fydhfzh's picture
Training in progress, step 4000
3f38182 verified
raw
history blame
35.8 kB
{
"best_metric": 0.8784962349040888,
"best_model_checkpoint": "fydhfzh/hubert-classifier-aug-fold-4\\checkpoint-4000",
"epoch": 10.78167115902965,
"eval_steps": 50,
"global_step": 4000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.13,
"eval_accuracy": 0.04588394062078273,
"eval_binary": 0.3221322537112049,
"eval_f1": 0.014671635517615288,
"eval_loss": 3.805781364440918,
"eval_precision": 0.03779948934318612,
"eval_recall": 0.04588394062078273,
"eval_runtime": 1.0685,
"eval_samples_per_second": 693.469,
"eval_steps_per_second": 22.461,
"step": 50
},
{
"epoch": 0.27,
"eval_accuracy": 0.1106612685560054,
"eval_binary": 0.374358974358979,
"eval_f1": 0.03914518880642342,
"eval_loss": 3.335998058319092,
"eval_precision": 0.0301847940928463,
"eval_recall": 0.1106612685560054,
"eval_runtime": 1.0612,
"eval_samples_per_second": 698.296,
"eval_steps_per_second": 22.617,
"step": 100
},
{
"epoch": 0.4,
"eval_accuracy": 0.23346828609986506,
"eval_binary": 0.4604588394062126,
"eval_f1": 0.14009724606429286,
"eval_loss": 3.007000684738159,
"eval_precision": 0.12454766261554863,
"eval_recall": 0.23346828609986506,
"eval_runtime": 1.0735,
"eval_samples_per_second": 690.246,
"eval_steps_per_second": 22.356,
"step": 150
},
{
"epoch": 0.54,
"eval_accuracy": 0.3468286099865047,
"eval_binary": 0.5406207827260503,
"eval_f1": 0.25241630871467685,
"eval_loss": 2.556727647781372,
"eval_precision": 0.24171104417236225,
"eval_recall": 0.3468286099865047,
"eval_runtime": 1.0985,
"eval_samples_per_second": 674.535,
"eval_steps_per_second": 21.847,
"step": 200
},
{
"epoch": 0.67,
"eval_accuracy": 0.42645074224021595,
"eval_binary": 0.5983805668016237,
"eval_f1": 0.35188011958617477,
"eval_loss": 2.177004337310791,
"eval_precision": 0.3602976489469129,
"eval_recall": 0.42645074224021595,
"eval_runtime": 1.1241,
"eval_samples_per_second": 659.219,
"eval_steps_per_second": 21.351,
"step": 250
},
{
"epoch": 0.81,
"eval_accuracy": 0.4966261808367072,
"eval_binary": 0.647098515519572,
"eval_f1": 0.43339387931632517,
"eval_loss": 1.8818293809890747,
"eval_precision": 0.4889656579714102,
"eval_recall": 0.4966261808367072,
"eval_runtime": 1.1323,
"eval_samples_per_second": 654.399,
"eval_steps_per_second": 21.195,
"step": 300
},
{
"epoch": 0.94,
"eval_accuracy": 0.5506072874493927,
"eval_binary": 0.6852901484480469,
"eval_f1": 0.5044747772286964,
"eval_loss": 1.6562384366989136,
"eval_precision": 0.5479471666625184,
"eval_recall": 0.5506072874493927,
"eval_runtime": 1.1248,
"eval_samples_per_second": 658.769,
"eval_steps_per_second": 21.337,
"step": 350
},
{
"epoch": 1.0,
"grad_norm": 4.053846836090088,
"learning_rate": 9.666666666666667e-05,
"loss": 2.9369,
"step": 371
},
{
"epoch": 1.08,
"eval_accuracy": 0.6045883940620783,
"eval_binary": 0.7234817813765205,
"eval_f1": 0.5598557405312042,
"eval_loss": 1.4207106828689575,
"eval_precision": 0.590639979026807,
"eval_recall": 0.6045883940620783,
"eval_runtime": 1.1338,
"eval_samples_per_second": 653.569,
"eval_steps_per_second": 21.168,
"step": 400
},
{
"epoch": 1.21,
"eval_accuracy": 0.6531713900134952,
"eval_binary": 0.7587044534412956,
"eval_f1": 0.6214387578818519,
"eval_loss": 1.256138801574707,
"eval_precision": 0.6601188715535435,
"eval_recall": 0.6531713900134952,
"eval_runtime": 1.1142,
"eval_samples_per_second": 665.023,
"eval_steps_per_second": 21.539,
"step": 450
},
{
"epoch": 1.35,
"eval_accuracy": 0.6653171390013495,
"eval_binary": 0.7655870445344135,
"eval_f1": 0.6326851465486375,
"eval_loss": 1.1726747751235962,
"eval_precision": 0.6850642147547112,
"eval_recall": 0.6653171390013495,
"eval_runtime": 1.176,
"eval_samples_per_second": 630.094,
"eval_steps_per_second": 20.408,
"step": 500
},
{
"epoch": 1.48,
"eval_accuracy": 0.7300944669365722,
"eval_binary": 0.809986504723346,
"eval_f1": 0.7136995231626676,
"eval_loss": 1.0430372953414917,
"eval_precision": 0.7479252809718396,
"eval_recall": 0.7300944669365722,
"eval_runtime": 1.1554,
"eval_samples_per_second": 641.328,
"eval_steps_per_second": 20.772,
"step": 550
},
{
"epoch": 1.62,
"eval_accuracy": 0.728744939271255,
"eval_binary": 0.809986504723346,
"eval_f1": 0.7074703954108517,
"eval_loss": 0.9479284286499023,
"eval_precision": 0.740343928828029,
"eval_recall": 0.728744939271255,
"eval_runtime": 1.1389,
"eval_samples_per_second": 650.651,
"eval_steps_per_second": 21.074,
"step": 600
},
{
"epoch": 1.75,
"eval_accuracy": 0.766531713900135,
"eval_binary": 0.8377867746288786,
"eval_f1": 0.7547522208317685,
"eval_loss": 0.8898987770080566,
"eval_precision": 0.7879262475416322,
"eval_recall": 0.766531713900135,
"eval_runtime": 1.156,
"eval_samples_per_second": 641.001,
"eval_steps_per_second": 20.761,
"step": 650
},
{
"epoch": 1.89,
"eval_accuracy": 0.815114709851552,
"eval_binary": 0.8699055330634268,
"eval_f1": 0.8043569052236664,
"eval_loss": 0.7485046982765198,
"eval_precision": 0.8233385560308638,
"eval_recall": 0.815114709851552,
"eval_runtime": 1.1399,
"eval_samples_per_second": 650.034,
"eval_steps_per_second": 21.054,
"step": 700
},
{
"epoch": 2.0,
"grad_norm": 4.031270980834961,
"learning_rate": 9.333333333333334e-05,
"loss": 1.3588,
"step": 742
},
{
"epoch": 2.02,
"eval_accuracy": 0.7921727395411606,
"eval_binary": 0.8553306342780018,
"eval_f1": 0.7800833014875206,
"eval_loss": 0.7457680702209473,
"eval_precision": 0.8196362229763039,
"eval_recall": 0.7921727395411606,
"eval_runtime": 1.1503,
"eval_samples_per_second": 644.196,
"eval_steps_per_second": 20.865,
"step": 750
},
{
"epoch": 2.16,
"eval_accuracy": 0.8137651821862348,
"eval_binary": 0.8686909581646418,
"eval_f1": 0.8106549167288675,
"eval_loss": 0.7069482803344727,
"eval_precision": 0.8318749701340793,
"eval_recall": 0.8137651821862348,
"eval_runtime": 1.1687,
"eval_samples_per_second": 634.021,
"eval_steps_per_second": 20.535,
"step": 800
},
{
"epoch": 2.29,
"eval_accuracy": 0.8137651821862348,
"eval_binary": 0.8685560053981101,
"eval_f1": 0.8088124908295262,
"eval_loss": 0.6896880269050598,
"eval_precision": 0.8376693376217073,
"eval_recall": 0.8137651821862348,
"eval_runtime": 1.1497,
"eval_samples_per_second": 644.512,
"eval_steps_per_second": 20.875,
"step": 850
},
{
"epoch": 2.43,
"eval_accuracy": 0.8070175438596491,
"eval_binary": 0.8650472334682855,
"eval_f1": 0.8034708035750392,
"eval_loss": 0.6972086429595947,
"eval_precision": 0.8367843338997185,
"eval_recall": 0.8070175438596491,
"eval_runtime": 1.1633,
"eval_samples_per_second": 636.983,
"eval_steps_per_second": 20.631,
"step": 900
},
{
"epoch": 2.56,
"eval_accuracy": 0.8164642375168691,
"eval_binary": 0.8704453441295538,
"eval_f1": 0.813993268748201,
"eval_loss": 0.6781038641929626,
"eval_precision": 0.8366608032093864,
"eval_recall": 0.8164642375168691,
"eval_runtime": 1.1823,
"eval_samples_per_second": 626.741,
"eval_steps_per_second": 20.299,
"step": 950
},
{
"epoch": 2.7,
"eval_accuracy": 0.815114709851552,
"eval_binary": 0.8703103913630216,
"eval_f1": 0.8100265350585438,
"eval_loss": 0.6554349660873413,
"eval_precision": 0.8380368300914861,
"eval_recall": 0.815114709851552,
"eval_runtime": 1.1543,
"eval_samples_per_second": 641.937,
"eval_steps_per_second": 20.791,
"step": 1000
},
{
"epoch": 2.83,
"eval_accuracy": 0.8218623481781376,
"eval_binary": 0.8759784075573539,
"eval_f1": 0.8182895287255976,
"eval_loss": 0.6537468433380127,
"eval_precision": 0.8421630941099861,
"eval_recall": 0.8218623481781376,
"eval_runtime": 1.1679,
"eval_samples_per_second": 634.474,
"eval_steps_per_second": 20.55,
"step": 1050
},
{
"epoch": 2.96,
"eval_accuracy": 0.8407557354925776,
"eval_binary": 0.8882591093117402,
"eval_f1": 0.8327839441497319,
"eval_loss": 0.6294392347335815,
"eval_precision": 0.8517352283784768,
"eval_recall": 0.8407557354925776,
"eval_runtime": 1.1945,
"eval_samples_per_second": 620.327,
"eval_steps_per_second": 20.092,
"step": 1100
},
{
"epoch": 3.0,
"grad_norm": 4.313427925109863,
"learning_rate": 9e-05,
"loss": 0.8613,
"step": 1113
},
{
"epoch": 3.1,
"eval_accuracy": 0.8340080971659919,
"eval_binary": 0.8840755735492567,
"eval_f1": 0.8305322924524646,
"eval_loss": 0.5901665687561035,
"eval_precision": 0.849580143203625,
"eval_recall": 0.8340080971659919,
"eval_runtime": 1.1609,
"eval_samples_per_second": 638.313,
"eval_steps_per_second": 20.674,
"step": 1150
},
{
"epoch": 3.23,
"eval_accuracy": 0.8286099865047234,
"eval_binary": 0.8800269905533059,
"eval_f1": 0.8244801946311869,
"eval_loss": 0.6372344493865967,
"eval_precision": 0.8429155377535945,
"eval_recall": 0.8286099865047234,
"eval_runtime": 1.1722,
"eval_samples_per_second": 632.156,
"eval_steps_per_second": 20.475,
"step": 1200
},
{
"epoch": 3.37,
"eval_accuracy": 0.8502024291497976,
"eval_binary": 0.8960863697705795,
"eval_f1": 0.8459444607208269,
"eval_loss": 0.5361122488975525,
"eval_precision": 0.8694864046584694,
"eval_recall": 0.8502024291497976,
"eval_runtime": 1.1813,
"eval_samples_per_second": 627.265,
"eval_steps_per_second": 20.316,
"step": 1250
},
{
"epoch": 3.5,
"eval_accuracy": 0.8690958164642375,
"eval_binary": 0.9082321187584337,
"eval_f1": 0.8673805683180668,
"eval_loss": 0.5066567659378052,
"eval_precision": 0.8832905286346582,
"eval_recall": 0.8690958164642375,
"eval_runtime": 1.1611,
"eval_samples_per_second": 638.193,
"eval_steps_per_second": 20.67,
"step": 1300
},
{
"epoch": 3.64,
"eval_accuracy": 0.8690958164642375,
"eval_binary": 0.908636977058029,
"eval_f1": 0.8662532335354076,
"eval_loss": 0.4976394772529602,
"eval_precision": 0.8768591769603915,
"eval_recall": 0.8690958164642375,
"eval_runtime": 1.1871,
"eval_samples_per_second": 624.2,
"eval_steps_per_second": 20.217,
"step": 1350
},
{
"epoch": 3.77,
"eval_accuracy": 0.863697705802969,
"eval_binary": 0.9044534412955457,
"eval_f1": 0.858932558800567,
"eval_loss": 0.5320582985877991,
"eval_precision": 0.8738656979547668,
"eval_recall": 0.863697705802969,
"eval_runtime": 1.1707,
"eval_samples_per_second": 632.936,
"eval_steps_per_second": 20.5,
"step": 1400
},
{
"epoch": 3.91,
"eval_accuracy": 0.8529014844804319,
"eval_binary": 0.8963562753036428,
"eval_f1": 0.8490350520713121,
"eval_loss": 0.6004767417907715,
"eval_precision": 0.8682368800789853,
"eval_recall": 0.8529014844804319,
"eval_runtime": 1.1646,
"eval_samples_per_second": 636.271,
"eval_steps_per_second": 20.608,
"step": 1450
},
{
"epoch": 4.0,
"grad_norm": 6.45872163772583,
"learning_rate": 8.666666666666667e-05,
"loss": 0.6427,
"step": 1484
},
{
"epoch": 4.04,
"eval_accuracy": 0.8609986504723347,
"eval_binary": 0.9025641025641017,
"eval_f1": 0.8585759542711945,
"eval_loss": 0.5946248173713684,
"eval_precision": 0.8751688934053783,
"eval_recall": 0.8609986504723347,
"eval_runtime": 1.1421,
"eval_samples_per_second": 648.789,
"eval_steps_per_second": 21.013,
"step": 1500
},
{
"epoch": 4.18,
"eval_accuracy": 0.8690958164642375,
"eval_binary": 0.9079622132253704,
"eval_f1": 0.8669378239477148,
"eval_loss": 0.5153417587280273,
"eval_precision": 0.8819396228303111,
"eval_recall": 0.8690958164642375,
"eval_runtime": 1.1485,
"eval_samples_per_second": 645.203,
"eval_steps_per_second": 20.897,
"step": 1550
},
{
"epoch": 4.31,
"eval_accuracy": 0.8569500674763832,
"eval_binary": 0.8986504723346818,
"eval_f1": 0.855017890388154,
"eval_loss": 0.569421648979187,
"eval_precision": 0.8722095081406823,
"eval_recall": 0.8569500674763832,
"eval_runtime": 1.1556,
"eval_samples_per_second": 641.21,
"eval_steps_per_second": 20.768,
"step": 1600
},
{
"epoch": 4.45,
"eval_accuracy": 0.8717948717948718,
"eval_binary": 0.9094466936572189,
"eval_f1": 0.869835417212902,
"eval_loss": 0.5046663880348206,
"eval_precision": 0.8909411654648616,
"eval_recall": 0.8717948717948718,
"eval_runtime": 1.1301,
"eval_samples_per_second": 655.718,
"eval_steps_per_second": 21.238,
"step": 1650
},
{
"epoch": 4.58,
"eval_accuracy": 0.8717948717948718,
"eval_binary": 0.9099865047233455,
"eval_f1": 0.8699350147474851,
"eval_loss": 0.5270289182662964,
"eval_precision": 0.8804698038503707,
"eval_recall": 0.8717948717948718,
"eval_runtime": 1.1458,
"eval_samples_per_second": 646.682,
"eval_steps_per_second": 20.945,
"step": 1700
},
{
"epoch": 4.72,
"eval_accuracy": 0.8569500674763832,
"eval_binary": 0.9001349527665304,
"eval_f1": 0.8530294562245342,
"eval_loss": 0.5278229117393494,
"eval_precision": 0.8713434076794401,
"eval_recall": 0.8569500674763832,
"eval_runtime": 1.1651,
"eval_samples_per_second": 635.992,
"eval_steps_per_second": 20.599,
"step": 1750
},
{
"epoch": 4.85,
"eval_accuracy": 0.8771929824561403,
"eval_binary": 0.9148448043184877,
"eval_f1": 0.8765031109078509,
"eval_loss": 0.5112797021865845,
"eval_precision": 0.8899662303508457,
"eval_recall": 0.8771929824561403,
"eval_runtime": 1.161,
"eval_samples_per_second": 638.269,
"eval_steps_per_second": 20.673,
"step": 1800
},
{
"epoch": 4.99,
"eval_accuracy": 0.8677462887989204,
"eval_binary": 0.90715249662618,
"eval_f1": 0.864758306975439,
"eval_loss": 0.5557437539100647,
"eval_precision": 0.8762526459894882,
"eval_recall": 0.8677462887989204,
"eval_runtime": 1.2253,
"eval_samples_per_second": 604.742,
"eval_steps_per_second": 19.587,
"step": 1850
},
{
"epoch": 5.0,
"grad_norm": 4.927316188812256,
"learning_rate": 8.333333333333334e-05,
"loss": 0.5222,
"step": 1855
},
{
"epoch": 5.12,
"eval_accuracy": 0.8717948717948718,
"eval_binary": 0.9093117408906874,
"eval_f1": 0.8695718983671247,
"eval_loss": 0.5396840572357178,
"eval_precision": 0.8828931646269107,
"eval_recall": 0.8717948717948718,
"eval_runtime": 1.1447,
"eval_samples_per_second": 647.325,
"eval_steps_per_second": 20.966,
"step": 1900
},
{
"epoch": 5.26,
"eval_accuracy": 0.8785425101214575,
"eval_binary": 0.9141700404858291,
"eval_f1": 0.8778235100479747,
"eval_loss": 0.5459949970245361,
"eval_precision": 0.8931182365392891,
"eval_recall": 0.8785425101214575,
"eval_runtime": 1.1802,
"eval_samples_per_second": 627.864,
"eval_steps_per_second": 20.336,
"step": 1950
},
{
"epoch": 5.39,
"eval_accuracy": 0.8717948717948718,
"eval_binary": 0.909986504723346,
"eval_f1": 0.8702993779030054,
"eval_loss": 0.5364383459091187,
"eval_precision": 0.88322276389078,
"eval_recall": 0.8717948717948718,
"eval_runtime": 1.1678,
"eval_samples_per_second": 634.529,
"eval_steps_per_second": 20.552,
"step": 2000
},
{
"epoch": 5.53,
"eval_accuracy": 0.8717948717948718,
"eval_binary": 0.909986504723346,
"eval_f1": 0.8693471297847463,
"eval_loss": 0.5403723120689392,
"eval_precision": 0.8857783354746918,
"eval_recall": 0.8717948717948718,
"eval_runtime": 1.2223,
"eval_samples_per_second": 606.259,
"eval_steps_per_second": 19.636,
"step": 2050
},
{
"epoch": 5.66,
"eval_accuracy": 0.8690958164642375,
"eval_binary": 0.9071524966261797,
"eval_f1": 0.8663898349797059,
"eval_loss": 0.6065850257873535,
"eval_precision": 0.8798577528334613,
"eval_recall": 0.8690958164642375,
"eval_runtime": 1.1672,
"eval_samples_per_second": 634.865,
"eval_steps_per_second": 20.562,
"step": 2100
},
{
"epoch": 5.8,
"eval_accuracy": 0.8569500674763832,
"eval_binary": 0.8995951417004041,
"eval_f1": 0.8543825643734608,
"eval_loss": 0.609086275100708,
"eval_precision": 0.8699245573739501,
"eval_recall": 0.8569500674763832,
"eval_runtime": 1.1615,
"eval_samples_per_second": 637.949,
"eval_steps_per_second": 20.662,
"step": 2150
},
{
"epoch": 5.93,
"eval_accuracy": 0.863697705802969,
"eval_binary": 0.9044534412955451,
"eval_f1": 0.8615752789891996,
"eval_loss": 0.5661621689796448,
"eval_precision": 0.8777716435611175,
"eval_recall": 0.863697705802969,
"eval_runtime": 1.1549,
"eval_samples_per_second": 641.6,
"eval_steps_per_second": 20.781,
"step": 2200
},
{
"epoch": 6.0,
"grad_norm": 3.95198392868042,
"learning_rate": 8e-05,
"loss": 0.4432,
"step": 2226
},
{
"epoch": 6.06,
"eval_accuracy": 0.8623481781376519,
"eval_binary": 0.9033738191632924,
"eval_f1": 0.8596934053855256,
"eval_loss": 0.6643653512001038,
"eval_precision": 0.8785115319528275,
"eval_recall": 0.8623481781376519,
"eval_runtime": 1.195,
"eval_samples_per_second": 620.059,
"eval_steps_per_second": 20.083,
"step": 2250
},
{
"epoch": 6.2,
"eval_accuracy": 0.8609986504723347,
"eval_binary": 0.9020242914979746,
"eval_f1": 0.858697058942677,
"eval_loss": 0.6462463736534119,
"eval_precision": 0.874255314336286,
"eval_recall": 0.8609986504723347,
"eval_runtime": 1.1742,
"eval_samples_per_second": 631.066,
"eval_steps_per_second": 20.439,
"step": 2300
},
{
"epoch": 6.33,
"eval_accuracy": 0.8771929824561403,
"eval_binary": 0.9130904183535752,
"eval_f1": 0.8725328911535255,
"eval_loss": 0.5374723672866821,
"eval_precision": 0.8823031219792354,
"eval_recall": 0.8771929824561403,
"eval_runtime": 1.1706,
"eval_samples_per_second": 632.998,
"eval_steps_per_second": 20.502,
"step": 2350
},
{
"epoch": 6.47,
"eval_accuracy": 0.863697705802969,
"eval_binary": 0.9044534412955454,
"eval_f1": 0.8571119491194448,
"eval_loss": 0.6624985337257385,
"eval_precision": 0.8729715216557322,
"eval_recall": 0.863697705802969,
"eval_runtime": 1.1671,
"eval_samples_per_second": 634.923,
"eval_steps_per_second": 20.564,
"step": 2400
},
{
"epoch": 6.6,
"eval_accuracy": 0.8704453441295547,
"eval_binary": 0.9105263157894729,
"eval_f1": 0.8676961482191342,
"eval_loss": 0.5533118844032288,
"eval_precision": 0.880092126347187,
"eval_recall": 0.8704453441295547,
"eval_runtime": 1.1466,
"eval_samples_per_second": 646.275,
"eval_steps_per_second": 20.932,
"step": 2450
},
{
"epoch": 6.74,
"eval_accuracy": 0.8744939271255061,
"eval_binary": 0.9122807017543849,
"eval_f1": 0.871742048083642,
"eval_loss": 0.6021706461906433,
"eval_precision": 0.8837557104763583,
"eval_recall": 0.8744939271255061,
"eval_runtime": 1.1589,
"eval_samples_per_second": 639.39,
"eval_steps_per_second": 20.709,
"step": 2500
},
{
"epoch": 6.87,
"eval_accuracy": 0.8663967611336032,
"eval_binary": 0.9063427800269895,
"eval_f1": 0.8625935560840108,
"eval_loss": 0.6515318751335144,
"eval_precision": 0.8741236309050073,
"eval_recall": 0.8663967611336032,
"eval_runtime": 1.1683,
"eval_samples_per_second": 634.261,
"eval_steps_per_second": 20.543,
"step": 2550
},
{
"epoch": 7.0,
"grad_norm": 4.3103928565979,
"learning_rate": 7.666666666666667e-05,
"loss": 0.3999,
"step": 2597
},
{
"epoch": 7.01,
"eval_accuracy": 0.8717948717948718,
"eval_binary": 0.9103913630229407,
"eval_f1": 0.8700599583697827,
"eval_loss": 0.5570635795593262,
"eval_precision": 0.8831367238249829,
"eval_recall": 0.8717948717948718,
"eval_runtime": 1.1765,
"eval_samples_per_second": 629.854,
"eval_steps_per_second": 20.4,
"step": 2600
},
{
"epoch": 7.14,
"eval_accuracy": 0.8785425101214575,
"eval_binary": 0.9151147098515511,
"eval_f1": 0.8766044389904654,
"eval_loss": 0.5620068311691284,
"eval_precision": 0.8879959446761065,
"eval_recall": 0.8785425101214575,
"eval_runtime": 1.1587,
"eval_samples_per_second": 639.508,
"eval_steps_per_second": 20.713,
"step": 2650
},
{
"epoch": 7.28,
"eval_accuracy": 0.8798920377867746,
"eval_binary": 0.9160593792172727,
"eval_f1": 0.8778735845274245,
"eval_loss": 0.5602428317070007,
"eval_precision": 0.887021746434702,
"eval_recall": 0.8798920377867746,
"eval_runtime": 1.1417,
"eval_samples_per_second": 649.02,
"eval_steps_per_second": 21.021,
"step": 2700
},
{
"epoch": 7.41,
"eval_accuracy": 0.8758434547908233,
"eval_binary": 0.9122807017543852,
"eval_f1": 0.8734723052118281,
"eval_loss": 0.5291705131530762,
"eval_precision": 0.884985605937023,
"eval_recall": 0.8758434547908233,
"eval_runtime": 1.1639,
"eval_samples_per_second": 636.647,
"eval_steps_per_second": 20.62,
"step": 2750
},
{
"epoch": 7.55,
"eval_accuracy": 0.873144399460189,
"eval_binary": 0.9109311740890679,
"eval_f1": 0.8728506306960341,
"eval_loss": 0.6643885374069214,
"eval_precision": 0.8896590139505118,
"eval_recall": 0.873144399460189,
"eval_runtime": 1.157,
"eval_samples_per_second": 640.431,
"eval_steps_per_second": 20.743,
"step": 2800
},
{
"epoch": 7.68,
"eval_accuracy": 0.8771929824561403,
"eval_binary": 0.9132253711201073,
"eval_f1": 0.8755511463124777,
"eval_loss": 0.587805449962616,
"eval_precision": 0.8871340834903588,
"eval_recall": 0.8771929824561403,
"eval_runtime": 1.1575,
"eval_samples_per_second": 640.186,
"eval_steps_per_second": 20.735,
"step": 2850
},
{
"epoch": 7.82,
"eval_accuracy": 0.8677462887989204,
"eval_binary": 0.9062078272604579,
"eval_f1": 0.8663130912154735,
"eval_loss": 0.6276022791862488,
"eval_precision": 0.8797659493003622,
"eval_recall": 0.8677462887989204,
"eval_runtime": 1.1612,
"eval_samples_per_second": 638.108,
"eval_steps_per_second": 20.667,
"step": 2900
},
{
"epoch": 7.95,
"eval_accuracy": 0.873144399460189,
"eval_binary": 0.9124156545209167,
"eval_f1": 0.8719900033412215,
"eval_loss": 0.5735969543457031,
"eval_precision": 0.8796167283009387,
"eval_recall": 0.873144399460189,
"eval_runtime": 1.2203,
"eval_samples_per_second": 607.227,
"eval_steps_per_second": 19.667,
"step": 2950
},
{
"epoch": 8.0,
"grad_norm": 5.361498832702637,
"learning_rate": 7.333333333333333e-05,
"loss": 0.3609,
"step": 2968
},
{
"epoch": 8.09,
"eval_accuracy": 0.8771929824561403,
"eval_binary": 0.9143049932523605,
"eval_f1": 0.8768958010180178,
"eval_loss": 0.6093275547027588,
"eval_precision": 0.8875668341255385,
"eval_recall": 0.8771929824561403,
"eval_runtime": 1.1465,
"eval_samples_per_second": 646.287,
"eval_steps_per_second": 20.932,
"step": 3000
},
{
"epoch": 8.22,
"eval_accuracy": 0.873144399460189,
"eval_binary": 0.9099865047233455,
"eval_f1": 0.8728942603104318,
"eval_loss": 0.5862379670143127,
"eval_precision": 0.8861477141841515,
"eval_recall": 0.873144399460189,
"eval_runtime": 1.1249,
"eval_samples_per_second": 658.698,
"eval_steps_per_second": 21.334,
"step": 3050
},
{
"epoch": 8.36,
"eval_accuracy": 0.8839406207827261,
"eval_binary": 0.9179487179487168,
"eval_f1": 0.8812727621395834,
"eval_loss": 0.5736364126205444,
"eval_precision": 0.8869233630367234,
"eval_recall": 0.8839406207827261,
"eval_runtime": 1.1449,
"eval_samples_per_second": 647.229,
"eval_steps_per_second": 20.963,
"step": 3100
},
{
"epoch": 8.49,
"eval_accuracy": 0.8798920377867746,
"eval_binary": 0.915114709851551,
"eval_f1": 0.8781431747563492,
"eval_loss": 0.5536712408065796,
"eval_precision": 0.8903398258661418,
"eval_recall": 0.8798920377867746,
"eval_runtime": 1.128,
"eval_samples_per_second": 656.932,
"eval_steps_per_second": 21.277,
"step": 3150
},
{
"epoch": 8.63,
"eval_accuracy": 0.8771929824561403,
"eval_binary": 0.9137651821862336,
"eval_f1": 0.8748654521289038,
"eval_loss": 0.5788822770118713,
"eval_precision": 0.8854517989133374,
"eval_recall": 0.8771929824561403,
"eval_runtime": 1.1554,
"eval_samples_per_second": 641.322,
"eval_steps_per_second": 20.772,
"step": 3200
},
{
"epoch": 8.76,
"eval_accuracy": 0.8758434547908233,
"eval_binary": 0.9129554655870437,
"eval_f1": 0.873432972085747,
"eval_loss": 0.6053817272186279,
"eval_precision": 0.8816829856303541,
"eval_recall": 0.8758434547908233,
"eval_runtime": 1.1318,
"eval_samples_per_second": 654.685,
"eval_steps_per_second": 21.204,
"step": 3250
},
{
"epoch": 8.89,
"eval_accuracy": 0.8866396761133604,
"eval_binary": 0.9203778677462877,
"eval_f1": 0.8852200459634372,
"eval_loss": 0.5305805206298828,
"eval_precision": 0.8933747127066964,
"eval_recall": 0.8866396761133604,
"eval_runtime": 1.1366,
"eval_samples_per_second": 651.943,
"eval_steps_per_second": 21.116,
"step": 3300
},
{
"epoch": 9.0,
"grad_norm": 2.9827425479888916,
"learning_rate": 7.000898472596586e-05,
"loss": 0.3284,
"step": 3339
},
{
"epoch": 9.03,
"eval_accuracy": 0.8866396761133604,
"eval_binary": 0.9203778677462882,
"eval_f1": 0.8851786363349267,
"eval_loss": 0.5871447920799255,
"eval_precision": 0.8929871747685514,
"eval_recall": 0.8866396761133604,
"eval_runtime": 1.1268,
"eval_samples_per_second": 657.63,
"eval_steps_per_second": 21.3,
"step": 3350
},
{
"epoch": 9.16,
"eval_accuracy": 0.8717948717948718,
"eval_binary": 0.9114709851551945,
"eval_f1": 0.8692786610310018,
"eval_loss": 0.6212562918663025,
"eval_precision": 0.8831649569503822,
"eval_recall": 0.8717948717948718,
"eval_runtime": 1.125,
"eval_samples_per_second": 658.664,
"eval_steps_per_second": 21.333,
"step": 3400
},
{
"epoch": 9.3,
"eval_accuracy": 0.8893387314439946,
"eval_binary": 0.921862348178137,
"eval_f1": 0.8877552195071403,
"eval_loss": 0.5114206671714783,
"eval_precision": 0.8954543919118818,
"eval_recall": 0.8893387314439946,
"eval_runtime": 1.1299,
"eval_samples_per_second": 655.79,
"eval_steps_per_second": 21.24,
"step": 3450
},
{
"epoch": 9.43,
"eval_accuracy": 0.8758434547908233,
"eval_binary": 0.9137651821862343,
"eval_f1": 0.8732295909377273,
"eval_loss": 0.5960059762001038,
"eval_precision": 0.881588392114708,
"eval_recall": 0.8758434547908233,
"eval_runtime": 1.1351,
"eval_samples_per_second": 652.779,
"eval_steps_per_second": 21.143,
"step": 3500
},
{
"epoch": 9.57,
"eval_accuracy": 0.8825910931174089,
"eval_binary": 0.9190283400809708,
"eval_f1": 0.8811241165340947,
"eval_loss": 0.6340000629425049,
"eval_precision": 0.8921271074105084,
"eval_recall": 0.8825910931174089,
"eval_runtime": 1.1227,
"eval_samples_per_second": 660.008,
"eval_steps_per_second": 21.377,
"step": 3550
},
{
"epoch": 9.7,
"eval_accuracy": 0.8758434547908233,
"eval_binary": 0.9126855600539802,
"eval_f1": 0.8735923320711677,
"eval_loss": 0.606035590171814,
"eval_precision": 0.8846766855876167,
"eval_recall": 0.8758434547908233,
"eval_runtime": 1.1444,
"eval_samples_per_second": 647.476,
"eval_steps_per_second": 20.971,
"step": 3600
},
{
"epoch": 9.84,
"eval_accuracy": 0.8704453441295547,
"eval_binary": 0.9099865047233459,
"eval_f1": 0.8673800362820931,
"eval_loss": 0.6465194821357727,
"eval_precision": 0.8800944669365723,
"eval_recall": 0.8704453441295547,
"eval_runtime": 1.1227,
"eval_samples_per_second": 659.991,
"eval_steps_per_second": 21.376,
"step": 3650
},
{
"epoch": 9.97,
"eval_accuracy": 0.8785425101214575,
"eval_binary": 0.9148448043184876,
"eval_f1": 0.8767805082692359,
"eval_loss": 0.6755142211914062,
"eval_precision": 0.8911575117040704,
"eval_recall": 0.8785425101214575,
"eval_runtime": 1.1379,
"eval_samples_per_second": 651.188,
"eval_steps_per_second": 21.091,
"step": 3700
},
{
"epoch": 10.0,
"grad_norm": 4.129298686981201,
"learning_rate": 6.667565139263254e-05,
"loss": 0.3062,
"step": 3710
},
{
"epoch": 10.11,
"eval_accuracy": 0.8785425101214575,
"eval_binary": 0.9161943319838048,
"eval_f1": 0.877249561583927,
"eval_loss": 0.6168139576911926,
"eval_precision": 0.8873360033279062,
"eval_recall": 0.8785425101214575,
"eval_runtime": 1.1239,
"eval_samples_per_second": 659.319,
"eval_steps_per_second": 21.354,
"step": 3750
},
{
"epoch": 10.24,
"eval_accuracy": 0.8812415654520918,
"eval_binary": 0.917543859649122,
"eval_f1": 0.8794928490642383,
"eval_loss": 0.6084047555923462,
"eval_precision": 0.8893881946513525,
"eval_recall": 0.8812415654520918,
"eval_runtime": 1.1288,
"eval_samples_per_second": 656.471,
"eval_steps_per_second": 21.262,
"step": 3800
},
{
"epoch": 10.38,
"eval_accuracy": 0.8704453441295547,
"eval_binary": 0.9106612685560047,
"eval_f1": 0.8674074511392077,
"eval_loss": 0.733109712600708,
"eval_precision": 0.8810352752660445,
"eval_recall": 0.8704453441295547,
"eval_runtime": 1.1179,
"eval_samples_per_second": 662.846,
"eval_steps_per_second": 21.469,
"step": 3850
},
{
"epoch": 10.51,
"eval_accuracy": 0.8704453441295547,
"eval_binary": 0.9085020242914971,
"eval_f1": 0.8681938294778445,
"eval_loss": 0.6585559248924255,
"eval_precision": 0.8790380185117028,
"eval_recall": 0.8704453441295547,
"eval_runtime": 1.1015,
"eval_samples_per_second": 672.696,
"eval_steps_per_second": 21.788,
"step": 3900
},
{
"epoch": 10.65,
"eval_accuracy": 0.865047233468286,
"eval_binary": 0.9035087719298243,
"eval_f1": 0.8619257548463832,
"eval_loss": 0.7438000440597534,
"eval_precision": 0.8722900768447327,
"eval_recall": 0.865047233468286,
"eval_runtime": 1.1141,
"eval_samples_per_second": 665.131,
"eval_steps_per_second": 21.543,
"step": 3950
},
{
"epoch": 10.78,
"eval_accuracy": 0.8798920377867746,
"eval_binary": 0.9157894736842094,
"eval_f1": 0.8784962349040888,
"eval_loss": 0.5920050740242004,
"eval_precision": 0.8895519684993368,
"eval_recall": 0.8798920377867746,
"eval_runtime": 1.1303,
"eval_samples_per_second": 655.59,
"eval_steps_per_second": 21.234,
"step": 4000
}
],
"logging_steps": 500,
"max_steps": 11130,
"num_input_tokens_seen": 0,
"num_train_epochs": 30,
"save_steps": 500,
"total_flos": 5.3953021213009e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}