0.012285062111914158,
"eval_runtime": 32.7767,
"eval_samples_per_second": 259.391,
"eval_steps_per_second": 1.037,
"step": 4000
},
{
"epoch": 1.68,
"learning_rate": 3.792192544760787e-05,
"loss": 0.0122,
"step": 4010
},
{
"epoch": 1.68,
"learning_rate": 3.788523627825066e-05,
"loss": 0.004,
"step": 4020
},
{
"epoch": 1.69,
"learning_rate": 3.7848547108893455e-05,
"loss": 0.0103,
"step": 4030
},
{
"epoch": 1.69,
"learning_rate": 3.781185793953625e-05,
"loss": 0.0058,
"step": 4040
},
{
"epoch": 1.69,
"learning_rate": 3.777516877017904e-05,
"loss": 0.004,
"step": 4050
},
{
"epoch": 1.7,
"learning_rate": 3.773847960082184e-05,
"loss": 0.0037,
"step": 4060
},
{
"epoch": 1.7,
"learning_rate": 3.7701790431464636e-05,
"loss": 0.0087,
"step": 4070
},
{
"epoch": 1.71,
"learning_rate": 3.766510126210743e-05,
"loss": 0.0073,
"step": 4080
},
{
"epoch": 1.71,
"learning_rate": 3.762841209275022e-05,
"loss": 0.0299,
"step": 4090
},
{
"epoch": 1.71,
"learning_rate": 3.7591722923393016e-05,
"loss": 0.0033,
"step": 4100
},
{
"epoch": 1.72,
"learning_rate": 3.755503375403581e-05,
"loss": 0.0043,
"step": 4110
},
{
"epoch": 1.72,
"learning_rate": 3.75183445846786e-05,
"loss": 0.0092,
"step": 4120
},
{
"epoch": 1.73,
"learning_rate": 3.74816554153214e-05,
"loss": 0.0055,
"step": 4130
},
{
"epoch": 1.73,
"learning_rate": 3.744496624596419e-05,
"loss": 0.0069,
"step": 4140
},
{
"epoch": 1.74,
"learning_rate": 3.740827707660699e-05,
"loss": 0.0059,
"step": 4150
},
{
"epoch": 1.74,
"learning_rate": 3.7371587907249784e-05,
"loss": 0.0081,
"step": 4160
},
{
"epoch": 1.74,
"learning_rate": 3.733489873789258e-05,
"loss": 0.0083,
"step": 4170
},
{
"epoch": 1.75,
"learning_rate": 3.729820956853537e-05,
"loss": 0.0109,
"step": 4180
},
{
"epoch": 1.75,
"learning_rate": 3.7261520399178165e-05,
"loss": 0.0163,
"step": 4190
},
{
"epoch": 1.76,
"learning_rate": 3.722483122982096e-05,
"loss": 0.0265,
"step": 4200
},
{
"epoch": 1.76,
"eval_f1": 0.9940958225544244,
"eval_f1_0": 0.9982137368330962,
"eval_f1_1": 0.9971425756232141,
"eval_f1_2": 0.9822802768870269,
"eval_f1_3": 0.9987467008743605,
"eval_loss": 0.011539779603481293,
"eval_runtime": 32.5807,
"eval_samples_per_second": 260.952,
"eval_steps_per_second": 1.044,
"step": 4200
},
{
"epoch": 1.76,
"learning_rate": 3.718814206046375e-05,
"loss": 0.0104,
"step": 4210
},
{
"epoch": 1.76,
"learning_rate": 3.7151452891106545e-05,
"loss": 0.0045,
"step": 4220
},
{
"epoch": 1.77,
"learning_rate": 3.711476372174934e-05,
"loss": 0.024,
"step": 4230
},
{
"epoch": 1.77,
"learning_rate": 3.707807455239213e-05,
"loss": 0.0058,
"step": 4240
},
{
"epoch": 1.78,
"learning_rate": 3.704138538303493e-05,
"loss": 0.0254,
"step": 4250
},
{
"epoch": 1.78,
"learning_rate": 3.7004696213677726e-05,
"loss": 0.0365,
"step": 4260
},
{
"epoch": 1.79,
"learning_rate": 3.696800704432052e-05,
"loss": 0.0147,
"step": 4270
},
{
"epoch": 1.79,
"learning_rate": 3.693131787496331e-05,
"loss": 0.0033,
"step": 4280
},
{
"epoch": 1.79,
"learning_rate": 3.6894628705606106e-05,
"loss": 0.004,
"step": 4290
},
{
"epoch": 1.8,
"learning_rate": 3.68579395362489e-05,
"loss": 0.0039,
"step": 4300
},
{
"epoch": 1.8,
"learning_rate": 3.6821250366891693e-05,
"loss": 0.031,
"step": 4310
},
{
"epoch": 1.81,
"learning_rate": 3.678456119753449e-05,
"loss": 0.0065,
"step": 4320
},
{
"epoch": 1.81,
"learning_rate": 3.674787202817728e-05,
"loss": 0.0035,
"step": 4330
},
{
"epoch": 1.82,
"learning_rate": 3.6711182858820074e-05,
"loss": 0.0031,
"step": 4340
},
{
"epoch": 1.82,
"learning_rate": 3.6674493689462874e-05,
"loss": 0.0035,
"step": 4350
},
{
"epoch": 1.82,
"learning_rate": 3.663780452010567e-05,
"loss": 0.0036,
"step": 4360
},
{
"epoch": 1.83,
"learning_rate": 3.660111535074846e-05,
"loss": 0.0028,
"step": 4370
},
{
"epoch": 1.83,
"learning_rate": 3.6564426181391255e-05,
"loss": 0.0095,
"step": 4380
},
{
"epoch": 1.84,
"learning_rate": 3.652773701203405e-05,
"loss": 0.0047,
"step": 4390
},
{
"epoch": 1.84,
"learning_rate": 3.649104784267685e-05,
"loss": 0.0056,
"step": 4400
},
{
"epoch": 1.84,
"eval_f1": 0.994136557447434,
"eval_f1_0": 0.9982309070139049,
"eval_f1_1": 0.9970766826126084,
"eval_f1_2": 0.9825211109980669,
"eval_f1_3": 0.9987175291651557,
"eval_loss": 0.011068220250308514,
"eval_runtime": 32.4093,
"eval_samples_per_second": 262.332,
"eval_steps_per_second": 1.049,
"step": 4400
},
{
"epoch": 1.84,
"learning_rate": 3.6454358673319635e-05,
"loss": 0.0378,
"step": 4410
},
{
"epoch": 1.85,
"learning_rate": 3.641766950396243e-05,
"loss": 0.0029,
"step": 4420
},
{
"epoch": 1.85,
"learning_rate": 3.638098033460522e-05,
"loss": 0.0175,
"step": 4430
},
{
"epoch": 1.86,
"learning_rate": 3.6344291165248016e-05,
"loss": 0.0108,
"step": 4440
},
{
"epoch": 1.86,
"learning_rate": 3.6307601995890816e-05,
"loss": 0.0042,
"step": 4450
},
{
"epoch": 1.87,
"learning_rate": 3.627091282653361e-05,
"loss": 0.0142,
"step": 4460
},
{
"epoch": 1.87,
"learning_rate": 3.62342236571764e-05,
"loss": 0.003,
"step": 4470
},
{
"epoch": 1.87,
"learning_rate": 3.61975344878192e-05,
"loss": 0.0041,
"step": 4480
},
{
"epoch": 1.88,
"learning_rate": 3.6160845318462e-05,
"loss": 0.0171,
"step": 4490
},
{
"epoch": 1.88,
"learning_rate": 3.612415614910479e-05,
"loss": 0.013,
"step": 4500
},
{
"epoch": 1.89,
"learning_rate": 3.6087466979747584e-05,
"loss": 0.0136,
"step": 4510
},
{
"epoch": 1.89,
"learning_rate": 3.605077781039037e-05,
"loss": 0.0128,
"step": 4520
},
{
"epoch": 1.89,
"learning_rate": 3.6014088641033164e-05,
"loss": 0.0033,
"step": 4530
},
{
"epoch": 1.9,
"learning_rate": 3.5977399471675964e-05,
"loss": 0.0073,
"step": 4540
},
{
"epoch": 1.9,
"learning_rate": 3.594071030231876e-05,
"loss": 0.0342,
"step": 4550
},
{
"epoch": 1.91,
"learning_rate": 3.590402113296155e-05,
"loss": 0.0042,
"step": 4560
},
{
"epoch": 1.91,
"learning_rate": 3.5867331963604345e-05,
"loss": 0.006,
"step": 4570
},
{
"epoch": 1.92,
"learning_rate": 3.583064279424714e-05,
"loss": 0.004,
"step": 4580
},
{
"epoch": 1.92,
"learning_rate": 3.579395362488994e-05,
"loss": 0.01,
"step": 4590
},
{
"epoch": 1.92,
"learning_rate": 3.575726445553273e-05,
"loss": 0.0085,
"step": 4600
},
{
"epoch": 1.92,
"eval_f1": 0.993926435925004,
"eval_f1_0": 0.9981605158099504,
"eval_f1_1": 0.9967335581819528,
"eval_f1_2": 0.9820825275519001,
"eval_f1_3": 0.9987291421562123,
"eval_loss": 0.01137313898652792,
"eval_runtime": 32.5633,
"eval_samples_per_second": 261.091,
"eval_steps_per_second": 1.044,
"step": 4600
},
{
"epoch": 1.93,
"learning_rate": 3.5720575286175526e-05,
"loss": 0.0068,
"step": 4610
},
{
"epoch": 1.93,
"learning_rate": 3.568388611681831e-05,
"loss": 0.0258,
"step": 4620
},
{
"epoch": 1.94,
"learning_rate": 3.5647196947461106e-05,
"loss": 0.0023,
"step": 4630
},
{
"epoch": 1.94,
"learning_rate": 3.5610507778103906e-05,
"loss": 0.0041,
"step": 4640
},
{
"epoch": 1.94,
"learning_rate": 3.55738186087467e-05,
"loss": 0.0185,
"step": 4650
},
{
"epoch": 1.95,
"learning_rate": 3.553712943938949e-05,
"loss": 0.0111,
"step": 4660
},
{
"epoch": 1.95,
"learning_rate": 3.550044027003229e-05,
"loss": 0.0045,
"step": 4670
},
{
"epoch": 1.96,
"learning_rate": 3.546375110067508e-05,
"loss": 0.011,
"step": 4680
},
{
"epoch": 1.96,
"learning_rate": 3.542706193131788e-05,
"loss": 0.0055,
"step": 4690
},
{
"epoch": 1.97,
"learning_rate": 3.5390372761960674e-05,
"loss": 0.0036,
"step": 4700
},
{
"epoch": 1.97,
"learning_rate": 3.535368359260347e-05,
"loss": 0.0025,
"step": 4710
},
{
"epoch": 1.97,
"learning_rate": 3.531699442324626e-05,
"loss": 0.0262,
"step": 4720
},
{
"epoch": 1.98,
"learning_rate": 3.528030525388905e-05,
"loss": 0.0045,
"step": 4730
},
{
"epoch": 1.98,
"learning_rate": 3.524361608453185e-05,
"loss": 0.0037,
"step": 4740
},
{
"epoch": 1.99,
"learning_rate": 3.520692691517464e-05,
"loss": 0.0032,
"step": 4750
},
{
"epoch": 1.99,
"learning_rate": 3.5170237745817435e-05,
"loss": 0.0457,
"step": 4760
},
{
"epoch": 1.99,
"learning_rate": 3.513354857646023e-05,
"loss": 0.0044,
"step": 4770
},
{
"epoch": 2.0,
"learning_rate": 3.509685940710302e-05,
"loss": 0.0215,
"step": 4780
},
{
"epoch": 2.0,
"learning_rate": 3.506017023774582e-05,
"loss": 0.0023,
"step": 4790
},
{
"epoch": 2.01,
"learning_rate": 3.5023481068388616e-05,
"loss": 0.0066,
"step": 4800
},
{
"epoch": 2.01,
"eval_f1": 0.9946477369501043,
"eval_f1_0": 0.9983790031087706,
"eval_f1_1": 0.9971663656127568,
"eval_f1_2": 0.9842573857912964,
"eval_f1_3": 0.9987881932875933,
"eval_loss": 0.010337233543395996,
"eval_runtime": 32.5797,
"eval_samples_per_second": 260.96,
"eval_steps_per_second": 1.044,
"step": 4800
},
{
"epoch": 2.01,
"learning_rate": 3.498679189903141e-05,
"loss": 0.0031,
"step": 4810
},
{
"epoch": 2.02,
"learning_rate": 3.49501027296742e-05,
"loss": 0.0173,
"step": 4820
},
{
"epoch": 2.02,
"learning_rate": 3.4913413560316996e-05,
"loss": 0.0085,
"step": 4830
},
{
"epoch": 2.02,
"learning_rate": 3.487672439095979e-05,
"loss": 0.002,
"step": 4840
},
{
"epoch": 2.03,
"learning_rate": 3.4840035221602583e-05,
"loss": 0.0116,
"step": 4850
},
{
"epoch": 2.03,
"learning_rate": 3.480334605224538e-05,
"loss": 0.0031,
"step": 4860
},
{
"epoch": 2.04,
"learning_rate": 3.476665688288817e-05,
"loss": 0.0031,
"step": 4870
},
{
"epoch": 2.04,
"learning_rate": 3.472996771353097e-05,
"loss": 0.0152,
"step": 4880
},
{
"epoch": 2.05,
"learning_rate": 3.4693278544173764e-05,
"loss": 0.0028,
"step": 4890
},
{
"epoch": 2.05,
"learning_rate": 3.465658937481656e-05,
"loss": 0.0033,
"step": 4900
},
{
"epoch": 2.05,
"learning_rate": 3.461990020545935e-05,
"loss": 0.0104,
"step": 4910
},
{
"epoch": 2.06,
"learning_rate": 3.4583211036102145e-05,
"loss": 0.0243,
"step": 4920
},
{
"epoch": 2.06,
"learning_rate": 3.454652186674494e-05,
"loss": 0.017,
"step": 4930
},
{
"epoch": 2.07,
"learning_rate": 3.450983269738773e-05,
"loss": 0.005,
"step": 4940
},
{
"epoch": 2.07,
"learning_rate": 3.4473143528030525e-05,
"loss": 0.0102,
"step": 4950
},
{
"epoch": 2.07,
"learning_rate": 3.443645435867332e-05,
"loss": 0.0032,
"step": 4960
},
{
"epoch": 2.08,
"learning_rate": 3.439976518931611e-05,
"loss": 0.0082,
"step": 4970
},
{
"epoch": 2.08,
"learning_rate": 3.436307601995891e-05,
"loss": 0.0033,
"step": 4980
},
{
"epoch": 2.09,
"learning_rate": 3.4326386850601706e-05,
"loss": 0.0125,
"step": 4990
},
{
"epoch": 2.09,
"learning_rate": 3.42896976812445e-05,
"loss": 0.0028,
"step": 5000
},
{
"epoch": 2.09,
"eval_f1": 0.9943125530635006,
"eval_f1_0": 0.9982833052354131,
"eval_f1_1": 0.9971853943764748,
"eval_f1_2": 0.9828960308636031,
"eval_f1_3": 0.9988854817785117,
"eval_loss": 0.01153592299669981,
"eval_runtime": 32.4834,
"eval_samples_per_second": 261.734,
"eval_steps_per_second": 1.047,
"step": 5000
},
{
"epoch": 2.1,
"learning_rate": 3.425300851188729e-05,
"loss": 0.0032,
"step": 5010
},
{
"epoch": 2.1,
"learning_rate": 3.4216319342530087e-05,
"loss": 0.0023,
"step": 5020
},
{
"epoch": 2.1,
"learning_rate": 3.417963017317288e-05,
"loss": 0.0024,
"step": 5030
},
{
"epoch": 2.11,
"learning_rate": 3.4142941003815674e-05,
"loss": 0.0189,
"step": 5040
},
{
"epoch": 2.11,
"learning_rate": 3.410625183445847e-05,
"loss": 0.004,
"step": 5050
},
{
"epoch": 2.12,
"learning_rate": 3.406956266510126e-05,
"loss": 0.0038,
"step": 5060
},
{
"epoch": 2.12,
"learning_rate": 3.4032873495744054e-05,
"loss": 0.0026,
"step": 5070
},
{
"epoch": 2.12,
"learning_rate": 3.3996184326386854e-05,
"loss": 0.0097,
"step": 5080
},
{
"epoch": 2.13,
"learning_rate": 3.395949515702965e-05,
"loss": 0.0027,
"step": 5090
},
{
"epoch": 2.13,
"learning_rate": 3.392280598767244e-05,
"loss": 0.0194,
"step": 5100
},
{
"epoch": 2.14,
"learning_rate": 3.3886116818315235e-05,
"loss": 0.0025,
"step": 5110
},
{
"epoch": 2.14,
"learning_rate": 3.3849427648958035e-05,
"loss": 0.0024,
"step": 5120
},
{
"epoch": 2.15,
"learning_rate": 3.381273847960082e-05,
"loss": 0.003,
"step": 5130
},
{
"epoch": 2.15,
"learning_rate": 3.3776049310243615e-05,
"loss": 0.0053,
"step": 5140
},
{
"epoch": 2.15,
"learning_rate": 3.373936014088641e-05,
"loss": 0.0028,
"step": 5150
},
{
"epoch": 2.16,
"learning_rate": 3.37026709715292e-05,
"loss": 0.0038,
"step": 5160
},
{
"epoch": 2.16,
"learning_rate": 3.3665981802172e-05,
"loss": 0.0133,
"step": 5170
},
{
"epoch": 2.17,
"learning_rate": 3.3629292632814796e-05,
"loss": 0.0039,
"step": 5180
},
{
"epoch": 2.17,
"learning_rate": 3.359260346345759e-05,
"loss": 0.0027,
"step": 5190
},
{
"epoch": 2.17,
"learning_rate": 3.355591429410038e-05,
"loss": 0.0019,
"step": 5200
},
{
"epoch": 2.17,
"eval_f1": 0.9946274522472097,
"eval_f1_0": 0.9983760287298878,
"eval_f1_1": 0.9972181690701332,
"eval_f1_2": 0.9840624373694744,
"eval_f1_3": 0.9988531738193439,
"eval_loss": 0.010496237315237522,
"eval_runtime": 32.6466,
"eval_samples_per_second": 260.425,
"eval_steps_per_second": 1.041,
"step": 5200
},
{
"epoch": 2.18,
"learning_rate": 3.351922512474318e-05,
"loss": 0.0064,
"step": 5210
},
{
"epoch": 2.18,
"learning_rate": 3.348253595538598e-05,
"loss": 0.0052,
"step": 5220
},
{
"epoch": 2.19,
"learning_rate": 3.344584678602877e-05,
"loss": 0.0024,
"step": 5230
},
{
"epoch": 2.19,
"learning_rate": 3.340915761667156e-05,
"loss": 0.0214,
"step": 5240
},
{
"epoch": 2.2,
"learning_rate": 3.337246844731435e-05,
"loss": 0.0029,
"step": 5250
},
{
"epoch": 2.2,
"learning_rate": 3.3335779277957144e-05,
"loss": 0.0105,
"step": 5260
},
{
"epoch": 2.2,
"learning_rate": 3.3299090108599944e-05,
"loss": 0.0048,
"step": 5270
},
{
"epoch": 2.21,
"learning_rate": 3.326240093924274e-05,
"loss": 0.0024,
"step": 5280
},
{
"epoch": 2.21,
"learning_rate": 3.322571176988553e-05,
"loss": 0.0058,
"step": 5290
},
{
"epoch": 2.22,
"learning_rate": 3.3189022600528325e-05,
"loss": 0.0057,
"step": 5300
},
{
"epoch": 2.22,
"learning_rate": 3.315233343117112e-05,
"loss": 0.0029,
"step": 5310
},
{
"epoch": 2.23,
"learning_rate": 3.311564426181392e-05,
"loss": 0.0083,
"step": 5320
},
{
"epoch": 2.23,
"learning_rate": 3.307895509245671e-05,
"loss": 0.0044,
"step": 5330
},
{
"epoch": 2.23,
"learning_rate": 3.3042265923099506e-05,
"loss": 0.005,
"step": 5340
},
{
"epoch": 2.24,
"learning_rate": 3.300557675374229e-05,
"loss": 0.0028,
"step": 5350
},
{
"epoch": 2.24,
"learning_rate": 3.2968887584385086e-05,
"loss": 0.0049,
"step": 5360
},
{
"epoch": 2.25,
"learning_rate": 3.2932198415027886e-05,
"loss": 0.0339,
"step": 5370
},
{
"epoch": 2.25,
"learning_rate": 3.289550924567068e-05,
"loss": 0.0029,
"step": 5380
},
{
"epoch": 2.25,
"learning_rate": 3.285882007631347e-05,
"loss": 0.0258,
"step": 5390
},
{
"epoch": 2.26,
"learning_rate": 3.282213090695627e-05,
"loss": 0.0029,
"step": 5400
},
{
"epoch": 2.26,
"eval_f1": 0.9947696032575243,
"eval_f1_0": 0.9984141504451356,
"eval_f1_1": 0.9972693180116997,
"eval_f1_2": 0.9845506594654737,
"eval_f1_3": 0.9988442851077881,
"eval_loss": 0.010481318458914757,
"eval_runtime": 32.5582,
"eval_samples_per_second": 261.133,
"eval_steps_per_second": 1.044,
"step": 5400
},
{
"epoch": 2.26,
"learning_rate": 3.278544173759906e-05,
"loss": 0.003,
"step": 5410
},
{
"epoch": 2.27,
"learning_rate": 3.274875256824186e-05,
"loss": 0.0026,
"step": 5420
},
{
"epoch": 2.27,
"learning_rate": 3.2712063398884654e-05,
"loss": 0.003,
"step": 5430
},
{
"epoch": 2.28,
"learning_rate": 3.267537422952745e-05,
"loss": 0.0085,
"step": 5440
},
{
"epoch": 2.28,
"learning_rate": 3.2638685060170234e-05,
"loss": 0.0056,
"step": 5450
},
{
"epoch": 2.28,
"learning_rate": 3.2601995890813035e-05,
"loss": 0.0051,
"step": 5460
},
{
"epoch": 2.29,
"learning_rate": 3.256530672145583e-05,
"loss": 0.0033,
"step": 5470
},
{
"epoch": 2.29,
"learning_rate": 3.252861755209862e-05,
"loss": 0.0068,
"step": 5480
},
{
"epoch": 2.3,
"learning_rate": 3.2491928382741415e-05,
"loss": 0.0029,
"step": 5490
},
{
"epoch": 2.3,
"learning_rate": 3.245523921338421e-05,
"loss": 0.0024,
"step": 5500
},
{
"epoch": 2.3,
"learning_rate": 3.241855004402701e-05,
"loss": 0.0042,
"step": 5510
},
{
"epoch": 2.31,
"learning_rate": 3.23818608746698e-05,
"loss": 0.0022,
"step": 5520
},
{
"epoch": 2.31,
"learning_rate": 3.2345171705312596e-05,
"loss": 0.003,
"step": 5530
},
{
"epoch": 2.32,
"learning_rate": 3.230848253595539e-05,
"loss": 0.003,
"step": 5540
},
{
"epoch": 2.32,
"learning_rate": 3.227179336659818e-05,
"loss": 0.0042,
"step": 5550
},
{
"epoch": 2.33,
"learning_rate": 3.2235104197240976e-05,
"loss": 0.0291,
"step": 5560
},
{
"epoch": 2.33,
"learning_rate": 3.219841502788377e-05,
"loss": 0.0207,
"step": 5570
},
{
"epoch": 2.33,
"learning_rate": 3.2161725858526563e-05,
"loss": 0.0036,
"step": 5580
},
{
"epoch": 2.34,
"learning_rate": 3.212503668916936e-05,
"loss": 0.0031,
"step": 5590
},
{
"epoch": 2.34,
"learning_rate": 3.208834751981215e-05,
"loss": 0.0035,
"step": 5600
},
{
"epoch": 2.34,
"eval_f1": 0.9950011532059437,
"eval_f1_0": 0.9984910804889736,
"eval_f1_1": 0.9973762834220544,
"eval_f1_2": 0.985278293189393,
"eval_f1_3": 0.9988589557233543,
"eval_loss": 0.010064340196549892,
"eval_runtime": 32.5686,
"eval_samples_per_second": 261.049,
"eval_steps_per_second": 1.044,
"step": 5600
},
{
"epoch": 2.35,
"learning_rate": 3.205165835045495e-05,
"loss": 0.0017,
"step": 5610
},
{
"epoch": 2.35,
"learning_rate": 3.2014969181097744e-05,
"loss": 0.0026,
"step": 5620
},
{
"epoch": 2.35,
"learning_rate": 3.197828001174054e-05,
"loss": 0.0026,
"step": 5630
},
{
"epoch": 2.36,
"learning_rate": 3.194159084238333e-05,
"loss": 0.0023,
"step": 5640
},
{
"epoch": 2.36,
"learning_rate": 3.1904901673026125e-05,
"loss": 0.0032,
"step": 5650
},
{
"epoch": 2.37,
"learning_rate": 3.186821250366892e-05,
"loss": 0.0032,
"step": 5660
},
{
"epoch": 2.37,
"learning_rate": 3.183152333431171e-05,
"loss": 0.0031,
"step": 5670
},
{
"epoch": 2.38,
"learning_rate": 3.1794834164954505e-05,
"loss": 0.0143,
"step": 5680
},
{
"epoch": 2.38,
"learning_rate": 3.17581449955973e-05,
"loss": 0.0022,
"step": 5690
},
{
"epoch": 2.38,
"learning_rate": 3.172145582624009e-05,
"loss": 0.0024,
"step": 5700
},
{
"epoch": 2.39,
"learning_rate": 3.168476665688289e-05,
"loss": 0.0104,
"step": 5710
},
{
"epoch": 2.39,
"learning_rate": 3.1648077487525686e-05,
"loss": 0.0038,
"step": 5720
},
{
"epoch": 2.4,
"learning_rate": 3.161138831816848e-05,
"loss": 0.0044,
"step": 5730
},
{
"epoch": 2.4,
"learning_rate": 3.157469914881127e-05,
"loss": 0.0037,
"step": 5740
},
{
"epoch": 2.4,
"learning_rate": 3.1538009979454067e-05,
"loss": 0.0058,
"step": 5750
},
{
"epoch": 2.41,
"learning_rate": 3.150132081009686e-05,
"loss": 0.0037,
"step": 5760
},
{
"epoch": 2.41,
"learning_rate": 3.1464631640739654e-05,
"loss": 0.002,
"step": 5770
},
{
"epoch": 2.42,
"learning_rate": 3.142794247138245e-05,
"loss": 0.0086,
"step": 5780
},
{
"epoch": 2.42,
"learning_rate": 3.139125330202524e-05,
"loss": 0.0043,
"step": 5790
},
{
"epoch": 2.43,
"learning_rate": 3.135456413266804e-05,
"loss": 0.0038,
"step": 5800
},
{
"epoch": 2.43,
"eval_f1": 0.9949292811210624,
"eval_f1_0": 0.9984637871650293,
"eval_f1_1": 0.9972766709054832,
"eval_f1_2": 0.9851559395098007,
"eval_f1_3": 0.9988207269039364,
"eval_loss": 0.010227196849882603,
"eval_runtime": 32.4605,
"eval_samples_per_second": 261.919,
"eval_steps_per_second": 1.047,
"step": 5800
},
{
"epoch": 2.43,
"learning_rate": 3.1317874963310834e-05,
"loss": 0.0179,
"step": 5810
},
{
"epoch": 2.43,
"learning_rate": 3.128118579395363e-05,
"loss": 0.0026,
"step": 5820
},
{
"epoch": 2.44,
"learning_rate": 3.124449662459642e-05,
"loss": 0.0024,
"step": 5830
},
{
"epoch": 2.44,
"learning_rate": 3.1207807455239215e-05,
"loss": 0.0043,
"step": 5840
},
{
"epoch": 2.45,
"learning_rate": 3.117111828588201e-05,
"loss": 0.0054,
"step": 5850
},
{
"epoch": 2.45,
"learning_rate": 3.11344291165248e-05,
"loss": 0.0073,
"step": 5860
},
{
"epoch": 2.46,
"learning_rate": 3.1097739947167595e-05,
"loss": 0.0109,
"step": 5870
},
{
"epoch": 2.46,
"learning_rate": 3.106105077781039e-05,
"loss": 0.0046,
"step": 5880
},
{
"epoch": 2.46,
"learning_rate": 3.102436160845318e-05,
"loss": 0.0023,
"step": 5890
},
{
"epoch": 2.47,
"learning_rate": 3.098767243909598e-05,
"loss": 0.0043,
"step": 5900
},
{
"epoch": 2.47,
"learning_rate": 3.0950983269738776e-05,
"loss": 0.0031,
"step": 5910
},
{
"epoch": 2.48,
"learning_rate": 3.091429410038157e-05,
"loss": 0.0032,
"step": 5920
},
{
"epoch": 2.48,
"learning_rate": 3.087760493102436e-05,
"loss": 0.0262,
"step": 5930
},
{
"epoch": 2.48,
"learning_rate": 3.084091576166716e-05,
"loss": 0.0132,
"step": 5940
},
{
"epoch": 2.49,
"learning_rate": 3.080422659230996e-05,
"loss": 0.0039,
"step": 5950
},
{
"epoch": 2.49,
"learning_rate": 3.0767537422952744e-05,
"loss": 0.0027,
"step": 5960
},
{
"epoch": 2.5,
"learning_rate": 3.073084825359554e-05,
"loss": 0.0027,
"step": 5970
},
{
"epoch": 2.5,
"learning_rate": 3.069415908423833e-05,
"loss": 0.0035,
"step": 5980
},
{
"epoch": 2.51,
"learning_rate": 3.0657469914881124e-05,
"loss": 0.0123,
"step": 5990
},
{
"epoch": 2.51,
"learning_rate": 3.0620780745523924e-05,
"loss": 0.0243,
"step": 6000
},
{
"epoch": 2.51,
"eval_f1": 0.9936627294427128,
"eval_f1_0": 0.9980992938460442,
"eval_f1_1": 0.9970706804845212,
"eval_f1_2": 0.9807339395751616,
"eval_f1_3": 0.9987470038651245,
"eval_loss": 0.012905139476060867,
"eval_runtime": 32.493,
"eval_samples_per_second": 261.656,
"eval_steps_per_second": 1.046,
"step": 6000
},
{
"epoch": 2.51,
"learning_rate": 3.058409157616672e-05,
"loss": 0.0021,
"step": 6010
},
{
"epoch": 2.52,
"learning_rate": 3.054740240680951e-05,
"loss": 0.003,
"step": 6020
},
{
"epoch": 2.52,
"learning_rate": 3.0510713237452305e-05,
"loss": 0.0052,
"step": 6030
},
{
"epoch": 2.53,
"learning_rate": 3.04740240680951e-05,
"loss": 0.0028,
"step": 6040
},
{
"epoch": 2.53,
"learning_rate": 3.0437334898737895e-05,
"loss": 0.0196,
"step": 6050
},
{
"epoch": 2.53,
"learning_rate": 3.040064572938069e-05,
"loss": 0.0335,
"step": 6060
},
{
"epoch": 2.54,
"learning_rate": 3.0363956560023482e-05,
"loss": 0.0075,
"step": 6070
},
{
"epoch": 2.54,
"learning_rate": 3.0327267390666276e-05,
"loss": 0.0134,
"step": 6080
},
{
"epoch": 2.55,
"learning_rate": 3.029057822130907e-05,
"loss": 0.0034,
"step": 6090
},
{
"epoch": 2.55,
"learning_rate": 3.0253889051951866e-05,
"loss": 0.0031,
"step": 6100
},
{
"epoch": 2.56,
"learning_rate": 3.021719988259466e-05,
"loss": 0.0073,
"step": 6110
},
{
"epoch": 2.56,
"learning_rate": 3.0180510713237453e-05,
"loss": 0.0025,
"step": 6120
},
{
"epoch": 2.56,
"learning_rate": 3.0143821543880247e-05,
"loss": 0.0051,
"step": 6130
},
{
"epoch": 2.57,
"learning_rate": 3.0107132374523044e-05,
"loss": 0.0026,
"step": 6140
},
{
"epoch": 2.57,
"learning_rate": 3.0070443205165837e-05,
"loss": 0.0073,
"step": 6150
},
{
"epoch": 2.58,
"learning_rate": 3.003375403580863e-05,
"loss": 0.004,
"step": 6160
},
{
"epoch": 2.58,
"learning_rate": 2.9997064866451424e-05,
"loss": 0.0038,
"step": 6170
},
{
"epoch": 2.58,
"learning_rate": 2.9960375697094218e-05,
"loss": 0.0058,
"step": 6180
},
{
"epoch": 2.59,
"learning_rate": 2.9923686527737015e-05,
"loss": 0.0069,
"step": 6190
},
{
"epoch": 2.59,
"learning_rate": 2.9886997358379808e-05,
"loss": 0.0022,
"step": 6200
},
{
"epoch": 2.59,
"eval_f1": 0.9942435755186948,
"eval_f1_0": 0.9982598041923024,
"eval_f1_1": 0.9973448862053478,
"eval_f1_2": 0.9825136010156198,
"eval_f1_3": 0.9988560106615089,
"eval_loss": 0.010344821028411388,
"eval_runtime": 32.4964,
"eval_samples_per_second": 261.629,
"eval_steps_per_second": 1.046,
"step": 6200
},
{
"epoch": 2.6,
"learning_rate": 2.98503081890226e-05,
"loss": 0.023,
"step": 6210
},
{
"epoch": 2.6,
"learning_rate": 2.9813619019665395e-05,
"loss": 0.008,
"step": 6220
},
{
"epoch": 2.61,
"learning_rate": 2.977692985030819e-05,
"loss": 0.0047,
"step": 6230
},
{
"epoch": 2.61,
"learning_rate": 2.9740240680950986e-05,
"loss": 0.0027,
"step": 6240
},
{
"epoch": 2.61,
"learning_rate": 2.970355151159378e-05,
"loss": 0.0023,
"step": 6250
},
{
"epoch": 2.62,
"learning_rate": 2.9666862342236573e-05,
"loss": 0.0048,
"step": 6260
},
{
"epoch": 2.62,
"learning_rate": 2.9630173172879366e-05,
"loss": 0.0037,
"step": 6270
},
{
"epoch": 2.63,
"learning_rate": 2.959348400352216e-05,
"loss": 0.0025,
"step": 6280
},
{
"epoch": 2.63,
"learning_rate": 2.9556794834164956e-05,
"loss": 0.0026,
"step": 6290
},
{
"epoch": 2.63,
"learning_rate": 2.952010566480775e-05,
"loss": 0.0047,
"step": 6300
},
{
"epoch": 2.64,
"learning_rate": 2.9483416495450543e-05,
"loss": 0.003,
"step": 6310
},
{
"epoch": 2.64,
"learning_rate": 2.9446727326093337e-05,
"loss": 0.0095,
"step": 6320
},
{
"epoch": 2.65,
"learning_rate": 2.941003815673613e-05,
"loss": 0.0033,
"step": 6330
},
{
"epoch": 2.65,
"learning_rate": 2.937334898737893e-05,
"loss": 0.0097,
"step": 6340
},
{
"epoch": 2.66,
"learning_rate": 2.933665981802172e-05,
"loss": 0.0021,
"step": 6350
},
{
"epoch": 2.66,
"learning_rate": 2.9299970648664514e-05,
"loss": 0.0027,
"step": 6360
},
{
"epoch": 2.66,
"learning_rate": 2.9263281479307308e-05,
"loss": 0.002,
"step": 6370
},
{
"epoch": 2.67,
"learning_rate": 2.92265923099501e-05,
"loss": 0.0117,
"step": 6380
},
{
"epoch": 2.67,
"learning_rate": 2.91899031405929e-05,
"loss": 0.0041,
"step": 6390
},
{
"epoch": 2.68,
"learning_rate": 2.9153213971235692e-05,
"loss": 0.0036,
"step": 6400
},
{
"epoch": 2.68,
"eval_f1": 0.9947850471072875,
"eval_f1_0": 0.9984259337649861,
"eval_f1_1": 0.9974449082024512,
"eval_f1_2": 0.9844015354193323,
"eval_f1_3": 0.9988678110423804,
"eval_loss": 0.010553164407610893,
"eval_runtime": 32.5079,
"eval_samples_per_second": 261.537,
"eval_steps_per_second": 1.046,
"step": 6400
},
{
"epoch": 2.68,
"learning_rate": 2.9116524801878485e-05,
"loss": 0.0032,
"step": 6410
},
{
"epoch": 2.69,
"learning_rate": 2.907983563252128e-05,
"loss": 0.0051,
"step": 6420
},
{
"epoch": 2.69,
"learning_rate": 2.9043146463164072e-05,
"loss": 0.0042,
"step": 6430
},
{
"epoch": 2.69,
"learning_rate": 2.9006457293806873e-05,
"loss": 0.0283,
"step": 6440
},
{
"epoch": 2.7,
"learning_rate": 2.8969768124449663e-05,
"loss": 0.0044,
"step": 6450
},
{
"epoch": 2.7,
"learning_rate": 2.8933078955092456e-05,
"loss": 0.0027,
"step": 6460
},
{
"epoch": 2.71,
"learning_rate": 2.889638978573525e-05,
"loss": 0.0031,
"step": 6470
},
{
"epoch": 2.71,
"learning_rate": 2.885970061637805e-05,
"loss": 0.0034,
"step": 6480
},
{
"epoch": 2.71,
"learning_rate": 2.8823011447020843e-05,
"loss": 0.005,
"step": 6490
},
{
"epoch": 2.72,
"learning_rate": 2.8786322277663637e-05,
"loss": 0.0024,
"step": 6500
},
{
"epoch": 2.72,
"learning_rate": 2.8749633108306427e-05,
"loss": 0.0027,
"step": 6510
},
{
"epoch": 2.73,
"learning_rate": 2.871294393894922e-05,
"loss": 0.0022,
"step": 6520
},
{
"epoch": 2.73,
"learning_rate": 2.867625476959202e-05,
"loss": 0.0021,
"step": 6530
},
{
"epoch": 2.74,
"learning_rate": 2.8639565600234814e-05,
"loss": 0.0051,
"step": 6540
},
{
"epoch": 2.74,
"learning_rate": 2.8602876430877608e-05,
"loss": 0.0032,
"step": 6550
},
{
"epoch": 2.74,
"learning_rate": 2.8566187261520398e-05,
"loss": 0.0023,
"step": 6560
},
{
"epoch": 2.75,
"learning_rate": 2.852949809216319e-05,
"loss": 0.0205,
"step": 6570
},
{
"epoch": 2.75,
"learning_rate": 2.8492808922805992e-05,
"loss": 0.0078,
"step": 6580
},
{
"epoch": 2.76,
"learning_rate": 2.8456119753448785e-05,
"loss": 0.005,
"step": 6590
},
{
"epoch": 2.76,
"learning_rate": 2.841943058409158e-05,
"loss": 0.0081,
"step": 6600
},
{
"epoch": 2.76,
"eval_f1": 0.9942146099405986,
"eval_f1_0": 0.9982458413827217,
"eval_f1_1": 0.9972979528108112,
"eval_f1_2": 0.9824526910788891,
"eval_f1_3": 0.9988619544899728,
"eval_loss": 0.010127900168299675,
"eval_runtime": 32.5613,
"eval_samples_per_second": 261.107,
"eval_steps_per_second": 1.044,
"step": 6600
},
{
"epoch": 2.76,
"learning_rate": 2.8382741414734372e-05,
"loss": 0.0027,
"step": 6610
},
{
"epoch": 2.77,
"learning_rate": 2.8346052245377162e-05,
"loss": 0.0064,
"step": 6620
},
{
"epoch": 2.77,
"learning_rate": 2.8309363076019963e-05,
"loss": 0.0023,
"step": 6630
},
{
"epoch": 2.78,
"learning_rate": 2.8272673906662756e-05,
"loss": 0.0224,
"step": 6640
},
{
"epoch": 2.78,
"learning_rate": 2.823598473730555e-05,
"loss": 0.0019,
"step": 6650
},
{
"epoch": 2.79,
"learning_rate": 2.8199295567948343e-05,
"loss": 0.0283,
"step": 6660
},
{
"epoch": 2.79,
"learning_rate": 2.8162606398591133e-05,
"loss": 0.0043,
"step": 6670
},
{
"epoch": 2.79,
"learning_rate": 2.8125917229233934e-05,
"loss": 0.0034,
"step": 6680
},
{
"epoch": 2.8,
"learning_rate": 2.8089228059876727e-05,
"loss": 0.0031,
"step": 6690
},
{
"epoch": 2.8,
"learning_rate": 2.805253889051952e-05,
"loss": 0.0027,
"step": 6700
},
{
"epoch": 2.81,
"learning_rate": 2.8015849721162314e-05,
"loss": 0.0115,
"step": 6710
},
{
"epoch": 2.81,
"learning_rate": 2.7979160551805104e-05,
"loss": 0.0032,
"step": 6720
},
{
"epoch": 2.81,
"learning_rate": 2.7942471382447905e-05,
"loss": 0.0027,
"step": 6730
},
{
"epoch": 2.82,
"learning_rate": 2.7905782213090698e-05,
"loss": 0.0104,
"step": 6740
},
{
"epoch": 2.82,
"learning_rate": 2.786909304373349e-05,
"loss": 0.007,
"step": 6750
},
{
"epoch": 2.83,
"learning_rate": 2.7832403874376285e-05,
"loss": 0.0058,
"step": 6760
},
{
"epoch": 2.83,
"learning_rate": 2.779571470501908e-05,
"loss": 0.0055,
"step": 6770
},
{
"epoch": 2.84,
"learning_rate": 2.7759025535661875e-05,
"loss": 0.0046,
"step": 6780
},
{
"epoch": 2.84,
"learning_rate": 2.772233636630467e-05,
"loss": 0.0062,
"step": 6790
},
{
"epoch": 2.84,
"learning_rate": 2.7685647196947462e-05,
"loss": 0.0231,
"step": 6800
},
{
"epoch": 2.84,
"eval_f1": 0.994625110377242,
"eval_f1_0": 0.9983772276312078,
"eval_f1_1": 0.9972981473551781,
"eval_f1_2": 0.983974844232225,
"eval_f1_3": 0.9988502222903572,
"eval_loss": 0.010598053224384785,
"eval_runtime": 32.5415,
"eval_samples_per_second": 261.267,
"eval_steps_per_second": 1.045,
"step": 6800
},
{
"epoch": 2.85,
"learning_rate": 2.7648958027590256e-05,
"loss": 0.0026,
"step": 6810
},
{
"epoch": 2.85,
"learning_rate": 2.7612268858233053e-05,
"loss": 0.0019,
"step": 6820
},
{
"epoch": 2.86,
"learning_rate": 2.7575579688875846e-05,
"loss": 0.0025,
"step": 6830
},
{
"epoch": 2.86,
"learning_rate": 2.753889051951864e-05,
"loss": 0.0063,
"step": 6840
},
{
"epoch": 2.86,
"learning_rate": 2.7502201350161433e-05,
"loss": 0.0023,
"step": 6850
},
{
"epoch": 2.87,
"learning_rate": 2.7465512180804227e-05,
"loss": 0.0091,
"step": 6860
},
{
"epoch": 2.87,
"learning_rate": 2.7428823011447024e-05,
"loss": 0.0027,
"step": 6870
},
{
"epoch": 2.88,
"learning_rate": 2.7392133842089817e-05,
"loss": 0.0029,
"step": 6880
},
{
"epoch": 2.88,
"learning_rate": 2.735544467273261e-05,
"loss": 0.0028,
"step": 6890
},
{
"epoch": 2.89,
"learning_rate": 2.7318755503375404e-05,
"loss": 0.0036,
"step": 6900
},
{
"epoch": 2.89,
"learning_rate": 2.7282066334018198e-05,
"loss": 0.0069,
"step": 6910
},
{
"epoch": 2.89,
"learning_rate": 2.7245377164660995e-05,
"loss": 0.0024,
"step": 6920
},
{
"epoch": 2.9,
"learning_rate": 2.7208687995303788e-05,
"loss": 0.0044,
"step": 6930
},
{
"epoch": 2.9,
"learning_rate": 2.717199882594658e-05,
"loss": 0.0065,
"step": 6940
},
{
"epoch": 2.91,
"learning_rate": 2.7135309656589375e-05,
"loss": 0.0042,
"step": 6950
},
{
"epoch": 2.91,
"learning_rate": 2.709862048723217e-05,
"loss": 0.0021,
"step": 6960
},
{
"epoch": 2.92,
"learning_rate": 2.7061931317874966e-05,
"loss": 0.0033,
"step": 6970
},
{
"epoch": 2.92,
"learning_rate": 2.702524214851776e-05,
"loss": 0.0062,
"step": 6980
},
{
"epoch": 2.92,
"learning_rate": 2.6988552979160553e-05,
"loss": 0.0026,
"step": 6990
},
{
"epoch": 2.93,
"learning_rate": 2.6951863809803346e-05,
"loss": 0.0025,
"step": 7000
},
{
"epoch": 2.93,
"eval_f1": 0.9946709754148209,
"eval_f1_0": 0.9983931458064326,
"eval_f1_1": 0.9973767409417955,
"eval_f1_2": 0.9840609154786092,
"eval_f1_3": 0.9988530994324464,
"eval_loss": 0.010840615257620811,
"eval_runtime": 32.6051,
"eval_samples_per_second": 260.757,
"eval_steps_per_second": 1.043,
"step": 7000
},
{
"epoch": 2.93,
"learning_rate": 2.691517464044614e-05,
"loss": 0.0023,
"step": 7010
},
{
"epoch": 2.94,
"learning_rate": 2.6878485471088936e-05,
"loss": 0.0025,
"step": 7020
},
{
"epoch": 2.94,
"learning_rate": 2.684179630173173e-05,
"loss": 0.0042,
"step": 7030
},
{
"epoch": 2.94,
"learning_rate": 2.6805107132374524e-05,
"loss": 0.0051,
"step": 7040
},
{
"epoch": 2.95,
"learning_rate": 2.6768417963017317e-05,
"loss": 0.0053,
"step": 7050
},
{
"epoch": 2.95,
"learning_rate": 2.673172879366011e-05,
"loss": 0.0044,
"step": 7060
},
{
"epoch": 2.96,
"learning_rate": 2.6695039624302907e-05,
"loss": 0.0023,
"step": 7070
},
{
"epoch": 2.96,
"learning_rate": 2.66583504549457e-05,
"loss": 0.0065,
"step": 7080
},
{
"epoch": 2.97,
"learning_rate": 2.6621661285588494e-05,
"loss": 0.0035,
"step": 7090
},
{
"epoch": 2.97,
"learning_rate": 2.6584972116231288e-05,
"loss": 0.0055,
"step": 7100
},
{
"epoch": 2.97,
"learning_rate": 2.6548282946874088e-05,
"loss": 0.0282,
"step": 7110
},
{
"epoch": 2.98,
"learning_rate": 2.651159377751688e-05,
"loss": 0.0028,
"step": 7120
},
{
"epoch": 2.98,
"learning_rate": 2.6474904608159672e-05,
"loss": 0.0052,
"step": 7130
},
{
"epoch": 2.99,
"learning_rate": 2.6438215438802465e-05,
"loss": 0.0029,
"step": 7140
},
{
"epoch": 2.99,
"learning_rate": 2.640152626944526e-05,
"loss": 0.0094,
"step": 7150
},
{
"epoch": 2.99,
"learning_rate": 2.636483710008806e-05,
"loss": 0.0036,
"step": 7160
},
{
"epoch": 3.0,
"learning_rate": 2.6328147930730853e-05,
"loss": 0.0048,
"step": 7170
},
{
"epoch": 3.0,
"learning_rate": 2.6291458761373643e-05,
"loss": 0.0025,
"step": 7180
},
{
"epoch": 3.01,
"learning_rate": 2.6254769592016436e-05,
"loss": 0.0243,
"step": 7190
},
{
"epoch": 3.01,
"learning_rate": 2.621808042265923e-05,
"loss": 0.0053,
"step": 7200
},
{
"epoch": 3.01,
"eval_f1": 0.9943055688326237,
"eval_f1_0": 0.9982727880800281,
"eval_f1_1": 0.9973206300119757,
"eval_f1_2": 0.9827817020526062,
"eval_f1_3": 0.9988471551858852,
"eval_loss": 0.010248737409710884,
"eval_runtime": 32.5278,
"eval_samples_per_second": 261.377,
"eval_steps_per_second": 1.045,
"step": 7200
},
{
"epoch": 3.02,
"learning_rate": 2.618139125330203e-05,
"loss": 0.0038,
"step": 7210
},
{
"epoch": 3.02,
"learning_rate": 2.6144702083944824e-05,
"loss": 0.0203,
"step": 7220
},
{
"epoch": 3.02,
"learning_rate": 2.6108012914587614e-05,
"loss": 0.0063,
"step": 7230
},
{
"epoch": 3.03,
"learning_rate": 2.6071323745230407e-05,
"loss": 0.0086,
"step": 7240
},
{
"epoch": 3.03,
"learning_rate": 2.60346345758732e-05,
"loss": 0.0043,
"step": 7250
},
{
"epoch": 3.04,
"learning_rate": 2.5997945406516e-05,
"loss": 0.0047,
"step": 7260
},
{
"epoch": 3.04,
"learning_rate": 2.5961256237158794e-05,
"loss": 0.0033,
"step": 7270
},
{
"epoch": 3.04,
"learning_rate": 2.5924567067801585e-05,
"loss": 0.0061,
"step": 7280
},
{
"epoch": 3.05,
"learning_rate": 2.5887877898444378e-05,
"loss": 0.0049,
"step": 7290
},
{
"epoch": 3.05,
"learning_rate": 2.585118872908717e-05,
"loss": 0.0028,
"step": 7300
},
{
"epoch": 3.06,
"learning_rate": 2.5814499559729972e-05,
"loss": 0.0028,
"step": 7310
},
{
"epoch": 3.06,
"learning_rate": 2.5777810390372765e-05,
"loss": 0.002,
"step": 7320
},
{
"epoch": 3.07,
"learning_rate": 2.574112122101556e-05,
"loss": 0.0019,
"step": 7330
},
{
"epoch": 3.07,
"learning_rate": 2.570443205165835e-05,
"loss": 0.0026,
"step": 7340
},
{
"epoch": 3.07,
"learning_rate": 2.5667742882301142e-05,
"loss": 0.0029,
"step": 7350
},
{
"epoch": 3.08,
"learning_rate": 2.5631053712943943e-05,
"loss": 0.0024,
"step": 7360
},
{
"epoch": 3.08,
"learning_rate": 2.5594364543586736e-05,
"loss": 0.0024,
"step": 7370
},
{
"epoch": 3.09,
"learning_rate": 2.555767537422953e-05,
"loss": 0.003,
"step": 7380
},
{
"epoch": 3.09,
"learning_rate": 2.552098620487232e-05,
"loss": 0.0076,
"step": 7390
},
{
"epoch": 3.09,
"learning_rate": 2.5484297035515113e-05,
"loss": 0.0123,
"step": 7400
},
{
"epoch": 3.09,
"eval_f1": 0.9942531602386351,
"eval_f1_0": 0.9982579119066933,
"eval_f1_1": 0.9973885094415428,
"eval_f1_2": 0.9824924187725632,
"eval_f1_3": 0.9988738008337411,
"eval_loss": 0.010519725270569324,
"eval_runtime": 32.5628,
"eval_samples_per_second": 261.095,
"eval_steps_per_second": 1.044,
"step": 7400
},
{
"epoch": 3.1,
"learning_rate": 2.5447607866157914e-05,
"loss": 0.0057,
"step": 7410
},
{
"epoch": 3.1,
"learning_rate": 2.5410918696800707e-05,
"loss": 0.0037,
"step": 7420
},
{
"epoch": 3.11,
"learning_rate": 2.53742295274435e-05,
"loss": 0.0022,
"step": 7430
},
{
"epoch": 3.11,
"learning_rate": 2.533754035808629e-05,
"loss": 0.0018,
"step": 7440
},
{
"epoch": 3.12,
"learning_rate": 2.530085118872909e-05,
"loss": 0.0014,
"step": 7450
},
{
"epoch": 3.12,
"learning_rate": 2.5264162019371885e-05,
"loss": 0.0031,
"step": 7460
},
{
"epoch": 3.12,
"learning_rate": 2.5227472850014678e-05,
"loss": 0.0022,
"step": 7470
},
{
"epoch": 3.13,
"learning_rate": 2.519078368065747e-05,
"loss": 0.0241,
"step": 7480
},
{
"epoch": 3.13,
"learning_rate": 2.5154094511300265e-05,
"loss": 0.0202,
"step": 7490
},
{
"epoch": 3.14,
"learning_rate": 2.5117405341943062e-05,
"loss": 0.0031,
"step": 7500
},
{
"epoch": 3.14,
"learning_rate": 2.5080716172585855e-05,
"loss": 0.0019,
"step": 7510
},
{
"epoch": 3.15,
"learning_rate": 2.504402700322865e-05,
"loss": 0.0026,
"step": 7520
},
{
"epoch": 3.15,
"learning_rate": 2.5007337833871442e-05,
"loss": 0.0035,
"step": 7530
},
{
"epoch": 3.15,
"learning_rate": 2.4970648664514236e-05,
"loss": 0.0023,
"step": 7540
},
{
"epoch": 3.16,
"learning_rate": 2.493395949515703e-05,
"loss": 0.0034,
"step": 7550
},
{
"epoch": 3.16,
"learning_rate": 2.4897270325799826e-05,
"loss": 0.0018,
"step": 7560
},
{
"epoch": 3.17,
"learning_rate": 2.486058115644262e-05,
"loss": 0.0033,
"step": 7570
},
{
"epoch": 3.17,
"learning_rate": 2.4823891987085413e-05,
"loss": 0.0023,
"step": 7580
},
{
"epoch": 3.17,
"learning_rate": 2.4787202817728207e-05,
"loss": 0.0034,
"step": 7590
},
{
"epoch": 3.18,
"learning_rate": 2.4750513648371e-05,
"loss": 0.0028,
"step": 7600
},
{
"epoch": 3.18,
"eval_f1": 0.9948112268005924,
"eval_f1_0": 0.9984296334269724,
"eval_f1_1": 0.9973807899325298,
"eval_f1_2": 0.9845489100607764,
"eval_f1_3": 0.9988855737820913,
"eval_loss": 0.010309526696801186,
"eval_runtime": 32.4974,
"eval_samples_per_second": 261.621,
"eval_steps_per_second": 1.046,
"step": 7600
},
{
"epoch": 3.18,
"learning_rate": 2.4713824479013797e-05,
"loss": 0.0055,
"step": 7610
},
{
"epoch": 3.19,
"learning_rate": 2.467713530965659e-05,
"loss": 0.0024,
"step": 7620
},
{
"epoch": 3.19,
"learning_rate": 2.4640446140299384e-05,
"loss": 0.0026,
"step": 7630
},
{
"epoch": 3.2,
"learning_rate": 2.4603756970942178e-05,
"loss": 0.0022,
"step": 7640
},
{
"epoch": 3.2,
"learning_rate": 2.456706780158497e-05,
"loss": 0.0018,
"step": 7650
},
{
"epoch": 3.2,
"learning_rate": 2.4530378632227768e-05,
"loss": 0.0023,
"step": 7660
},
{
"epoch": 3.21,
"learning_rate": 2.4493689462870562e-05,
"loss": 0.0023,
"step": 7670
},
{
"epoch": 3.21,
"learning_rate": 2.4457000293513355e-05,
"loss": 0.0039,
"step": 7680
},
{
"epoch": 3.22,
"learning_rate": 2.4420311124156152e-05,
"loss": 0.0065,
"step": 7690
},
{
"epoch": 3.22,
"learning_rate": 2.4383621954798942e-05,
"loss": 0.0025,
"step": 7700
},
{
"epoch": 3.22,
"learning_rate": 2.434693278544174e-05,
"loss": 0.0029,
"step": 7710
},
{
"epoch": 3.23,
"learning_rate": 2.4310243616084533e-05,
"loss": 0.0021,
"step": 7720
},
{
"epoch": 3.23,
"learning_rate": 2.427355444672733e-05,
"loss": 0.0026,
"step": 7730
},
{
"epoch": 3.24,
"learning_rate": 2.4236865277370123e-05,
"loss": 0.0023,
"step": 7740
},
{
"epoch": 3.24,
"learning_rate": 2.4200176108012913e-05,
"loss": 0.0034,
"step": 7750
},
{
"epoch": 3.25,
"learning_rate": 2.416348693865571e-05,
"loss": 0.0095,
"step": 7760
},
{
"epoch": 3.25,
"learning_rate": 2.4126797769298504e-05,
"loss": 0.0068,
"step": 7770
},
{
"epoch": 3.25,
"learning_rate": 2.40901085999413e-05,
"loss": 0.0027,
"step": 7780
},
{
"epoch": 3.26,
"learning_rate": 2.4053419430584094e-05,
"loss": 0.0035,
"step": 7790
},
{
"epoch": 3.26,
"learning_rate": 2.4016730261226884e-05,
"loss": 0.0029,
"step": 7800
},
{
"epoch": 3.26,
"eval_f1": 0.9947889953932831,
"eval_f1_0": 0.9984267205831371,
"eval_f1_1": 0.9974202817787735,
"eval_f1_2": 0.9844470314318976,
"eval_f1_3": 0.998861947779324,
"eval_loss": 0.010554310865700245,
"eval_runtime": 32.7178,
"eval_samples_per_second": 259.858,
"eval_steps_per_second": 1.039,
"step": 7800
},
{
"epoch": 3.27,
"learning_rate": 2.398004109186968e-05,
"loss": 0.0017,
"step": 7810
},
{
"epoch": 3.27,
"learning_rate": 2.3943351922512474e-05,
"loss": 0.002,
"step": 7820
},
{
"epoch": 3.27,
"learning_rate": 2.390666275315527e-05,
"loss": 0.0021,
"step": 7830
},
{
"epoch": 3.28,
"learning_rate": 2.3869973583798065e-05,
"loss": 0.0024,
"step": 7840
},
{
"epoch": 3.28,
"learning_rate": 2.383328441444086e-05,
"loss": 0.0026,
"step": 7850
},
{
"epoch": 3.29,
"learning_rate": 2.3796595245083652e-05,
"loss": 0.0042,
"step": 7860
},
{
"epoch": 3.29,
"learning_rate": 2.3759906075726445e-05,
"loss": 0.0054,
"step": 7870
},
{
"epoch": 3.3,
"learning_rate": 2.3723216906369242e-05,
"loss": 0.0215,
"step": 7880
},
{
"epoch": 3.3,
"learning_rate": 2.3686527737012036e-05,
"loss": 0.0024,
"step": 7890
},
{
"epoch": 3.3,
"learning_rate": 2.364983856765483e-05,
"loss": 0.0028,
"step": 7900
},
{
"epoch": 3.31,
"learning_rate": 2.3613149398297623e-05,
"loss": 0.0047,
"step": 7910
},
{
"epoch": 3.31,
"learning_rate": 2.3576460228940416e-05,
"loss": 0.0032,
"step": 7920
},
{
"epoch": 3.32,
"learning_rate": 2.3539771059583213e-05,
"loss": 0.0206,
"step": 7930
},
{
"epoch": 3.32,
"learning_rate": 2.3503081890226007e-05,
"loss": 0.0031,
"step": 7940
},
{
"epoch": 3.32,
"learning_rate": 2.34663927208688e-05,
"loss": 0.0052,
"step": 7950
},
{
"epoch": 3.33,
"learning_rate": 2.3429703551511594e-05,
"loss": 0.0044,
"step": 7960
},
{
"epoch": 3.33,
"learning_rate": 2.3393014382154387e-05,
"loss": 0.002,
"step": 7970
},
{
"epoch": 3.34,
"learning_rate": 2.3356325212797184e-05,
"loss": 0.0035,
"step": 7980
},
{
"epoch": 3.34,
"learning_rate": 2.3319636043439978e-05,
"loss": 0.0173,
"step": 7990
},
{
"epoch": 3.35,
"learning_rate": 2.328294687408277e-05,
"loss": 0.0021,
"step": 8000
},
{
"epoch": 3.35,
"eval_f1": 0.9950137070205632,
"eval_f1_0": 0.9984852943781941,
"eval_f1_1": 0.9972922939329573,
"eval_f1_2": 0.9853916659890104,
"eval_f1_3": 0.9988855737820913,
"eval_loss": 0.010200351476669312,
"eval_runtime": 32.4179,
"eval_samples_per_second": 262.263,
"eval_steps_per_second": 1.049,
"step": 8000
},
{
"epoch": 3.35,
"learning_rate": 2.3246257704725565e-05,
"loss": 0.0029,
"step": 8010
},
{
"epoch": 3.35,
"learning_rate": 2.320956853536836e-05,
"loss": 0.0081,
"step": 8020
},
{
"epoch": 3.36,
"learning_rate": 2.3172879366011155e-05,
"loss": 0.0023,
"step": 8030
},
{
"epoch": 3.36,
"learning_rate": 2.313619019665395e-05,
"loss": 0.0019,
"step": 8040
},
{
"epoch": 3.37,
"learning_rate": 2.3099501027296745e-05,
"loss": 0.0027,
"step": 8050
},
{
"epoch": 3.37,
"learning_rate": 2.3062811857939536e-05,
"loss": 0.0024,
"step": 8060
},
{
"epoch": 3.38,
"learning_rate": 2.3026122688582332e-05,
"loss": 0.0018,
"step": 8070
},
{
"epoch": 3.38,
"learning_rate": 2.2989433519225126e-05,
"loss": 0.0021,
"step": 8080
},
{
"epoch": 3.38,
"learning_rate": 2.295274434986792e-05,
"loss": 0.0023,
"step": 8090
},
{
"epoch": 3.39,
"learning_rate": 2.2916055180510716e-05,
"loss": 0.0138,
"step": 8100
},
{
"epoch": 3.39,
"learning_rate": 2.2879366011153506e-05,
"loss": 0.0022,
"step": 8110
},
{
"epoch": 3.4,
"learning_rate": 2.2842676841796303e-05,
"loss": 0.0021,
"step": 8120
},
{
"epoch": 3.4,
"learning_rate": 2.2805987672439097e-05,
"loss": 0.0032,
"step": 8130
},
{
"epoch": 3.4,
"learning_rate": 2.276929850308189e-05,
"loss": 0.0025,
"step": 8140
},
{
"epoch": 3.41,
"learning_rate": 2.2732609333724687e-05,
"loss": 0.0026,
"step": 8150
},
{
"epoch": 3.41,
"learning_rate": 2.269592016436748e-05,
"loss": 0.0027,
"step": 8160
},
{
"epoch": 3.42,
"learning_rate": 2.2659230995010274e-05,
"loss": 0.0019,
"step": 8170
},
{
"epoch": 3.42,
"learning_rate": 2.2622541825653068e-05,
"loss": 0.0022,
"step": 8180
},
{
"epoch": 3.43,
"learning_rate": 2.2585852656295865e-05,
"loss": 0.0101,
"step": 8190
},
{
"epoch": 3.43,
"learning_rate": 2.2549163486938658e-05,
"loss": 0.0018,
"step": 8200
},
{
"epoch": 3.43,
"eval_f1": 0.9947056254494651,
"eval_f1_0": 0.998400933444462,
"eval_f1_1": 0.9973489578694641,
"eval_f1_2": 0.9842401872314661,
"eval_f1_3": 0.9988324232524678,
"eval_loss": 0.010863890871405602,
"eval_runtime": 32.6157,
"eval_samples_per_second": 260.672,
"eval_steps_per_second": 1.042,
"step": 8200
},
{
"epoch": 3.43,
"learning_rate": 2.251247431758145e-05,
"loss": 0.0023,
"step": 8210
},
{
"epoch": 3.44,
"learning_rate": 2.2475785148224245e-05,
"loss": 0.0022,
"step": 8220
},
{
"epoch": 3.44,
"learning_rate": 2.243909597886704e-05,
"loss": 0.0034,
"step": 8230
},
{
"epoch": 3.45,
"learning_rate": 2.2402406809509836e-05,
"loss": 0.0026,
"step": 8240
},
{
"epoch": 3.45,
"learning_rate": 2.236571764015263e-05,
"loss": 0.0082,
"step": 8250
},
{
"epoch": 3.45,
"learning_rate": 2.2329028470795423e-05,
"loss": 0.008,
"step": 8260
},
{
"epoch": 3.46,
"learning_rate": 2.2292339301438216e-05,
"loss": 0.0036,
"step": 8270
},
{
"epoch": 3.46,
"learning_rate": 2.225565013208101e-05,
"loss": 0.0054,
"step": 8280
},
{
"epoch": 3.47,
"learning_rate": 2.2218960962723806e-05,
"loss": 0.003,
"step": 8290
},
{
"epoch": 3.47,
"learning_rate": 2.21822717933666e-05,
"loss": 0.0021,
"step": 8300
},
{
"epoch": 3.48,
"learning_rate": 2.2145582624009393e-05,
"loss": 0.0041,
"step": 8310
},
{
"epoch": 3.48,
"learning_rate": 2.2108893454652187e-05,
"loss": 0.006,
"step": 8320
},
{
"epoch": 3.48,
"learning_rate": 2.207220428529498e-05,
"loss": 0.0029,
"step": 8330
},
{
"epoch": 3.49,
"learning_rate": 2.2035515115937777e-05,
"loss": 0.0097,
"step": 8340
},
{
"epoch": 3.49,
"learning_rate": 2.199882594658057e-05,
"loss": 0.0026,
"step": 8350
},
{
"epoch": 3.5,
"learning_rate": 2.1962136777223364e-05,
"loss": 0.0025,
"step": 8360
},
{
"epoch": 3.5,
"learning_rate": 2.1925447607866158e-05,
"loss": 0.0048,
"step": 8370
},
{
"epoch": 3.5,
"learning_rate": 2.188875843850895e-05,
"loss": 0.004,
"step": 8380
},
{
"epoch": 3.51,
"learning_rate": 2.1852069269151748e-05,
"loss": 0.0042,
"step": 8390
},
{
"epoch": 3.51,
"learning_rate": 2.1815380099794542e-05,
"loss": 0.0024,
"step": 8400
},
{
"epoch": 3.51,
"eval_f1": 0.9948874932296581,
"eval_f1_0": 0.9984523681619899,
"eval_f1_1": 0.9973191770726172,
"eval_f1_2": 0.9849578120453885,
"eval_f1_3": 0.9988206156386367,
"eval_loss": 0.01073378324508667,
"eval_runtime": 32.4716,
"eval_samples_per_second": 261.829,
"eval_steps_per_second": 1.047,
"step": 8400
},
{
"epoch": 3.52,
"learning_rate": 2.177869093043734e-05,
"loss": 0.0073,
"step": 8410
},
{
"epoch": 3.52,
"learning_rate": 2.174200176108013e-05,
"loss": 0.0024,
"step": 8420
},
{
"epoch": 3.53,
"learning_rate": 2.1705312591722922e-05,
"loss": 0.0034,
"step": 8430
},
{
"epoch": 3.53,
"learning_rate": 2.166862342236572e-05,
"loss": 0.0036,
"step": 8440
},
{
"epoch": 3.53,
"learning_rate": 2.1631934253008513e-05,
"loss": 0.004,
"step": 8450
},
{
"epoch": 3.54,
"learning_rate": 2.159524508365131e-05,
"loss": 0.0018,
"step": 8460
},
{
"epoch": 3.54,
"learning_rate": 2.15585559142941e-05,
"loss": 0.0026,
"step": 8470
},
{
"epoch": 3.55,
"learning_rate": 2.1521866744936893e-05,
"loss": 0.0019,
"step": 8480
},
{
"epoch": 3.55,
"learning_rate": 2.148517757557969e-05,
"loss": 0.0019,
"step": 8490
},
{
"epoch": 3.55,
"learning_rate": 2.1448488406222484e-05,
"loss": 0.0017,
"step": 8500
},
{
"epoch": 3.56,
"learning_rate": 2.141179923686528e-05,
"loss": 0.0039,
"step": 8510
},
{
"epoch": 3.56,
"learning_rate": 2.1375110067508074e-05,
"loss": 0.0216,
"step": 8520
},
{
"epoch": 3.57,
"learning_rate": 2.1338420898150867e-05,
"loss": 0.0019,
"step": 8530
},
{
"epoch": 3.57,
"learning_rate": 2.130173172879366e-05,
"loss": 0.0033,
"step": 8540
},
{
"epoch": 3.58,
"learning_rate": 2.1265042559436454e-05,
"loss": 0.0025,
"step": 8550
},
{
"epoch": 3.58,
"learning_rate": 2.122835339007925e-05,
"loss": 0.0019,
"step": 8560
},
{
"epoch": 3.58,
"learning_rate": 2.1191664220722045e-05,
"loss": 0.0025,
"step": 8570
},
{
"epoch": 3.59,
"learning_rate": 2.115497505136484e-05,
"loss": 0.0024,
"step": 8580
},
{
"epoch": 3.59,
"learning_rate": 2.1118285882007632e-05,
"loss": 0.0031,
"step": 8590
},
{
"epoch": 3.6,
"learning_rate": 2.1081596712650425e-05,
"loss": 0.0032,
"step": 8600
},
{
"epoch": 3.6,
"eval_f1": 0.9946927738123031,
"eval_f1_0": 0.9984006617253486,
"eval_f1_1": 0.9973827694885125,
"eval_f1_2": 0.9841640895298778,
"eval_f1_3": 0.9988235745054738,
"eval_loss": 0.011195490136742592,
"eval_runtime": 32.6122,
"eval_samples_per_second": 260.7,
"eval_steps_per_second": 1.043,
"step": 8600
},
{
"epoch": 3.6,
"learning_rate": 2.1044907543293222e-05,
"loss": 0.025,
"step": 8610
},
{
"epoch": 3.61,
"learning_rate": 2.1008218373936016e-05,
"loss": 0.0021,
"step": 8620
},
{
"epoch": 3.61,
"learning_rate": 2.097152920457881e-05,
"loss": 0.0026,
"step": 8630
},
{
"epoch": 3.61,
"learning_rate": 2.0934840035221603e-05,
"loss": 0.0028,
"step": 8640
},
{
"epoch": 3.62,
"learning_rate": 2.0898150865864396e-05,
"loss": 0.0022,
"step": 8650
},
{
"epoch": 3.62,
"learning_rate": 2.0861461696507193e-05,
"loss": 0.0075,
"step": 8660
},
{
"epoch": 3.63,
"learning_rate": 2.0824772527149987e-05,
"loss": 0.0027,
"step": 8670
},
{
"epoch": 3.63,
"learning_rate": 2.078808335779278e-05,
"loss": 0.0024,
"step": 8680
},
{
"epoch": 3.63,
"learning_rate": 2.0751394188435574e-05,
"loss": 0.002,
"step": 8690
},
{
"epoch": 3.64,
"learning_rate": 2.071470501907837e-05,
"loss": 0.0026,
"step": 8700
},
{
"epoch": 3.64,
"learning_rate": 2.0678015849721164e-05,
"loss": 0.0104,
"step": 8710
},
{
"epoch": 3.65,
"learning_rate": 2.0641326680363958e-05,
"loss": 0.0024,
"step": 8720
},
{
"epoch": 3.65,
"learning_rate": 2.060463751100675e-05,
"loss": 0.0123,
"step": 8730
},
{
"epoch": 3.66,
"learning_rate": 2.0567948341649545e-05,
"loss": 0.0038,
"step": 8740
},
{
"epoch": 3.66,
"learning_rate": 2.053125917229234e-05,
"loss": 0.0039,
"step": 8750
},
{
"epoch": 3.66,
"learning_rate": 2.0494570002935135e-05,
"loss": 0.0272,
"step": 8760
},
{
"epoch": 3.67,
"learning_rate": 2.045788083357793e-05,
"loss": 0.0022,
"step": 8770
},
{
"epoch": 3.67,
"learning_rate": 2.0421191664220722e-05,
"loss": 0.0021,
"step": 8780
},
{
"epoch": 3.68,
"learning_rate": 2.0384502494863516e-05,
"loss": 0.0027,
"step": 8790
},
{
"epoch": 3.68,
"learning_rate": 2.0347813325506312e-05,
"loss": 0.0105,
"step": 8800
},
{
"epoch": 3.68,
"eval_f1": 0.9945755834792885,
"eval_f1_0": 0.9983544586712249,
"eval_f1_1": 0.9973788996685247,
"eval_f1_2": 0.9837011178027719,
"eval_f1_3": 0.9988678577746329,
"eval_loss": 0.010032573714852333,
"eval_runtime": 32.3921,
"eval_samples_per_second": 262.472,
"eval_steps_per_second": 1.05,
"step": 8800
},
{
"epoch": 3.68,
"learning_rate": 2.0311124156149106e-05,
"loss": 0.0178,
"step": 8810
},
{
"epoch": 3.69,
"learning_rate": 2.02744349867919e-05,
"loss": 0.0029,
"step": 8820
},
{
"epoch": 3.69,
"learning_rate": 2.0237745817434693e-05,
"loss": 0.0017,
"step": 8830
},
{
"epoch": 3.7,
"learning_rate": 2.0201056648077486e-05,
"loss": 0.0211,
"step": 8840
},
{
"epoch": 3.7,
"learning_rate": 2.0164367478720283e-05,
"loss": 0.0049,
"step": 8850
},
{
"epoch": 3.71,
"learning_rate": 2.0127678309363077e-05,
"loss": 0.0023,
"step": 8860
},
{
"epoch": 3.71,
"learning_rate": 2.0090989140005874e-05,
"loss": 0.0029,
"step": 8870
},
{
"epoch": 3.71,
"learning_rate": 2.0054299970648667e-05,
"loss": 0.003,
"step": 8880
},
{
"epoch": 3.72,
"learning_rate": 2.0017610801291457e-05,
"loss": 0.0094,
"step": 8890
},
{
"epoch": 3.72,
"learning_rate": 1.9980921631934254e-05,
"loss": 0.0015,
"step": 8900
},
{
"epoch": 3.73,
"learning_rate": 1.9944232462577048e-05,
"loss": 0.0021,
"step": 8910
},
{
"epoch": 3.73,
"learning_rate": 1.9907543293219845e-05,
"loss": 0.005,
"step": 8920
},
{
"epoch": 3.73,
"learning_rate": 1.9870854123862638e-05,
"loss": 0.0017,
"step": 8930
},
{
"epoch": 3.74,
"learning_rate": 1.9834164954505428e-05,
"loss": 0.0042,
"step": 8940
},
{
"epoch": 3.74,
"learning_rate": 1.9797475785148225e-05,
"loss": 0.0024,
"step": 8950
},
{
"epoch": 3.75,
"learning_rate": 1.976078661579102e-05,
"loss": 0.0021,
"step": 8960
},
{
"epoch": 3.75,
"learning_rate": 1.9724097446433816e-05,
"loss": 0.0031,
"step": 8970
},
{
"epoch": 3.76,
"learning_rate": 1.968740827707661e-05,
"loss": 0.0022,
"step": 8980
},
{
"epoch": 3.76,
"learning_rate": 1.96507191077194e-05,
"loss": 0.0274,
"step": 8990
},
{
"epoch": 3.76,
"learning_rate": 1.9614029938362196e-05,
"loss": 0.0041,
"step": 9000
},
{
"epoch": 3.76,
"eval_f1": 0.9948214097130661,
"eval_f1_0": 0.9984300524144331,
"eval_f1_1": 0.9974088795631298,
"eval_f1_2": 0.9846053119495689,
"eval_f1_3": 0.9988413949251329,
"eval_loss": 0.009919887408614159,
"eval_runtime": 32.5423,
"eval_samples_per_second": 261.26,
"eval_steps_per_second": 1.045,
"step": 9000
},
{
"epoch": 3.77,
"learning_rate": 1.957734076900499e-05,
"loss": 0.0061,
"step": 9010
},
{
"epoch": 3.77,
"learning_rate": 1.9540651599647786e-05,
"loss": 0.0033,
"step": 9020
},
{
"epoch": 3.78,
"learning_rate": 1.950396243029058e-05,
"loss": 0.0032,
"step": 9030
},
{
"epoch": 3.78,
"learning_rate": 1.9467273260933373e-05,
"loss": 0.0031,
"step": 9040
},
{
"epoch": 3.79,
"learning_rate": 1.9430584091576167e-05,
"loss": 0.0034,
"step": 9050
},
{
"epoch": 3.79,
"learning_rate": 1.939389492221896e-05,
"loss": 0.0026,
"step": 9060
},
{
"epoch": 3.79,
"learning_rate": 1.9357205752861757e-05,
"loss": 0.0023,
"step": 9070
},
{
"epoch": 3.8,
"learning_rate": 1.932051658350455e-05,
"loss": 0.002,
"step": 9080
},
{
"epoch": 3.8,
"learning_rate": 1.9283827414147344e-05,
"loss": 0.003,
"step": 9090
},
{
"epoch": 3.81,
"learning_rate": 1.9247138244790138e-05,
"loss": 0.0028,
"step": 9100
},
{
"epoch": 3.81,
"learning_rate": 1.921044907543293e-05,
"loss": 0.0032,
"step": 9110
},
{
"epoch": 3.81,
"learning_rate": 1.9173759906075728e-05,
"loss": 0.0039,
"step": 9120
},
{
"epoch": 3.82,
"learning_rate": 1.9137070736718522e-05,
"loss": 0.0019,
"step": 9130
},
{
"epoch": 3.82,
"learning_rate": 1.9100381567361315e-05,
"loss": 0.0037,
"step": 9140
},
{
"epoch": 3.83,
"learning_rate": 1.906369239800411e-05,
"loss": 0.0027,
"step": 9150
},
{
"epoch": 3.83,
"learning_rate": 1.9027003228646902e-05,
"loss": 0.002,
"step": 9160
},
{
"epoch": 3.84,
"learning_rate": 1.89903140592897e-05,
"loss": 0.0081,
"step": 9170
},
{
"epoch": 3.84,
"learning_rate": 1.8953624889932493e-05,
"loss": 0.0024,
"step": 9180
},
{
"epoch": 3.84,
"learning_rate": 1.8916935720575286e-05,
"loss": 0.0018,
"step": 9190
},
{
"epoch": 3.85,
"learning_rate": 1.888024655121808e-05,
"loss": 0.0028,
"step": 9200
},
{
"epoch": 3.85,
"eval_f1": 0.9949803917828218,
"eval_f1_0": 0.9984804396876756,
"eval_f1_1": 0.9974206826757432,
"eval_f1_2": 0.9851408268771353,
"eval_f1_3": 0.9988796178907332,
"eval_loss": 0.010286178439855576,
"eval_runtime": 32.4074,
"eval_samples_per_second": 262.348,
"eval_steps_per_second": 1.049,
"step": 9200
},
{
"epoch": 3.85,
"learning_rate": 1.8843557381860877e-05,
"loss": 0.0033,
"step": 9210
},
{
"epoch": 3.86,
"learning_rate": 1.880686821250367e-05,
"loss": 0.0029,
"step": 9220
},
{
"epoch": 3.86,
"learning_rate": 1.8770179043146464e-05,
"loss": 0.0025,
"step": 9230
},
{
"epoch": 3.86,
"learning_rate": 1.873348987378926e-05,
"loss": 0.0265,
"step": 9240
},
{
"epoch": 3.87,
"learning_rate": 1.869680070443205e-05,
"loss": 0.0241,
"step": 9250
},
{
"epoch": 3.87,
"learning_rate": 1.8660111535074848e-05,
"loss": 0.0024,
"step": 9260
},
{
"epoch": 3.88,
"learning_rate": 1.862342236571764e-05,
"loss": 0.0031,
"step": 9270
},
{
"epoch": 3.88,
"learning_rate": 1.8586733196360435e-05,
"loss": 0.0045,
"step": 9280
},
{
"epoch": 3.89,
"learning_rate": 1.855004402700323e-05,
"loss": 0.0022,
"step": 9290
},
{
"epoch": 3.89,
"learning_rate": 1.851335485764602e-05,
"loss": 0.0065,
"step": 9300
},
{
"epoch": 3.89,
"learning_rate": 1.847666568828882e-05,
"loss": 0.0021,
"step": 9310
},
{
"epoch": 3.9,
"learning_rate": 1.8439976518931612e-05,
"loss": 0.0025,
"step": 9320
},
{
"epoch": 3.9,
"learning_rate": 1.8403287349574405e-05,
"loss": 0.0018,
"step": 9330
},
{
"epoch": 3.91,
"learning_rate": 1.8366598180217202e-05,
"loss": 0.0025,
"step": 9340
},
{
"epoch": 3.91,
"learning_rate": 1.8329909010859992e-05,
"loss": 0.003,
"step": 9350
},
{
"epoch": 3.91,
"learning_rate": 1.829321984150279e-05,
"loss": 0.0043,
"step": 9360
},
{
"epoch": 3.92,
"learning_rate": 1.8256530672145583e-05,
"loss": 0.0015,
"step": 9370
},
{
"epoch": 3.92,
"learning_rate": 1.821984150278838e-05,
"loss": 0.0034,
"step": 9380
},
{
"epoch": 3.93,
"learning_rate": 1.8183152333431173e-05,
"loss": 0.0032,
"step": 9390
},
{
"epoch": 3.93,
"learning_rate": 1.8146463164073967e-05,
"loss": 0.0029,
"step": 9400
},
{
"epoch": 3.93,
"eval_f1": 0.9949698111031801,
"eval_f1_0": 0.9984751134230747,
"eval_f1_1": 0.9973547780742178,
"eval_f1_2": 0.9851696821743005,
"eval_f1_3": 0.9988796707411273,
"eval_loss": 0.010252884589135647,
"eval_runtime": 32.5179,
"eval_samples_per_second": 261.456,
"eval_steps_per_second": 1.046,
"step": 9400
},
{
"epoch": 3.94,
"learning_rate": 1.810977399471676e-05,
"loss": 0.0184,
"step": 9410
},
{
"epoch": 3.94,
"learning_rate": 1.8073084825359554e-05,
"loss": 0.0025,
"step": 9420
},
{
"epoch": 3.94,
"learning_rate": 1.803639565600235e-05,
"loss": 0.002,
"step": 9430
},
{
"epoch": 3.95,
"learning_rate": 1.7999706486645144e-05,
"loss": 0.0079,
"step": 9440
},
{
"epoch": 3.95,
"learning_rate": 1.7963017317287938e-05,
"loss": 0.0063,
"step": 9450
},
{
"epoch": 3.96,
"learning_rate": 1.792632814793073e-05,
"loss": 0.0017,
"step": 9460
},
{
"epoch": 3.96,
"learning_rate": 1.7889638978573525e-05,
"loss": 0.0325,
"step": 9470
},
{
"epoch": 3.96,
"learning_rate": 1.785294980921632e-05,
"loss": 0.0031,
"step": 9480
},
{
"epoch": 3.97,
"learning_rate": 1.7816260639859115e-05,
"loss": 0.0037,
"step": 9490
},
{
"epoch": 3.97,
"learning_rate": 1.777957147050191e-05,
"loss": 0.002,
"step": 9500
},
{
"epoch": 3.98,
"learning_rate": 1.7742882301144702e-05,
"loss": 0.0243,
"step": 9510
},
{
"epoch": 3.98,
"learning_rate": 1.7706193131787496e-05,
"loss": 0.0028,
"step": 9520
},
{
"epoch": 3.99,
"learning_rate": 1.7669503962430292e-05,
"loss": 0.0023,
"step": 9530
},
{
"epoch": 3.99,
"learning_rate": 1.7632814793073086e-05,
"loss": 0.0023,
"step": 9540
},
{
"epoch": 3.99,
"learning_rate": 1.759612562371588e-05,
"loss": 0.0023,
"step": 9550
},
{
"epoch": 4.0,
"learning_rate": 1.7559436454358673e-05,
"loss": 0.0054,
"step": 9560
},
{
"epoch": 4.0,
"learning_rate": 1.7522747285001466e-05,
"loss": 0.0037,
"step": 9570
},
{
"epoch": 4.01,
"learning_rate": 1.7486058115644263e-05,
"loss": 0.0023,
"step": 9580
},
{
"epoch": 4.01,
"learning_rate": 1.7449368946287057e-05,
"loss": 0.0022,
"step": 9590
},
{
"epoch": 4.02,
"learning_rate": 1.7412679776929854e-05,
"loss": 0.002,
"step": 9600
},
{
"epoch": 4.02,
"eval_f1": 0.9949615006981284,
"eval_f1_0": 0.9984786527115893,
"eval_f1_1": 0.9974094295249231,
"eval_f1_2": 0.9850930411883666,
"eval_f1_3": 0.9988648793676346,
"eval_loss": 0.010591627098619938,
"eval_runtime": 32.5761,
"eval_samples_per_second": 260.989,
"eval_steps_per_second": 1.044,
"step": 9600
},
{
"epoch": 4.02,
"learning_rate": 1.7375990607572644e-05,
"loss": 0.002,
"step": 9610
},
{
"epoch": 4.02,
"learning_rate": 1.7339301438215437e-05,
"loss": 0.0053,
"step": 9620
},
{
"epoch": 4.03,
"learning_rate": 1.7302612268858234e-05,
"loss": 0.0022,
"step": 9630
},
{
"epoch": 4.03,
"learning_rate": 1.7265923099501028e-05,
"loss": 0.007,
"step": 9640
},
{
"epoch": 4.04,
"learning_rate": 1.7229233930143825e-05,
"loss": 0.0063,
"step": 9650
},
{
"epoch": 4.04,
"learning_rate": 1.7192544760786615e-05,
"loss": 0.0063,
"step": 9660
},
{
"epoch": 4.04,
"learning_rate": 1.7155855591429412e-05,
"loss": 0.0189,
"step": 9670
},
{
"epoch": 4.05,
"learning_rate": 1.7119166422072205e-05,
"loss": 0.0018,
"step": 9680
},
{
"epoch": 4.05,
"learning_rate": 1.7082477252715e-05,
"loss": 0.0021,
"step": 9690
},
{
"epoch": 4.06,
"learning_rate": 1.7045788083357796e-05,
"loss": 0.0018,
"step": 9700
},
{
"epoch": 4.06,
"learning_rate": 1.7009098914000586e-05,
"loss": 0.0221,
"step": 9710
},
{
"epoch": 4.07,
"learning_rate": 1.6972409744643383e-05,
"loss": 0.0028,
"step": 9720
},
{
"epoch": 4.07,
"learning_rate": 1.6935720575286176e-05,
"loss": 0.0027,
"step": 9730
},
{
"epoch": 4.07,
"learning_rate": 1.689903140592897e-05,
"loss": 0.0023,
"step": 9740
},
{
"epoch": 4.08,
"learning_rate": 1.6862342236571767e-05,
"loss": 0.0024,
"step": 9750
},
{
"epoch": 4.08,
"learning_rate": 1.682565306721456e-05,
"loss": 0.0067,
"step": 9760
},
{
"epoch": 4.09,
"learning_rate": 1.6788963897857354e-05,
"loss": 0.0035,
"step": 9770
},
{
"epoch": 4.09,
"learning_rate": 1.6752274728500147e-05,
"loss": 0.003,
"step": 9780
},
{
"epoch": 4.09,
"learning_rate": 1.671558555914294e-05,
"loss": 0.0034,
"step": 9790
},
{
"epoch": 4.1,
"learning_rate": 1.6678896389785737e-05,
"loss": 0.0035,
"step": 9800
},
{
"epoch": 4.1,
"eval_f1": 0.9950048143092218,
"eval_f1_0": 0.9984898833225512,
"eval_f1_1": 0.9973566579060562,
"eval_f1_2": 0.9853166648725864,
"eval_f1_3": 0.9988560511356935,
"eval_loss": 0.010622835718095303,
"eval_runtime": 32.5826,
"eval_samples_per_second": 260.937,
"eval_steps_per_second": 1.044,
"step": 9800
},
{
"epoch": 4.1,
"learning_rate": 1.664220722042853e-05,
"loss": 0.0021,
"step": 9810
},
{
"epoch": 4.11,
"learning_rate": 1.6605518051071324e-05,
"loss": 0.0058,
"step": 9820
},
{
"epoch": 4.11,
"learning_rate": 1.6568828881714118e-05,
"loss": 0.0032,
"step": 9830
},
{
"epoch": 4.12,
"learning_rate": 1.6532139712356915e-05,
"loss": 0.0035,
"step": 9840
},
{
"epoch": 4.12,
"learning_rate": 1.649545054299971e-05,
"loss": 0.0023,
"step": 9850
},
{
"epoch": 4.12,
"learning_rate": 1.6458761373642502e-05,
"loss": 0.0015,
"step": 9860
},
{
"epoch": 4.13,
"learning_rate": 1.6422072204285295e-05,
"loss": 0.0078,
"step": 9870
},
{
"epoch": 4.13,
"learning_rate": 1.638538303492809e-05,
"loss": 0.0023,
"step": 9880
},
{
"epoch": 4.14,
"learning_rate": 1.6348693865570886e-05,
"loss": 0.0018,
"step": 9890
},
{
"epoch": 4.14,
"learning_rate": 1.631200469621368e-05,
"loss": 0.003,
"step": 9900
},
{
"epoch": 4.14,
"learning_rate": 1.6275315526856473e-05,
"loss": 0.0166,
"step": 9910
},
{
"epoch": 4.15,
"learning_rate": 1.6238626357499266e-05,
"loss": 0.0031,
"step": 9920
},
{
"epoch": 4.15,
"learning_rate": 1.620193718814206e-05,
"loss": 0.0033,
"step": 9930
},
{
"epoch": 4.16,
"learning_rate": 1.6165248018784857e-05,
"loss": 0.002,
"step": 9940
},
{
"epoch": 4.16,
"learning_rate": 1.612855884942765e-05,
"loss": 0.0039,
"step": 9950
},
{
"epoch": 4.17,
"learning_rate": 1.6091869680070444e-05,
"loss": 0.0019,
"step": 9960
},
{
"epoch": 4.17,
"learning_rate": 1.6055180510713237e-05,
"loss": 0.002,
"step": 9970
},
{
"epoch": 4.17,
"learning_rate": 1.601849134135603e-05,
"loss": 0.003,
"step": 9980
},
{
"epoch": 4.18,
"learning_rate": 1.5981802171998828e-05,
"loss": 0.0045,
"step": 9990
},
{
"epoch": 4.18,
"learning_rate": 1.594511300264162e-05,
"loss": 0.0221,
"step": 10000
},
{
"epoch": 4.18,
"eval_f1": 0.9949715452686114,
"eval_f1_0": 0.9984762593755379,
"eval_f1_1": 0.9973998309511085,
"eval_f1_2": 0.9851274684774881,
"eval_f1_3": 0.9988826222703111,
"eval_loss": 0.010112923569977283,
"eval_runtime": 32.6397,
"eval_samples_per_second": 260.48,
"eval_steps_per_second": 1.042,
"step": 10000
},
{
"epoch": 4.19,
"learning_rate": 1.5908423833284418e-05,
"loss": 0.0027,
"step": 10010
},
{
"epoch": 4.19,
"learning_rate": 1.5871734663927208e-05,
"loss": 0.0028,
"step": 10020
},
{
"epoch": 4.19,
"learning_rate": 1.583504549457e-05,
"loss": 0.0136,
"step": 10030
},
{
"epoch": 4.2,
"learning_rate": 1.57983563252128e-05,
"loss": 0.002,
"step": 10040
},
{
"epoch": 4.2,
"learning_rate": 1.5761667155855592e-05,
"loss": 0.0022,
"step": 10050
},
{
"epoch": 4.21,
"learning_rate": 1.572497798649839e-05,
"loss": 0.0033,
"step": 10060
},
{
"epoch": 4.21,
"learning_rate": 1.5688288817141182e-05,
"loss": 0.0027,
"step": 10070
},
{
"epoch": 4.22,
"learning_rate": 1.5651599647783972e-05,
"loss": 0.0016,
"step": 10080
},
{
"epoch": 4.22,
"learning_rate": 1.561491047842677e-05,
"loss": 0.0019,
"step": 10090
},
{
"epoch": 4.22,
"learning_rate": 1.5578221309069563e-05,
"loss": 0.0026,
"step": 10100
},
{
"epoch": 4.23,
"learning_rate": 1.554153213971236e-05,
"loss": 0.0118,
"step": 10110
},
{
"epoch": 4.23,
"learning_rate": 1.5504842970355153e-05,
"loss": 0.0022,
"step": 10120
},
{
"epoch": 4.24,
"learning_rate": 1.5468153800997943e-05,
"loss": 0.007,
"step": 10130
},
{
"epoch": 4.24,
"learning_rate": 1.543146463164074e-05,
"loss": 0.0053,
"step": 10140
},
{
"epoch": 4.25,
"learning_rate": 1.5394775462283534e-05,
"loss": 0.0025,
"step": 10150
},
{
"epoch": 4.25,
"learning_rate": 1.535808629292633e-05,
"loss": 0.0038,
"step": 10160
},
{
"epoch": 4.25,
"learning_rate": 1.5321397123569124e-05,
"loss": 0.0028,
"step": 10170
},
{
"epoch": 4.26,
"learning_rate": 1.5284707954211918e-05,
"loss": 0.0022,
"step": 10180
},
{
"epoch": 4.26,
"learning_rate": 1.5248018784854711e-05,
"loss": 0.0022,
"step": 10190
},
{
"epoch": 4.27,
"learning_rate": 1.5211329615497505e-05,
"loss": 0.0032,
"step": 10200
},
{
"epoch": 4.27,
"eval_f1": 0.9949587645299518,
"eval_f1_0": 0.9984761385940906,
"eval_f1_1": 0.997407323174892,
"eval_f1_2": 0.9850689806691169,
"eval_f1_3": 0.9988826156817077,
"eval_loss": 0.010558743961155415,
"eval_runtime": 32.4942,
"eval_samples_per_second": 261.647,
"eval_steps_per_second": 1.046,
"step": 10200
},
{
"epoch": 4.27,
"learning_rate": 1.5174640446140302e-05,
"loss": 0.0039,
"step": 10210
},
{
"epoch": 4.27,
"learning_rate": 1.5137951276783093e-05,
"loss": 0.003,
"step": 10220
},
{
"epoch": 4.28,
"learning_rate": 1.510126210742589e-05,
"loss": 0.003,
"step": 10230
},
{
"epoch": 4.28,
"learning_rate": 1.5064572938068682e-05,
"loss": 0.0032,
"step": 10240
},
{
"epoch": 4.29,
"learning_rate": 1.5027883768711476e-05,
"loss": 0.0022,
"step": 10250
},
{
"epoch": 4.29,
"learning_rate": 1.4991194599354273e-05,
"loss": 0.0022,
"step": 10260
},
{
"epoch": 4.3,
"learning_rate": 1.4954505429997064e-05,
"loss": 0.0029,
"step": 10270
},
{
"epoch": 4.3,
"learning_rate": 1.4917816260639861e-05,
"loss": 0.0016,
"step": 10280
},
{
"epoch": 4.3,
"learning_rate": 1.4881127091282655e-05,
"loss": 0.0021,
"step": 10290
},
{
"epoch": 4.31,
"learning_rate": 1.4844437921925447e-05,
"loss": 0.0027,
"step": 10300
},
{
"epoch": 4.31,
"learning_rate": 1.4807748752568243e-05,
"loss": 0.0064,
"step": 10310
},
{
"epoch": 4.32,
"learning_rate": 1.4771059583211035e-05,
"loss": 0.0091,
"step": 10320
},
{
"epoch": 4.32,
"learning_rate": 1.4734370413853832e-05,
"loss": 0.003,
"step": 10330
},
{
"epoch": 4.32,
"learning_rate": 1.4697681244496626e-05,
"loss": 0.0033,
"step": 10340
},
{
"epoch": 4.33,
"learning_rate": 1.466099207513942e-05,
"loss": 0.0016,
"step": 10350
},
{
"epoch": 4.33,
"learning_rate": 1.4624302905782214e-05,
"loss": 0.0031,
"step": 10360
},
{
"epoch": 4.34,
"learning_rate": 1.4587613736425008e-05,
"loss": 0.0015,
"step": 10370
},
{
"epoch": 4.34,
"learning_rate": 1.4550924567067803e-05,
"loss": 0.0016,
"step": 10380
},
{
"epoch": 4.35,
"learning_rate": 1.4514235397710597e-05,
"loss": 0.0081,
"step": 10390
},
{
"epoch": 4.35,
"learning_rate": 1.4477546228353392e-05,
"loss": 0.0025,
"step": 10400
},
{
"epoch": 4.35,
"eval_f1": 0.9952178768165258,
"eval_f1_0": 0.9985507536472163,
"eval_f1_1": 0.9973867480267958,
"eval_f1_2": 0.9860572996445295,
"eval_f1_3": 0.998876705947562,
"eval_loss": 0.010311787948012352,
"eval_runtime": 32.4819,
"eval_samples_per_second": 261.746,
"eval_steps_per_second": 1.047,
"step": 10400
},
{
"epoch": 4.35,
"learning_rate": 1.4440857058996185e-05,
"loss": 0.0025,
"step": 10410
},
{
"epoch": 4.36,
"learning_rate": 1.4404167889638979e-05,
"loss": 0.0029,
"step": 10420
},
{
"epoch": 4.36,
"learning_rate": 1.4367478720281774e-05,
"loss": 0.0151,
"step": 10430
},
{
"epoch": 4.37,
"learning_rate": 1.4330789550924567e-05,
"loss": 0.0039,
"step": 10440
},
{
"epoch": 4.37,
"learning_rate": 1.4294100381567363e-05,
"loss": 0.007,
"step": 10450
},
{
"epoch": 4.37,
"learning_rate": 1.4257411212210156e-05,
"loss": 0.0017,
"step": 10460
},
{
"epoch": 4.38,
"learning_rate": 1.422072204285295e-05,
"loss": 0.0022,
"step": 10470
},
{
"epoch": 4.38,
"learning_rate": 1.4184032873495745e-05,
"loss": 0.0033,
"step": 10480
},
{
"epoch": 4.39,
"learning_rate": 1.4147343704138538e-05,
"loss": 0.002,
"step": 10490
},
{
"epoch": 4.39,
"learning_rate": 1.4110654534781334e-05,
"loss": 0.0142,
"step": 10500
},
{
"epoch": 4.4,
"learning_rate": 1.4073965365424127e-05,
"loss": 0.006,
"step": 10510
},
{
"epoch": 4.4,
"learning_rate": 1.4037276196066922e-05,
"loss": 0.0098,
"step": 10520
},
{
"epoch": 4.4,
"learning_rate": 1.4000587026709716e-05,
"loss": 0.0022,
"step": 10530
},
{
"epoch": 4.41,
"learning_rate": 1.396389785735251e-05,
"loss": 0.0026,
"step": 10540
},
{
"epoch": 4.41,
"learning_rate": 1.3927208687995304e-05,
"loss": 0.0068,
"step": 10550
},
{
"epoch": 4.42,
"learning_rate": 1.3890519518638098e-05,
"loss": 0.0034,
"step": 10560
},
{
"epoch": 4.42,
"learning_rate": 1.3853830349280895e-05,
"loss": 0.0062,
"step": 10570
},
{
"epoch": 4.42,
"learning_rate": 1.3817141179923687e-05,
"loss": 0.0021,
"step": 10580
},
{
"epoch": 4.43,
"learning_rate": 1.378045201056648e-05,
"loss": 0.0032,
"step": 10590
},
{
"epoch": 4.43,
"learning_rate": 1.3743762841209275e-05,
"loss": 0.0026,
"step": 10600
},
{
"epoch": 4.43,
"eval_f1": 0.9951592936118213,
"eval_f1_0": 0.9985337624603158,
"eval_f1_1": 0.9973454597129459,
"eval_f1_2": 0.9858871495415432,
"eval_f1_3": 0.9988708027324805,
"eval_loss": 0.010438648983836174,
"eval_runtime": 32.5008,
"eval_samples_per_second": 261.593,
"eval_steps_per_second": 1.046,
"step": 10600
},
{
"epoch": 4.44,
"learning_rate": 1.3707073671852069e-05,
"loss": 0.0045,
"step": 10610
},
{
"epoch": 4.44,
"learning_rate": 1.3670384502494866e-05,
"loss": 0.0044,
"step": 10620
},
{
"epoch": 4.45,
"learning_rate": 1.3633695333137658e-05,
"loss": 0.004,
"step": 10630
},
{
"epoch": 4.45,
"learning_rate": 1.3597006163780451e-05,
"loss": 0.0024,
"step": 10640
},
{
"epoch": 4.45,
"learning_rate": 1.3560316994423248e-05,
"loss": 0.0029,
"step": 10650
},
{
"epoch": 4.46,
"learning_rate": 1.352362782506604e-05,
"loss": 0.0187,
"step": 10660
},
{
"epoch": 4.46,
"learning_rate": 1.3486938655708837e-05,
"loss": 0.0018,
"step": 10670
},
{
"epoch": 4.47,
"learning_rate": 1.3450249486351629e-05,
"loss": 0.003,
"step": 10680
},
{
"epoch": 4.47,
"learning_rate": 1.3413560316994425e-05,
"loss": 0.0022,
"step": 10690
},
{
"epoch": 4.48,
"learning_rate": 1.3376871147637219e-05,
"loss": 0.0049,
"step": 10700
},
{
"epoch": 4.48,
"learning_rate": 1.334018197828001e-05,
"loss": 0.0018,
"step": 10710
},
{
"epoch": 4.48,
"learning_rate": 1.3303492808922808e-05,
"loss": 0.0038,
"step": 10720
},
{
"epoch": 4.49,
"learning_rate": 1.3266803639565601e-05,
"loss": 0.0158,
"step": 10730
},
{
"epoch": 4.49,
"learning_rate": 1.3230114470208396e-05,
"loss": 0.0032,
"step": 10740
},
{
"epoch": 4.5,
"learning_rate": 1.319342530085119e-05,
"loss": 0.0025,
"step": 10750
},
{
"epoch": 4.5,
"learning_rate": 1.3156736131493982e-05,
"loss": 0.0019,
"step": 10760
},
{
"epoch": 4.5,
"learning_rate": 1.3120046962136779e-05,
"loss": 0.0032,
"step": 10770
},
{
"epoch": 4.51,
"learning_rate": 1.3083357792779572e-05,
"loss": 0.002,
"step": 10780
},
{
"epoch": 4.51,
"learning_rate": 1.3046668623422367e-05,
"loss": 0.002,
"step": 10790
},
{
"epoch": 4.52,
"learning_rate": 1.300997945406516e-05,
"loss": 0.0039,
"step": 10800
},
{
"epoch": 4.52,
"eval_f1": 0.9949895546366804,
"eval_f1_0": 0.9984832294695261,
"eval_f1_1": 0.9973774468004472,
"eval_f1_2": 0.9852444496072253,
"eval_f1_3": 0.9988530926695226,
"eval_loss": 0.010345346294343472,
"eval_runtime": 32.5558,
"eval_samples_per_second": 261.152,
"eval_steps_per_second": 1.044,
"step": 10800
},
{
"epoch": 4.52,
"learning_rate": 1.2973290284707954e-05,
"loss": 0.0033,
"step": 10810
},
{
"epoch": 4.53,
"learning_rate": 1.293660111535075e-05,
"loss": 0.0021,
"step": 10820
},
{
"epoch": 4.53,
"learning_rate": 1.2899911945993543e-05,
"loss": 0.0052,
"step": 10830
},
{
"epoch": 4.53,
"learning_rate": 1.2863222776636338e-05,
"loss": 0.0036,
"step": 10840
},
{
"epoch": 4.54,
"learning_rate": 1.2826533607279132e-05,
"loss": 0.0039,
"step": 10850
},
{
"epoch": 4.54,
"learning_rate": 1.2789844437921927e-05,
"loss": 0.0032,
"step": 10860
},
{
"epoch": 4.55,
"learning_rate": 1.275315526856472e-05,
"loss": 0.003,
"step": 10870
},
{
"epoch": 4.55,
"learning_rate": 1.2716466099207514e-05,
"loss": 0.002,
"step": 10880
},
{
"epoch": 4.55,
"learning_rate": 1.2679776929850309e-05,
"loss": 0.0024,
"step": 10890
},
{
"epoch": 4.56,
"learning_rate": 1.2643087760493103e-05,
"loss": 0.0034,
"step": 10900
},
{
"epoch": 4.56,
"learning_rate": 1.2606398591135898e-05,
"loss": 0.0053,
"step": 10910
},
{
"epoch": 4.57,
"learning_rate": 1.2569709421778691e-05,
"loss": 0.0025,
"step": 10920
},
{
"epoch": 4.57,
"learning_rate": 1.2533020252421485e-05,
"loss": 0.0044,
"step": 10930
},
{
"epoch": 4.58,
"learning_rate": 1.249633108306428e-05,
"loss": 0.005,
"step": 10940
},
{
"epoch": 4.58,
"learning_rate": 1.2459641913707075e-05,
"loss": 0.0032,
"step": 10950
},
{
"epoch": 4.58,
"learning_rate": 1.2422952744349869e-05,
"loss": 0.0021,
"step": 10960
},
{
"epoch": 4.59,
"learning_rate": 1.2386263574992662e-05,
"loss": 0.0019,
"step": 10970
},
{
"epoch": 4.59,
"learning_rate": 1.2349574405635457e-05,
"loss": 0.0012,
"step": 10980
},
{
"epoch": 4.6,
"learning_rate": 1.2312885236278251e-05,
"loss": 0.0023,
"step": 10990
},
{
"epoch": 4.6,
"learning_rate": 1.2276196066921046e-05,
"loss": 0.0052,
"step": 11000
},
{
"epoch": 4.6,
"eval_f1": 0.9949347575263713,
"eval_f1_0": 0.9984661405002416,
"eval_f1_1": 0.997367957296102,
"eval_f1_2": 0.9850518396396187,
"eval_f1_3": 0.9988530926695226,
"eval_loss": 0.010258505120873451,
"eval_runtime": 32.6307,
"eval_samples_per_second": 260.552,
"eval_steps_per_second": 1.042,
"step": 11000
},
{
"epoch": 4.6,
"learning_rate": 1.223950689756384e-05,
"loss": 0.0025,
"step": 11010
},
{
"epoch": 4.61,
"learning_rate": 1.2202817728206633e-05,
"loss": 0.0022,
"step": 11020
},
{
"epoch": 4.61,
"learning_rate": 1.2166128558849428e-05,
"loss": 0.0029,
"step": 11030
},
{
"epoch": 4.62,
"learning_rate": 1.2129439389492222e-05,
"loss": 0.0026,
"step": 11040
},
{
"epoch": 4.62,
"learning_rate": 1.2092750220135017e-05,
"loss": 0.0031,
"step": 11050
},
{
"epoch": 4.63,
"learning_rate": 1.2056061050777812e-05,
"loss": 0.0039,
"step": 11060
},
{
"epoch": 4.63,
"learning_rate": 1.2019371881420604e-05,
"loss": 0.0034,
"step": 11070
},
{
"epoch": 4.63,
"learning_rate": 1.19826827120634e-05,
"loss": 0.002,
"step": 11080
},
{
"epoch": 4.64,
"learning_rate": 1.1945993542706194e-05,
"loss": 0.0143,
"step": 11090
},
{
"epoch": 4.64,
"learning_rate": 1.1909304373348988e-05,
"loss": 0.0027,
"step": 11100
},
{
"epoch": 4.65,
"learning_rate": 1.1872615203991783e-05,
"loss": 0.0174,
"step": 11110
},
{
"epoch": 4.65,
"learning_rate": 1.1835926034634577e-05,
"loss": 0.0036,
"step": 11120
},
{
"epoch": 4.65,
"learning_rate": 1.179923686527737e-05,
"loss": 0.0017,
"step": 11130
},
{
"epoch": 4.66,
"learning_rate": 1.1762547695920165e-05,
"loss": 0.0014,
"step": 11140
},
{
"epoch": 4.66,
"learning_rate": 1.1725858526562959e-05,
"loss": 0.0026,
"step": 11150
},
{
"epoch": 4.67,
"learning_rate": 1.1689169357205754e-05,
"loss": 0.0021,
"step": 11160
},
{
"epoch": 4.67,
"learning_rate": 1.1652480187848547e-05,
"loss": 0.0027,
"step": 11170
},
{
"epoch": 4.68,
"learning_rate": 1.1615791018491341e-05,
"loss": 0.0194,
"step": 11180
},
{
"epoch": 4.68,
"learning_rate": 1.1579101849134136e-05,
"loss": 0.0122,
"step": 11190
},
{
"epoch": 4.68,
"learning_rate": 1.154241267977693e-05,
"loss": 0.008,
"step": 11200
},
{
"epoch": 4.68,
"eval_f1": 0.9950206826719094,
"eval_f1_0": 0.9984907277205156,
"eval_f1_1": 0.9973716674467654,
"eval_f1_2": 0.9853554360725758,
"eval_f1_3": 0.998864899447781,
"eval_loss": 0.01006374228745699,
"eval_runtime": 32.5681,
"eval_samples_per_second": 261.053,
"eval_steps_per_second": 1.044,
"step": 11200
},
{
"epoch": 4.69,
"learning_rate": 1.1505723510419725e-05,
"loss": 0.0025,
"step": 11210
},
{
"epoch": 4.69,
"learning_rate": 1.1469034341062518e-05,
"loss": 0.0018,
"step": 11220
},
{
"epoch": 4.7,
"learning_rate": 1.1432345171705314e-05,
"loss": 0.0029,
"step": 11230
},
{
"epoch": 4.7,
"learning_rate": 1.1395656002348107e-05,
"loss": 0.0025,
"step": 11240
},
{
"epoch": 4.71,
"learning_rate": 1.13589668329909e-05,
"loss": 0.002,
"step": 11250
},
{
"epoch": 4.71,
"learning_rate": 1.1322277663633696e-05,
"loss": 0.0018,
"step": 11260
},
{
"epoch": 4.71,
"learning_rate": 1.1285588494276491e-05,
"loss": 0.0034,
"step": 11270
},
{
"epoch": 4.72,
"learning_rate": 1.1248899324919285e-05,
"loss": 0.0026,
"step": 11280
},
{
"epoch": 4.72,
"learning_rate": 1.121221015556208e-05,
"loss": 0.0021,
"step": 11290
},
{
"epoch": 4.73,
"learning_rate": 1.1175520986204872e-05,
"loss": 0.0024,
"step": 11300
},
{
"epoch": 4.73,
"learning_rate": 1.1138831816847667e-05,
"loss": 0.0079,
"step": 11310
},
{
"epoch": 4.73,
"learning_rate": 1.1102142647490462e-05,
"loss": 0.0019,
"step": 11320
},
{
"epoch": 4.74,
"learning_rate": 1.1065453478133255e-05,
"loss": 0.0024,
"step": 11330
},
{
"epoch": 4.74,
"learning_rate": 1.102876430877605e-05,
"loss": 0.0039,
"step": 11340
},
{
"epoch": 4.75,
"learning_rate": 1.0992075139418844e-05,
"loss": 0.004,
"step": 11350
},
{
"epoch": 4.75,
"learning_rate": 1.0955385970061638e-05,
"loss": 0.002,
"step": 11360
},
{
"epoch": 4.76,
"learning_rate": 1.0918696800704433e-05,
"loss": 0.0017,
"step": 11370
},
{
"epoch": 4.76,
"learning_rate": 1.0882007631347226e-05,
"loss": 0.0051,
"step": 11380
},
{
"epoch": 4.76,
"learning_rate": 1.0845318461990022e-05,
"loss": 0.0029,
"step": 11390
},
{
"epoch": 4.77,
"learning_rate": 1.0808629292632815e-05,
"loss": 0.0023,
"step": 11400
},
{
"epoch": 4.77,
"eval_f1": 0.9950140081754068,
"eval_f1_0": 0.9984891951915634,
"eval_f1_1": 0.9973849273275094,
"eval_f1_2": 0.9853288310391183,
"eval_f1_3": 0.9988530791434359,
"eval_loss": 0.010309702716767788,
"eval_runtime": 32.5852,
"eval_samples_per_second": 260.916,
"eval_steps_per_second": 1.043,
"step": 11400
},
{
"epoch": 4.77,
"learning_rate": 1.0771940123275609e-05,
"loss": 0.002,
"step": 11410
},
{
"epoch": 4.78,
"learning_rate": 1.0735250953918404e-05,
"loss": 0.0024,
"step": 11420
},
{
"epoch": 4.78,
"learning_rate": 1.0698561784561197e-05,
"loss": 0.0033,
"step": 11430
},
{
"epoch": 4.78,
"learning_rate": 1.0661872615203992e-05,
"loss": 0.0314,
"step": 11440
},
{
"epoch": 4.79,
"learning_rate": 1.0625183445846788e-05,
"loss": 0.0054,
"step": 11450
},
{
"epoch": 4.79,
"learning_rate": 1.0588494276489581e-05,
"loss": 0.0018,
"step": 11460
},
{
"epoch": 4.8,
"learning_rate": 1.0551805107132375e-05,
"loss": 0.0223,
"step": 11470
},
{
"epoch": 4.8,
"learning_rate": 1.051511593777517e-05,
"loss": 0.0031,
"step": 11480
},
{
"epoch": 4.81,
"learning_rate": 1.0478426768417963e-05,
"loss": 0.0042,
"step": 11490
},
{
"epoch": 4.81,
"learning_rate": 1.0441737599060759e-05,
"loss": 0.0031,
"step": 11500
},
{
"epoch": 4.81,
"learning_rate": 1.0405048429703552e-05,
"loss": 0.0028,
"step": 11510
},
{
"epoch": 4.82,
"learning_rate": 1.0368359260346346e-05,
"loss": 0.0025,
"step": 11520
},
{
"epoch": 4.82,
"learning_rate": 1.033167009098914e-05,
"loss": 0.0044,
"step": 11530
},
{
"epoch": 4.83,
"learning_rate": 1.0294980921631934e-05,
"loss": 0.0041,
"step": 11540
},
{
"epoch": 4.83,
"learning_rate": 1.025829175227473e-05,
"loss": 0.0024,
"step": 11550
},
{
"epoch": 4.83,
"learning_rate": 1.0221602582917523e-05,
"loss": 0.0022,
"step": 11560
},
{
"epoch": 4.84,
"learning_rate": 1.0184913413560318e-05,
"loss": 0.002,
"step": 11570
},
{
"epoch": 4.84,
"learning_rate": 1.0148224244203112e-05,
"loss": 0.0156,
"step": 11580
},
{
"epoch": 4.85,
"learning_rate": 1.0111535074845905e-05,
"loss": 0.0026,
"step": 11590
},
{
"epoch": 4.85,
"learning_rate": 1.00748459054887e-05,
"loss": 0.0023,
"step": 11600
},
{
"epoch": 4.85,
"eval_f1": 0.9949580181432467,
"eval_f1_0": 0.9984742295273287,
"eval_f1_1": 0.9973811472924625,
"eval_f1_2": 0.9851176996597336,
"eval_f1_3": 0.9988589960934621,
"eval_loss": 0.010243140161037445,
"eval_runtime": 32.5481,
"eval_samples_per_second": 261.214,
"eval_steps_per_second": 1.045,
"step": 11600
},
{
"epoch": 4.86,
"learning_rate": 1.0038156736131494e-05,
"loss": 0.0197,
"step": 11610
},
{
"epoch": 4.86,
"learning_rate": 1.0001467566774289e-05,
"loss": 0.0025,
"step": 11620
},
{
"epoch": 4.86,
"learning_rate": 9.964778397417084e-06,
"loss": 0.0059,
"step": 11630
},
{
"epoch": 4.87,
"learning_rate": 9.928089228059876e-06,
"loss": 0.0203,
"step": 11640
},
{
"epoch": 4.87,
"learning_rate": 9.891400058702671e-06,
"loss": 0.0051,
"step": 11650
},
{
"epoch": 4.88,
"learning_rate": 9.854710889345466e-06,
"loss": 0.0023,
"step": 11660
},
{
"epoch": 4.88,
"learning_rate": 9.81802171998826e-06,
"loss": 0.0028,
"step": 11670
},
{
"epoch": 4.88,
"learning_rate": 9.781332550631055e-06,
"loss": 0.0019,
"step": 11680
},
{
"epoch": 4.89,
"learning_rate": 9.744643381273847e-06,
"loss": 0.0023,
"step": 11690
},
{
"epoch": 4.89,
"learning_rate": 9.707954211916642e-06,
"loss": 0.0024,
"step": 11700
},
{
"epoch": 4.9,
"learning_rate": 9.671265042559437e-06,
"loss": 0.0024,
"step": 11710
},
{
"epoch": 4.9,
"learning_rate": 9.634575873202231e-06,
"loss": 0.002,
"step": 11720
},
{
"epoch": 4.91,
"learning_rate": 9.597886703845026e-06,
"loss": 0.0041,
"step": 11730
},
{
"epoch": 4.91,
"learning_rate": 9.56119753448782e-06,
"loss": 0.0026,
"step": 11740
},
{
"epoch": 4.91,
"learning_rate": 9.524508365130613e-06,
"loss": 0.0023,
"step": 11750
},
{
"epoch": 4.92,
"learning_rate": 9.487819195773408e-06,
"loss": 0.0072,
"step": 11760
},
{
"epoch": 4.92,
"learning_rate": 9.451130026416202e-06,
"loss": 0.0114,
"step": 11770
},
{
"epoch": 4.93,
"learning_rate": 9.414440857058997e-06,
"loss": 0.002,
"step": 11780
},
{
"epoch": 4.93,
"learning_rate": 9.37775168770179e-06,
"loss": 0.0025,
"step": 11790
},
{
"epoch": 4.94,
"learning_rate": 9.341062518344586e-06,
"loss": 0.0031,
"step": 11800
},
{
"epoch": 4.94,
"eval_f1": 0.9950169707602463,
"eval_f1_0": 0.9984893331734841,
"eval_f1_1": 0.9973791976896007,
"eval_f1_2": 0.9853256044702073,
"eval_f1_3": 0.9988737477076933,
"eval_loss": 0.010131525807082653,
"eval_runtime": 32.6241,
"eval_samples_per_second": 260.605,
"eval_steps_per_second": 1.042,
"step": 11800
},
{
"epoch": 4.94,
"learning_rate": 9.30437334898738e-06,
"loss": 0.0048,
"step": 11810
},
{
"epoch": 4.94,
"learning_rate": 9.267684179630173e-06,
"loss": 0.0028,
"step": 11820
},
{
"epoch": 4.95,
"learning_rate": 9.230995010272968e-06,
"loss": 0.0023,
"step": 11830
},
{
"epoch": 4.95,
"learning_rate": 9.194305840915763e-06,
"loss": 0.0021,
"step": 11840
},
{
"epoch": 4.96,
"learning_rate": 9.157616671558557e-06,
"loss": 0.0023,
"step": 11850
},
{
"epoch": 4.96,
"learning_rate": 9.120927502201352e-06,
"loss": 0.0029,
"step": 11860
},
{
"epoch": 4.96,
"learning_rate": 9.084238332844144e-06,
"loss": 0.0024,
"step": 11870
},
{
"epoch": 4.97,
"learning_rate": 9.047549163486939e-06,
"loss": 0.0034,
"step": 11880
},
{
"epoch": 4.97,
"learning_rate": 9.010859994129734e-06,
"loss": 0.0025,
"step": 11890
},
{
"epoch": 4.98,
"learning_rate": 8.974170824772528e-06,
"loss": 0.002,
"step": 11900
},
{
"epoch": 4.98,
"learning_rate": 8.937481655415323e-06,
"loss": 0.0031,
"step": 11910
},
{
"epoch": 4.99,
"learning_rate": 8.900792486058116e-06,
"loss": 0.0025,
"step": 11920
},
{
"epoch": 4.99,
"learning_rate": 8.86410331670091e-06,
"loss": 0.0025,
"step": 11930
},
{
"epoch": 4.99,
"learning_rate": 8.827414147343705e-06,
"loss": 0.0042,
"step": 11940
},
{
"epoch": 5.0,
"learning_rate": 8.790724977986498e-06,
"loss": 0.0067,
"step": 11950
},
{
"epoch": 5.0,
"learning_rate": 8.754035808629294e-06,
"loss": 0.0017,
"step": 11960
},
{
"epoch": 5.01,
"learning_rate": 8.717346639272087e-06,
"loss": 0.003,
"step": 11970
},
{
"epoch": 5.01,
"learning_rate": 8.68065746991488e-06,
"loss": 0.0037,
"step": 11980
},
{
"epoch": 5.01,
"learning_rate": 8.643968300557676e-06,
"loss": 0.0023,
"step": 11990
},
{
"epoch": 5.02,
"learning_rate": 8.60727913120047e-06,
"loss": 0.0021,
"step": 12000
},
{
"epoch": 5.02,
"eval_f1": 0.9949525636250813,
"eval_f1_0": 0.9984716580232431,
"eval_f1_1": 0.9974000280463446,
"eval_f1_2": 0.9850825105496254,
"eval_f1_3": 0.9988560578811125,
"eval_loss": 0.010191216133534908,
"eval_runtime": 32.6102,
"eval_samples_per_second": 260.716,
"eval_steps_per_second": 1.043,
"step": 12000
},
{
"epoch": 5.02,
"learning_rate": 8.570589961843265e-06,
"loss": 0.0018,
"step": 12010
},
{
"epoch": 5.03,
"learning_rate": 8.53390079248606e-06,
"loss": 0.002,
"step": 12020
},
{
"epoch": 5.03,
"learning_rate": 8.497211623128853e-06,
"loss": 0.0053,
"step": 12030
},
{
"epoch": 5.04,
"learning_rate": 8.460522453771647e-06,
"loss": 0.0023,
"step": 12040
},
{
"epoch": 5.04,
"learning_rate": 8.42383328441444e-06,
"loss": 0.0019,
"step": 12050
},
{
"epoch": 5.04,
"learning_rate": 8.387144115057235e-06,
"loss": 0.0015,
"step": 12060
},
{
"epoch": 5.05,
"learning_rate": 8.35045494570003e-06,
"loss": 0.0017,
"step": 12070
},
{
"epoch": 5.05,
"learning_rate": 8.313765776342824e-06,
"loss": 0.002,
"step": 12080
},
{
"epoch": 5.06,
"learning_rate": 8.277076606985618e-06,
"loss": 0.006,
"step": 12090
},
{
"epoch": 5.06,
"learning_rate": 8.240387437628413e-06,
"loss": 0.0018,
"step": 12100
},
{
"epoch": 5.06,
"learning_rate": 8.203698268271206e-06,
"loss": 0.0424,
"step": 12110
},
{
"epoch": 5.07,
"learning_rate": 8.167009098914002e-06,
"loss": 0.0199,
"step": 12120
},
{
"epoch": 5.07,
"learning_rate": 8.130319929556795e-06,
"loss": 0.0017,
"step": 12130
},
{
"epoch": 5.08,
"learning_rate": 8.09363076019959e-06,
"loss": 0.0022,
"step": 12140
},
{
"epoch": 5.08,
"learning_rate": 8.056941590842384e-06,
"loss": 0.0024,
"step": 12150
},
{
"epoch": 5.09,
"learning_rate": 8.020252421485177e-06,
"loss": 0.0028,
"step": 12160
},
{
"epoch": 5.09,
"learning_rate": 7.983563252127972e-06,
"loss": 0.002,
"step": 12170
},
{
"epoch": 5.09,
"learning_rate": 7.946874082770766e-06,
"loss": 0.026,
"step": 12180
},
{
"epoch": 5.1,
"learning_rate": 7.910184913413561e-06,
"loss": 0.0026,
"step": 12190
},
{
"epoch": 5.1,
"learning_rate": 7.873495744056356e-06,
"loss": 0.0021,
"step": 12200
},
{
"epoch": 5.1,
"eval_f1": 0.9949571129143104,
"eval_f1_0": 0.9984739780116615,
"eval_f1_1": 0.9974094786173021,
"eval_f1_2": 0.9850889236565663,
"eval_f1_3": 0.998856071371712,
"eval_loss": 0.010159976780414581,
"eval_runtime": 32.4244,
"eval_samples_per_second": 262.21,
"eval_steps_per_second": 1.049,
"step": 12200
},
{
"epoch": 5.11,
"learning_rate": 7.836806574699148e-06,
"loss": 0.0025,
"step": 12210
},
{
"epoch": 5.11,
"learning_rate": 7.800117405341943e-06,
"loss": 0.0029,
"step": 12220
},
{
"epoch": 5.12,
"learning_rate": 7.763428235984737e-06,
"loss": 0.0019,
"step": 12230
},
{
"epoch": 5.12,
"learning_rate": 7.726739066627532e-06,
"loss": 0.0016,
"step": 12240
},
{
"epoch": 5.12,
"learning_rate": 7.690049897270327e-06,
"loss": 0.0021,
"step": 12250
},
{
"epoch": 5.13,
"learning_rate": 7.653360727913119e-06,
"loss": 0.0144,
"step": 12260
},
{
"epoch": 5.13,
"learning_rate": 7.616671558555914e-06,
"loss": 0.0055,
"step": 12270
},
{
"epoch": 5.14,
"learning_rate": 7.579982389198709e-06,
"loss": 0.0018,
"step": 12280
},
{
"epoch": 5.14,
"learning_rate": 7.543293219841503e-06,
"loss": 0.0044,
"step": 12290
},
{
"epoch": 5.14,
"learning_rate": 7.506604050484297e-06,
"loss": 0.0024,
"step": 12300
},
{
"epoch": 5.15,
"learning_rate": 7.4699148811270926e-06,
"loss": 0.0033,
"step": 12310
},
{
"epoch": 5.15,
"learning_rate": 7.433225711769885e-06,
"loss": 0.0025,
"step": 12320
},
{
"epoch": 5.16,
"learning_rate": 7.3965365424126796e-06,
"loss": 0.0058,
"step": 12330
},
{
"epoch": 5.16,
"learning_rate": 7.359847373055474e-06,
"loss": 0.0021,
"step": 12340
},
{
"epoch": 5.17,
"learning_rate": 7.323158203698269e-06,
"loss": 0.0052,
"step": 12350
},
{
"epoch": 5.17,
"learning_rate": 7.2864690343410635e-06,
"loss": 0.0025,
"step": 12360
},
{
"epoch": 5.17,
"learning_rate": 7.249779864983858e-06,
"loss": 0.0018,
"step": 12370
},
{
"epoch": 5.18,
"learning_rate": 7.2130906956266505e-06,
"loss": 0.0058,
"step": 12380
},
{
"epoch": 5.18,
"learning_rate": 7.176401526269446e-06,
"loss": 0.007,
"step": 12390
},
{
"epoch": 5.19,
"learning_rate": 7.13971235691224e-06,
"loss": 0.0034,
"step": 12400
},
{
"epoch": 5.19,
"eval_f1": 0.9949354140816,
"eval_f1_0": 0.9984670736062381,
"eval_f1_1": 0.9973962379997651,
"eval_f1_2": 0.9850252385251065,
"eval_f1_3": 0.9988531061952903,
"eval_loss": 0.010212548077106476,
"eval_runtime": 32.4928,
"eval_samples_per_second": 261.658,
"eval_steps_per_second": 1.046,
"step": 12400
},
{
"epoch": 5.19,
"learning_rate": 7.103023187555034e-06,
"loss": 0.0026,
"step": 12410
},
{
"epoch": 5.19,
"learning_rate": 7.066334018197829e-06,
"loss": 0.0018,
"step": 12420
},
{
"epoch": 5.2,
"learning_rate": 7.029644848840622e-06,
"loss": 0.002,
"step": 12430
},
{
"epoch": 5.2,
"learning_rate": 6.992955679483417e-06,
"loss": 0.0023,
"step": 12440
},
{
"epoch": 5.21,
"learning_rate": 6.956266510126211e-06,
"loss": 0.0024,
"step": 12450
},
{
"epoch": 5.21,
"learning_rate": 6.919577340769005e-06,
"loss": 0.0038,
"step": 12460
},
{
"epoch": 5.22,
"learning_rate": 6.8828881714118e-06,
"loss": 0.0053,
"step": 12470
},
{
"epoch": 5.22,
"learning_rate": 6.846199002054594e-06,
"loss": 0.0039,
"step": 12480
},
{
"epoch": 5.22,
"learning_rate": 6.8095098326973875e-06,
"loss": 0.0149,
"step": 12490
},
{
"epoch": 5.23,
"learning_rate": 6.772820663340182e-06,
"loss": 0.0016,
"step": 12500
},
{
"epoch": 5.23,
"learning_rate": 6.736131493982976e-06,
"loss": 0.0026,
"step": 12510
},
{
"epoch": 5.24,
"learning_rate": 6.6994423246257706e-06,
"loss": 0.0019,
"step": 12520
},
{
"epoch": 5.24,
"learning_rate": 6.662753155268566e-06,
"loss": 0.007,
"step": 12530
},
{
"epoch": 5.24,
"learning_rate": 6.62606398591136e-06,
"loss": 0.0027,
"step": 12540
},
{
"epoch": 5.25,
"learning_rate": 6.589374816554153e-06,
"loss": 0.0027,
"step": 12550
},
{
"epoch": 5.25,
"learning_rate": 6.552685647196947e-06,
"loss": 0.0023,
"step": 12560
},
{
"epoch": 5.26,
"learning_rate": 6.515996477839742e-06,
"loss": 0.0028,
"step": 12570
},
{
"epoch": 5.26,
"learning_rate": 6.479307308482537e-06,
"loss": 0.0017,
"step": 12580
},
{
"epoch": 5.27,
"learning_rate": 6.442618139125331e-06,
"loss": 0.0032,
"step": 12590
},
{
"epoch": 5.27,
"learning_rate": 6.405928969768124e-06,
"loss": 0.003,
"step": 12600
},
{
"epoch": 5.27,
"eval_f1": 0.994954489327091,
"eval_f1_0": 0.9984730716662295,
"eval_f1_1": 0.9973792572856837,
"eval_f1_2": 0.9851066053511706,
"eval_f1_3": 0.9988590230052803,
"eval_loss": 0.010264579206705093,
"eval_runtime": 32.5663,
"eval_samples_per_second": 261.068,
"eval_steps_per_second": 1.044,
"step": 12600
},
{
"epoch": 5.27,
"learning_rate": 6.369239800410919e-06,
"loss": 0.0036,
"step": 12610
},
{
"epoch": 5.28,
"learning_rate": 6.332550631053713e-06,
"loss": 0.002,
"step": 12620
},
{
"epoch": 5.28,
"learning_rate": 6.2958614616965076e-06,
"loss": 0.011,
"step": 12630
},
{
"epoch": 5.29,
"learning_rate": 6.259172292339302e-06,
"loss": 0.0019,
"step": 12640
},
{
"epoch": 5.29,
"learning_rate": 6.2224831229820954e-06,
"loss": 0.0058,
"step": 12650
},
{
"epoch": 5.29,
"learning_rate": 6.185793953624891e-06,
"loss": 0.0041,
"step": 12660
},
{
"epoch": 5.3,
"learning_rate": 6.149104784267685e-06,
"loss": 0.0022,
"step": 12670
},
{
"epoch": 5.3,
"learning_rate": 6.1124156149104785e-06,
"loss": 0.0015,
"step": 12680
},
{
"epoch": 5.31,
"learning_rate": 6.075726445553273e-06,
"loss": 0.0143,
"step": 12690
},
{
"epoch": 5.31,
"learning_rate": 6.039037276196067e-06,
"loss": 0.002,
"step": 12700
},
{
"epoch": 5.32,
"learning_rate": 6.0023481068388615e-06,
"loss": 0.0021,
"step": 12710
},
{
"epoch": 5.32,
"learning_rate": 5.965658937481656e-06,
"loss": 0.002,
"step": 12720
},
{
"epoch": 5.32,
"learning_rate": 5.928969768124449e-06,
"loss": 0.0015,
"step": 12730
},
{
"epoch": 5.33,
"learning_rate": 5.892280598767244e-06,
"loss": 0.0015,
"step": 12740
},
{
"epoch": 5.33,
"learning_rate": 5.855591429410039e-06,
"loss": 0.0025,
"step": 12750
},
{
"epoch": 5.34,
"learning_rate": 5.8189022600528324e-06,
"loss": 0.0017,
"step": 12760
},
{
"epoch": 5.34,
"learning_rate": 5.782213090695627e-06,
"loss": 0.0014,
"step": 12770
},
{
"epoch": 5.35,
"learning_rate": 5.745523921338421e-06,
"loss": 0.0193,
"step": 12780
},
{
"epoch": 5.35,
"learning_rate": 5.7088347519812155e-06,
"loss": 0.0028,
"step": 12790
},
{
"epoch": 5.35,
"learning_rate": 5.67214558262401e-06,
"loss": 0.0031,
"step": 12800
},
{
"epoch": 5.35,
"eval_f1": 0.9949471144513108,
"eval_f1_0": 0.9984707819490605,
"eval_f1_1": 0.9973735674004002,
"eval_f1_2": 0.9850939539637191,
"eval_f1_3": 0.9988501544920632,
"eval_loss": 0.010289383120834827,
"eval_runtime": 32.4196,
"eval_samples_per_second": 262.249,
"eval_steps_per_second": 1.049,
"step": 12800
},
{
"epoch": 5.36,
"learning_rate": 5.635456413266804e-06,
"loss": 0.006,
"step": 12810
},
{
"epoch": 5.36,
"learning_rate": 5.598767243909598e-06,
"loss": 0.0024,
"step": 12820
},
{
"epoch": 5.37,
"learning_rate": 5.562078074552392e-06,
"loss": 0.0025,
"step": 12830
},
{
"epoch": 5.37,
"learning_rate": 5.525388905195187e-06,
"loss": 0.0021,
"step": 12840
},
{
"epoch": 5.37,
"learning_rate": 5.488699735837981e-06,
"loss": 0.0021,
"step": 12850
},
{
"epoch": 5.38,
"learning_rate": 5.452010566480775e-06,
"loss": 0.0021,
"step": 12860
},
{
"epoch": 5.38,
"learning_rate": 5.4153213971235695e-06,
"loss": 0.0065,
"step": 12870
},
{
"epoch": 5.39,
"learning_rate": 5.378632227766364e-06,
"loss": 0.0139,
"step": 12880
},
{
"epoch": 5.39,
"learning_rate": 5.341943058409158e-06,
"loss": 0.0024,
"step": 12890
},
{
"epoch": 5.4,
"learning_rate": 5.305253889051952e-06,
"loss": 0.0039,
"step": 12900
},
{
"epoch": 5.4,
"learning_rate": 5.268564719694746e-06,
"loss": 0.0026,
"step": 12910
},
{
"epoch": 5.4,
"learning_rate": 5.23187555033754e-06,
"loss": 0.0178,
"step": 12920
},
{
"epoch": 5.41,
"learning_rate": 5.195186380980335e-06,
"loss": 0.0018,
"step": 12930
},
{
"epoch": 5.41,
"learning_rate": 5.158497211623129e-06,
"loss": 0.0017,
"step": 12940
},
{
"epoch": 5.42,
"learning_rate": 5.1218080422659234e-06,
"loss": 0.0026,
"step": 12950
},
{
"epoch": 5.42,
"learning_rate": 5.085118872908717e-06,
"loss": 0.0033,
"step": 12960
},
{
"epoch": 5.42,
"learning_rate": 5.048429703551512e-06,
"loss": 0.0021,
"step": 12970
},
{
"epoch": 5.43,
"learning_rate": 5.0117405341943065e-06,
"loss": 0.0192,
"step": 12980
},
{
"epoch": 5.43,
"learning_rate": 4.9750513648371e-06,
"loss": 0.0024,
"step": 12990
},
{
"epoch": 5.44,
"learning_rate": 4.938362195479894e-06,
"loss": 0.0025,
"step": 13000
},
{
"epoch": 5.44,
"eval_f1": 0.9949550327798844,
"eval_f1_0": 0.9984730470718446,
"eval_f1_1": 0.9973811472924625,
"eval_f1_2": 0.9851157822631673,
"eval_f1_3": 0.9988501544920632,
"eval_loss": 0.01024584285914898,
"eval_runtime": 32.3462,
"eval_samples_per_second": 262.844,
"eval_steps_per_second": 1.051,
"step": 13000
},
{
"epoch": 5.44,
"learning_rate": 4.901673026122689e-06,
"loss": 0.0156,
"step": 13010
},
{
"epoch": 5.45,
"learning_rate": 4.864983856765483e-06,
"loss": 0.0027,
"step": 13020
},
{
"epoch": 5.45,
"learning_rate": 4.828294687408277e-06,
"loss": 0.0048,
"step": 13030
},
{
"epoch": 5.45,
"learning_rate": 4.791605518051072e-06,
"loss": 0.0022,
"step": 13040
},
{
"epoch": 5.46,
"learning_rate": 4.754916348693866e-06,
"loss": 0.0024,
"step": 13050
},
{
"epoch": 5.46,
"learning_rate": 4.7182271793366605e-06,
"loss": 0.0021,
"step": 13060
},
{
"epoch": 5.47,
"learning_rate": 4.681538009979454e-06,
"loss": 0.0027,
"step": 13070
},
{
"epoch": 5.47,
"learning_rate": 4.644848840622248e-06,
"loss": 0.004,
"step": 13080
},
{
"epoch": 5.47,
"learning_rate": 4.608159671265043e-06,
"loss": 0.0032,
"step": 13090
},
{
"epoch": 5.48,
"learning_rate": 4.571470501907837e-06,
"loss": 0.003,
"step": 13100
},
{
"epoch": 5.48,
"learning_rate": 4.534781332550631e-06,
"loss": 0.0034,
"step": 13110
},
{
"epoch": 5.49,
"learning_rate": 4.498092163193426e-06,
"loss": 0.0046,
"step": 13120
},
{
"epoch": 5.49,
"learning_rate": 4.461402993836219e-06,
"loss": 0.0128,
"step": 13130
},
{
"epoch": 5.5,
"learning_rate": 4.424713824479014e-06,
"loss": 0.0027,
"step": 13140
},
{
"epoch": 5.5,
"learning_rate": 4.388024655121809e-06,
"loss": 0.0028,
"step": 13150
},
{
"epoch": 5.5,
"learning_rate": 4.351335485764602e-06,
"loss": 0.0022,
"step": 13160
},
{
"epoch": 5.51,
"learning_rate": 4.314646316407397e-06,
"loss": 0.0023,
"step": 13170
},
{
"epoch": 5.51,
"learning_rate": 4.277957147050191e-06,
"loss": 0.0032,
"step": 13180
},
{
"epoch": 5.52,
"learning_rate": 4.241267977692985e-06,
"loss": 0.0114,
"step": 13190
},
{
"epoch": 5.52,
"learning_rate": 4.20457880833578e-06,
"loss": 0.0021,
"step": 13200
},
{
"epoch": 5.52,
"eval_f1": 0.9949300885740563,
"eval_f1_0": 0.9984650164807667,
"eval_f1_1": 0.9973792672181007,
"eval_f1_2": 0.9850229779278351,
"eval_f1_3": 0.9988530926695226,
"eval_loss": 0.01022119726985693,
"eval_runtime": 32.6065,
"eval_samples_per_second": 260.746,
"eval_steps_per_second": 1.043,
"step": 13200
},
{
"epoch": 5.52,
"learning_rate": 4.167889638978574e-06,
"loss": 0.002,
"step": 13210
},
{
"epoch": 5.53,
"learning_rate": 4.1312004696213675e-06,
"loss": 0.0218,
"step": 13220
},
{
"epoch": 5.53,
"learning_rate": 4.094511300264163e-06,
"loss": 0.0025,
"step": 13230
},
{
"epoch": 5.54,
"learning_rate": 4.057822130906956e-06,
"loss": 0.0028,
"step": 13240
},
{
"epoch": 5.54,
"learning_rate": 4.021132961549751e-06,
"loss": 0.0017,
"step": 13250
},
{
"epoch": 5.55,
"learning_rate": 3.984443792192545e-06,
"loss": 0.0031,
"step": 13260
},
{
"epoch": 5.55,
"learning_rate": 3.947754622835339e-06,
"loss": 0.0052,
"step": 13270
},
{
"epoch": 5.55,
"learning_rate": 3.911065453478134e-06,
"loss": 0.0043,
"step": 13280
},
{
"epoch": 5.56,
"learning_rate": 3.874376284120928e-06,
"loss": 0.0015,
"step": 13290
},
{
"epoch": 5.56,
"learning_rate": 3.8376871147637215e-06,
"loss": 0.0028,
"step": 13300
},
{
"epoch": 5.57,
"learning_rate": 3.8009979454065163e-06,
"loss": 0.0016,
"step": 13310
},
{
"epoch": 5.57,
"learning_rate": 3.7643087760493106e-06,
"loss": 0.0213,
"step": 13320
},
{
"epoch": 5.58,
"learning_rate": 3.7276196066921046e-06,
"loss": 0.0024,
"step": 13330
},
{
"epoch": 5.58,
"learning_rate": 3.690930437334899e-06,
"loss": 0.0034,
"step": 13340
},
{
"epoch": 5.58,
"learning_rate": 3.6542412679776933e-06,
"loss": 0.0024,
"step": 13350
},
{
"epoch": 5.59,
"learning_rate": 3.617552098620487e-06,
"loss": 0.0022,
"step": 13360
},
{
"epoch": 5.59,
"learning_rate": 3.5808629292632815e-06,
"loss": 0.0029,
"step": 13370
},
{
"epoch": 5.6,
"learning_rate": 3.5441737599060763e-06,
"loss": 0.002,
"step": 13380
},
{
"epoch": 5.6,
"learning_rate": 3.50748459054887e-06,
"loss": 0.0047,
"step": 13390
},
{
"epoch": 5.6,
"learning_rate": 3.4707954211916646e-06,
"loss": 0.0271,
"step": 13400
},
{
"epoch": 5.6,
"eval_f1": 0.994927417880647,
"eval_f1_0": 0.9984645513385877,
"eval_f1_1": 0.9973887271845249,
"eval_f1_2": 0.9850121487812029,
"eval_f1_3": 0.9988442442182728,
"eval_loss": 0.010328641161322594,
"eval_runtime": 32.3867,
"eval_samples_per_second": 262.515,
"eval_steps_per_second": 1.05,
"step": 13400
},
{
"epoch": 5.61,
"learning_rate": 3.434106251834459e-06,
"loss": 0.0029,
"step": 13410
},
{
"epoch": 5.61,
"learning_rate": 3.397417082477253e-06,
"loss": 0.0024,
"step": 13420
},
{
"epoch": 5.62,
"learning_rate": 3.3607279131200472e-06,
"loss": 0.0033,
"step": 13430
},
{
"epoch": 5.62,
"learning_rate": 3.324038743762841e-06,
"loss": 0.005,
"step": 13440
},
{
"epoch": 5.63,
"learning_rate": 3.2873495744056355e-06,
"loss": 0.0017,
"step": 13450
},
{
"epoch": 5.63,
"learning_rate": 3.25066040504843e-06,
"loss": 0.0022,
"step": 13460
},
{
"epoch": 5.63,
"learning_rate": 3.2139712356912238e-06,
"loss": 0.0016,
"step": 13470
},
{
"epoch": 5.64,
"learning_rate": 3.177282066334018e-06,
"loss": 0.0056,
"step": 13480
},
{
"epoch": 5.64,
"learning_rate": 3.140592896976813e-06,
"loss": 0.003,
"step": 13490
},
{
"epoch": 5.65,
"learning_rate": 3.103903727619607e-06,
"loss": 0.0024,
"step": 13500
},
{
"epoch": 5.65,
"learning_rate": 3.067214558262401e-06,
"loss": 0.0152,
"step": 13510
},
{
"epoch": 5.65,
"learning_rate": 3.030525388905195e-06,
"loss": 0.003,
"step": 13520
},
{
"epoch": 5.66,
"learning_rate": 2.99383621954799e-06,
"loss": 0.0015,
"step": 13530
},
{
"epoch": 5.66,
"learning_rate": 2.957147050190784e-06,
"loss": 0.0054,
"step": 13540
},
{
"epoch": 5.67,
"learning_rate": 2.920457880833578e-06,
"loss": 0.0031,
"step": 13550
},
{
"epoch": 5.67,
"learning_rate": 2.883768711476372e-06,
"loss": 0.0026,
"step": 13560
},
{
"epoch": 5.68,
"learning_rate": 2.8470795421191665e-06,
"loss": 0.0029,
"step": 13570
},
{
"epoch": 5.68,
"learning_rate": 2.810390372761961e-06,
"loss": 0.0023,
"step": 13580
},
{
"epoch": 5.68,
"learning_rate": 2.7737012034047547e-06,
"loss": 0.0023,
"step": 13590
},
{
"epoch": 5.69,
"learning_rate": 2.7370120340475495e-06,
"loss": 0.0093,
"step": 13600
},
{
"epoch": 5.69,
"eval_f1": 0.9949661674545234,
"eval_f1_0": 0.9984748963053742,
"eval_f1_1": 0.9973754673468344,
"eval_f1_2": 0.985152351675912,
"eval_f1_3": 0.9988619544899728,
"eval_loss": 0.010220842435956001,
"eval_runtime": 32.5579,
"eval_samples_per_second": 261.135,
"eval_steps_per_second": 1.044,
"step": 13600
},
{
"epoch": 5.69,
"learning_rate": 2.7003228646903434e-06,
"loss": 0.0024,
"step": 13610
},
{
"epoch": 5.7,
"learning_rate": 2.6636336953331378e-06,
"loss": 0.0024,
"step": 13620
},
{
"epoch": 5.7,
"learning_rate": 2.626944525975932e-06,
"loss": 0.0021,
"step": 13630
},
{
"epoch": 5.7,
"learning_rate": 2.5902553566187265e-06,
"loss": 0.0029,
"step": 13640
},
{
"epoch": 5.71,
"learning_rate": 2.5535661872615204e-06,
"loss": 0.0049,
"step": 13650
},
{
"epoch": 5.71,
"learning_rate": 2.5168770179043148e-06,
"loss": 0.002,
"step": 13660
},
{
"epoch": 5.72,
"learning_rate": 2.480187848547109e-06,
"loss": 0.0015,
"step": 13670
},
{
"epoch": 5.72,
"learning_rate": 2.443498679189903e-06,
"loss": 0.0023,
"step": 13680
},
{
"epoch": 5.73,
"learning_rate": 2.4068095098326974e-06,
"loss": 0.002,
"step": 13690
},
{
"epoch": 5.73,
"learning_rate": 2.3701203404754918e-06,
"loss": 0.0031,
"step": 13700
},
{
"epoch": 5.73,
"learning_rate": 2.333431171118286e-06,
"loss": 0.0026,
"step": 13710
},
{
"epoch": 5.74,
"learning_rate": 2.29674200176108e-06,
"loss": 0.0015,
"step": 13720
},
{
"epoch": 5.74,
"learning_rate": 2.260052832403875e-06,
"loss": 0.0026,
"step": 13730
},
{
"epoch": 5.75,
"learning_rate": 2.2233636630466687e-06,
"loss": 0.0017,
"step": 13740
},
{
"epoch": 5.75,
"learning_rate": 2.186674493689463e-06,
"loss": 0.0027,
"step": 13750
},
{
"epoch": 5.75,
"learning_rate": 2.149985324332257e-06,
"loss": 0.0025,
"step": 13760
},
{
"epoch": 5.76,
"learning_rate": 2.1132961549750514e-06,
"loss": 0.0053,
"step": 13770
},
{
"epoch": 5.76,
"learning_rate": 2.0766069856178457e-06,
"loss": 0.0024,
"step": 13780
},
{
"epoch": 5.77,
"learning_rate": 2.0399178162606396e-06,
"loss": 0.0046,
"step": 13790
},
{
"epoch": 5.77,
"learning_rate": 2.0032286469034344e-06,
"loss": 0.0025,
"step": 13800
},
{
"epoch": 5.77,
"eval_f1": 0.9949519911102203,
"eval_f1_0": 0.998471221035115,
"eval_f1_1": 0.9973641175099532,
"eval_f1_2": 0.9851165680147006,
"eval_f1_3": 0.9988560578811125,
"eval_loss": 0.010221057571470737,
"eval_runtime": 32.5003,
"eval_samples_per_second": 261.598,
"eval_steps_per_second": 1.046,
"step": 13800
},
{
"epoch": 5.78,
"learning_rate": 1.9665394775462283e-06,
"loss": 0.0016,
"step": 13810
},
{
"epoch": 5.78,
"learning_rate": 1.9298503081890227e-06,
"loss": 0.0019,
"step": 13820
},
{
"epoch": 5.78,
"learning_rate": 1.8931611388318168e-06,
"loss": 0.0022,
"step": 13830
},
{
"epoch": 5.79,
"learning_rate": 1.8564719694746112e-06,
"loss": 0.0036,
"step": 13840
},
{
"epoch": 5.79,
"learning_rate": 1.8197828001174053e-06,
"loss": 0.0049,
"step": 13850
},
{
"epoch": 5.8,
"learning_rate": 1.7830936307601995e-06,
"loss": 0.0025,
"step": 13860
},
{
"epoch": 5.8,
"learning_rate": 1.746404461402994e-06,
"loss": 0.0029,
"step": 13870
},
{
"epoch": 5.81,
"learning_rate": 1.7097152920457882e-06,
"loss": 0.0058,
"step": 13880
},
{
"epoch": 5.81,
"learning_rate": 1.6730261226885823e-06,
"loss": 0.0023,
"step": 13890
},
{
"epoch": 5.81,
"learning_rate": 1.6363369533313769e-06,
"loss": 0.002,
"step": 13900
},
{
"epoch": 5.82,
"learning_rate": 1.599647783974171e-06,
"loss": 0.0031,
"step": 13910
},
{
"epoch": 5.82,
"learning_rate": 1.5629586146169652e-06,
"loss": 0.0033,
"step": 13920
},
{
"epoch": 5.83,
"learning_rate": 1.5262694452597595e-06,
"loss": 0.0041,
"step": 13930
},
{
"epoch": 5.83,
"learning_rate": 1.4895802759025536e-06,
"loss": 0.0031,
"step": 13940
},
{
"epoch": 5.83,
"learning_rate": 1.4528911065453478e-06,
"loss": 0.0028,
"step": 13950
},
{
"epoch": 5.84,
"learning_rate": 1.4162019371881421e-06,
"loss": 0.0029,
"step": 13960
},
{
"epoch": 5.84,
"learning_rate": 1.3795127678309363e-06,
"loss": 0.0025,
"step": 13970
},
{
"epoch": 5.85,
"learning_rate": 1.3428235984737306e-06,
"loss": 0.0028,
"step": 13980
},
{
"epoch": 5.85,
"learning_rate": 1.3061344291165248e-06,
"loss": 0.0045,
"step": 13990
},
{
"epoch": 5.86,
"learning_rate": 1.2694452597593191e-06,
"loss": 0.0098,
"step": 14000
},
{
"epoch": 5.86,
"eval_f1": 0.9949290140199564,
"eval_f1_0": 0.9984645485121445,
"eval_f1_1": 0.9973886875975926,
"eval_f1_2": 0.9850156171986577,
"eval_f1_3": 0.9988472027714307,
"eval_loss": 0.010264057666063309,
"eval_runtime": 32.5008,
"eval_samples_per_second": 261.594,
"eval_steps_per_second": 1.046,
"step": 14000
},
{
"epoch": 5.86,
"learning_rate": 1.2327560904021135e-06,
"loss": 0.003,
"step": 14010
},
{
"epoch": 5.86,
"learning_rate": 1.1960669210449076e-06,
"loss": 0.0024,
"step": 14020
},
{
"epoch": 5.87,
"learning_rate": 1.159377751687702e-06,
"loss": 0.0022,
"step": 14030
},
{
"epoch": 5.87,
"learning_rate": 1.122688582330496e-06,
"loss": 0.0022,
"step": 14040
},
{
"epoch": 5.88,
"learning_rate": 1.0859994129732902e-06,
"loss": 0.0031,
"step": 14050
},
{
"epoch": 5.88,
"learning_rate": 1.0493102436160846e-06,
"loss": 0.0056,
"step": 14060
},
{
"epoch": 5.88,
"learning_rate": 1.0126210742588787e-06,
"loss": 0.0021,
"step": 14070
},
{
"epoch": 5.89,
"learning_rate": 9.75931904901673e-07,
"loss": 0.0022,
"step": 14080
},
{
"epoch": 5.89,
"learning_rate": 9.392427355444672e-07,
"loss": 0.0039,
"step": 14090
},
{
"epoch": 5.9,
"learning_rate": 9.025535661872616e-07,
"loss": 0.0164,
"step": 14100
},
{
"epoch": 5.9,
"learning_rate": 8.658643968300559e-07,
"loss": 0.0016,
"step": 14110
},
{
"epoch": 5.91,
"learning_rate": 8.291752274728501e-07,
"loss": 0.0041,
"step": 14120
},
{
"epoch": 5.91,
"learning_rate": 7.924860581156443e-07,
"loss": 0.0019,
"step": 14130
},
{
"epoch": 5.91,
"learning_rate": 7.557968887584386e-07,
"loss": 0.0022,
"step": 14140
},
{
"epoch": 5.92,
"learning_rate": 7.191077194012328e-07,
"loss": 0.0081,
"step": 14150
},
{
"epoch": 5.92,
"learning_rate": 6.82418550044027e-07,
"loss": 0.003,
"step": 14160
},
{
"epoch": 5.93,
"learning_rate": 6.457293806868213e-07,
"loss": 0.0019,
"step": 14170
},
{
"epoch": 5.93,
"learning_rate": 6.090402113296154e-07,
"loss": 0.002,
"step": 14180
},
{
"epoch": 5.93,
"learning_rate": 5.723510419724098e-07,
"loss": 0.002,
"step": 14190
},
{
"epoch": 5.94,
"learning_rate": 5.35661872615204e-07,
"loss": 0.0046,
"step": 14200
},
{
"epoch": 5.94,
"eval_f1": 0.9949533482927058,
"eval_f1_0": 0.9984721421349605,
"eval_f1_1": 0.9973773772256289,
"eval_f1_2": 0.9851078159291213,
"eval_f1_3": 0.9988560578811125,
"eval_loss": 0.010254278779029846,
"eval_runtime": 32.6225,
"eval_samples_per_second": 260.618,
"eval_steps_per_second": 1.042,
"step": 14200
},
{
"epoch": 5.94,
"learning_rate": 4.989727032579983e-07,
"loss": 0.0047,
"step": 14210
},
{
"epoch": 5.95,
"learning_rate": 4.6228353390079247e-07,
"loss": 0.0029,
"step": 14220
},
{
"epoch": 5.95,
"learning_rate": 4.255943645435867e-07,
"loss": 0.0046,
"step": 14230
},
{
"epoch": 5.96,
"learning_rate": 3.8890519518638096e-07,
"loss": 0.0139,
"step": 14240
},
{
"epoch": 5.96,
"learning_rate": 3.5221602582917526e-07,
"loss": 0.0018,
"step": 14250
},
{
"epoch": 5.96,
"learning_rate": 3.155268564719695e-07,
"loss": 0.003,
"step": 14260
},
{
"epoch": 5.97,
"learning_rate": 2.788376871147637e-07,
"loss": 0.0025,
"step": 14270
},
{
"epoch": 5.97,
"learning_rate": 2.42148517757558e-07,
"loss": 0.002,
"step": 14280
},
{
"epoch": 5.98,
"learning_rate": 2.0545934840035222e-07,
"loss": 0.0087,
"step": 14290
},
{
"epoch": 5.98,
"learning_rate": 1.6877017904314646e-07,
"loss": 0.002,
"step": 14300
},
{
"epoch": 5.98,
"learning_rate": 1.320810096859407e-07,
"loss": 0.0016,
"step": 14310
},
{
"epoch": 5.99,
"learning_rate": 9.539184032873497e-08,
"loss": 0.0032,
"step": 14320
},
{
"epoch": 5.99,
"learning_rate": 5.87026709715292e-08,
"loss": 0.0021,
"step": 14330
},
{
"epoch": 6.0,
"learning_rate": 2.2013501614323454e-08,
"loss": 0.0028,
"step": 14340
},
{
"epoch": 6.0,
"step": 14346,
"total_flos": 2.7863876783570944e+17,
"train_loss": 0.03144244249522162,
"train_runtime": 5824.0528,
"train_samples_per_second": 78.82,
"train_steps_per_second": 2.463
}
],
"logging_steps": 10,
"max_steps": 14346,
"num_train_epochs": 6,
"save_steps": 200,
"total_flos": 2.7863876783570944e+17,
"trial_name": null,
"trial_params": null
}