kobert-sentiment / trainer_state.json
lkw2024's picture
Upload 11 files
1a82a3e verified
{
"best_metric": 0.6842610364683301,
"best_model_checkpoint": "./checkpoints/skt_kobert-base-v1\\checkpoint-3126",
"epoch": 3.0,
"eval_steps": 500,
"global_step": 4689,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03198976327575176,
"grad_norm": 2.0306639671325684,
"learning_rate": 1.9786734911494988e-05,
"loss": 1.013,
"step": 50
},
{
"epoch": 0.06397952655150352,
"grad_norm": 2.591745138168335,
"learning_rate": 1.9573469822989978e-05,
"loss": 0.9463,
"step": 100
},
{
"epoch": 0.09596928982725528,
"grad_norm": 2.405771017074585,
"learning_rate": 1.9360204734484968e-05,
"loss": 0.912,
"step": 150
},
{
"epoch": 0.12795905310300704,
"grad_norm": 3.699129104614258,
"learning_rate": 1.9146939645979955e-05,
"loss": 0.8349,
"step": 200
},
{
"epoch": 0.1599488163787588,
"grad_norm": 3.0695576667785645,
"learning_rate": 1.8933674557474945e-05,
"loss": 0.8967,
"step": 250
},
{
"epoch": 0.19193857965451055,
"grad_norm": 2.762423038482666,
"learning_rate": 1.872040946896993e-05,
"loss": 0.8609,
"step": 300
},
{
"epoch": 0.22392834293026231,
"grad_norm": 5.626348495483398,
"learning_rate": 1.8507144380464918e-05,
"loss": 0.8619,
"step": 350
},
{
"epoch": 0.2559181062060141,
"grad_norm": 2.3783187866210938,
"learning_rate": 1.8293879291959908e-05,
"loss": 0.847,
"step": 400
},
{
"epoch": 0.28790786948176583,
"grad_norm": 2.9742441177368164,
"learning_rate": 1.8080614203454897e-05,
"loss": 0.8161,
"step": 450
},
{
"epoch": 0.3198976327575176,
"grad_norm": 1.8092529773712158,
"learning_rate": 1.7867349114949884e-05,
"loss": 0.7752,
"step": 500
},
{
"epoch": 0.35188739603326935,
"grad_norm": 2.72413969039917,
"learning_rate": 1.7654084026444874e-05,
"loss": 0.8281,
"step": 550
},
{
"epoch": 0.3838771593090211,
"grad_norm": 3.9095618724823,
"learning_rate": 1.744081893793986e-05,
"loss": 0.7984,
"step": 600
},
{
"epoch": 0.41586692258477287,
"grad_norm": 3.821829319000244,
"learning_rate": 1.7227553849434847e-05,
"loss": 0.845,
"step": 650
},
{
"epoch": 0.44785668586052463,
"grad_norm": 6.882596492767334,
"learning_rate": 1.7014288760929837e-05,
"loss": 0.8218,
"step": 700
},
{
"epoch": 0.4798464491362764,
"grad_norm": 3.819617986679077,
"learning_rate": 1.6801023672424827e-05,
"loss": 0.8316,
"step": 750
},
{
"epoch": 0.5118362124120281,
"grad_norm": 2.8060302734375,
"learning_rate": 1.6587758583919813e-05,
"loss": 0.8009,
"step": 800
},
{
"epoch": 0.5438259756877799,
"grad_norm": 6.112952709197998,
"learning_rate": 1.6374493495414803e-05,
"loss": 0.7959,
"step": 850
},
{
"epoch": 0.5758157389635317,
"grad_norm": 6.098555564880371,
"learning_rate": 1.616122840690979e-05,
"loss": 0.822,
"step": 900
},
{
"epoch": 0.6078055022392834,
"grad_norm": 8.28978157043457,
"learning_rate": 1.5947963318404776e-05,
"loss": 0.7781,
"step": 950
},
{
"epoch": 0.6397952655150352,
"grad_norm": 3.776757001876831,
"learning_rate": 1.5734698229899766e-05,
"loss": 0.8017,
"step": 1000
},
{
"epoch": 0.6717850287907869,
"grad_norm": 2.7059381008148193,
"learning_rate": 1.5521433141394756e-05,
"loss": 0.7822,
"step": 1050
},
{
"epoch": 0.7037747920665387,
"grad_norm": 6.732589244842529,
"learning_rate": 1.5308168052889743e-05,
"loss": 0.7965,
"step": 1100
},
{
"epoch": 0.7357645553422905,
"grad_norm": 8.803948402404785,
"learning_rate": 1.5094902964384733e-05,
"loss": 0.772,
"step": 1150
},
{
"epoch": 0.7677543186180422,
"grad_norm": 3.7579758167266846,
"learning_rate": 1.488163787587972e-05,
"loss": 0.877,
"step": 1200
},
{
"epoch": 0.799744081893794,
"grad_norm": 3.5114402770996094,
"learning_rate": 1.4668372787374708e-05,
"loss": 0.7834,
"step": 1250
},
{
"epoch": 0.8317338451695457,
"grad_norm": 3.5781891345977783,
"learning_rate": 1.4455107698869698e-05,
"loss": 0.7746,
"step": 1300
},
{
"epoch": 0.8637236084452975,
"grad_norm": 4.3330278396606445,
"learning_rate": 1.4241842610364684e-05,
"loss": 0.8138,
"step": 1350
},
{
"epoch": 0.8957133717210493,
"grad_norm": 3.0052335262298584,
"learning_rate": 1.4028577521859672e-05,
"loss": 0.7651,
"step": 1400
},
{
"epoch": 0.927703134996801,
"grad_norm": 3.8835694789886475,
"learning_rate": 1.3815312433354662e-05,
"loss": 0.7984,
"step": 1450
},
{
"epoch": 0.9596928982725528,
"grad_norm": 5.093181133270264,
"learning_rate": 1.3602047344849649e-05,
"loss": 0.7943,
"step": 1500
},
{
"epoch": 0.9916826615483045,
"grad_norm": 2.615415096282959,
"learning_rate": 1.3388782256344637e-05,
"loss": 0.7568,
"step": 1550
},
{
"epoch": 1.0,
"eval_accuracy": 0.673064619321817,
"eval_f1": 0.5774602220992764,
"eval_loss": 0.755181610584259,
"eval_runtime": 146.1897,
"eval_samples_per_second": 21.383,
"eval_steps_per_second": 1.341,
"step": 1563
},
{
"epoch": 1.0236724248240563,
"grad_norm": 3.9289746284484863,
"learning_rate": 1.3175517167839627e-05,
"loss": 0.7968,
"step": 1600
},
{
"epoch": 1.055662188099808,
"grad_norm": 3.1284472942352295,
"learning_rate": 1.2962252079334613e-05,
"loss": 0.7693,
"step": 1650
},
{
"epoch": 1.0876519513755598,
"grad_norm": 6.5303544998168945,
"learning_rate": 1.2748986990829602e-05,
"loss": 0.7763,
"step": 1700
},
{
"epoch": 1.1196417146513116,
"grad_norm": 4.240009784698486,
"learning_rate": 1.2535721902324592e-05,
"loss": 0.7499,
"step": 1750
},
{
"epoch": 1.1516314779270633,
"grad_norm": 3.231553554534912,
"learning_rate": 1.2322456813819578e-05,
"loss": 0.8081,
"step": 1800
},
{
"epoch": 1.183621241202815,
"grad_norm": 3.248556613922119,
"learning_rate": 1.2109191725314566e-05,
"loss": 0.7372,
"step": 1850
},
{
"epoch": 1.2156110044785668,
"grad_norm": 4.725924968719482,
"learning_rate": 1.1895926636809556e-05,
"loss": 0.7564,
"step": 1900
},
{
"epoch": 1.2476007677543186,
"grad_norm": 7.154216766357422,
"learning_rate": 1.1682661548304543e-05,
"loss": 0.7358,
"step": 1950
},
{
"epoch": 1.2795905310300704,
"grad_norm": 5.529367923736572,
"learning_rate": 1.1469396459799531e-05,
"loss": 0.7921,
"step": 2000
},
{
"epoch": 1.3115802943058221,
"grad_norm": 2.6566271781921387,
"learning_rate": 1.1256131371294521e-05,
"loss": 0.7684,
"step": 2050
},
{
"epoch": 1.3435700575815739,
"grad_norm": 3.7145371437072754,
"learning_rate": 1.1042866282789508e-05,
"loss": 0.7612,
"step": 2100
},
{
"epoch": 1.3755598208573256,
"grad_norm": 4.2554707527160645,
"learning_rate": 1.0829601194284496e-05,
"loss": 0.7622,
"step": 2150
},
{
"epoch": 1.4075495841330774,
"grad_norm": 3.207913398742676,
"learning_rate": 1.0616336105779486e-05,
"loss": 0.7331,
"step": 2200
},
{
"epoch": 1.4395393474088292,
"grad_norm": 7.0538153648376465,
"learning_rate": 1.0403071017274472e-05,
"loss": 0.7578,
"step": 2250
},
{
"epoch": 1.471529110684581,
"grad_norm": 4.054543972015381,
"learning_rate": 1.018980592876946e-05,
"loss": 0.7711,
"step": 2300
},
{
"epoch": 1.5035188739603327,
"grad_norm": 5.265899658203125,
"learning_rate": 9.976540840264449e-06,
"loss": 0.742,
"step": 2350
},
{
"epoch": 1.5355086372360844,
"grad_norm": 5.515292167663574,
"learning_rate": 9.763275751759437e-06,
"loss": 0.763,
"step": 2400
},
{
"epoch": 1.5674984005118362,
"grad_norm": 3.4068896770477295,
"learning_rate": 9.550010663254427e-06,
"loss": 0.7787,
"step": 2450
},
{
"epoch": 1.599488163787588,
"grad_norm": 3.8786113262176514,
"learning_rate": 9.336745574749414e-06,
"loss": 0.7665,
"step": 2500
},
{
"epoch": 1.6314779270633397,
"grad_norm": 7.389112949371338,
"learning_rate": 9.123480486244403e-06,
"loss": 0.7324,
"step": 2550
},
{
"epoch": 1.6634676903390915,
"grad_norm": 3.790463924407959,
"learning_rate": 8.910215397739392e-06,
"loss": 0.7479,
"step": 2600
},
{
"epoch": 1.6954574536148432,
"grad_norm": 3.296069622039795,
"learning_rate": 8.696950309234378e-06,
"loss": 0.6861,
"step": 2650
},
{
"epoch": 1.727447216890595,
"grad_norm": 3.9884438514709473,
"learning_rate": 8.483685220729368e-06,
"loss": 0.7621,
"step": 2700
},
{
"epoch": 1.7594369801663468,
"grad_norm": 4.0090532302856445,
"learning_rate": 8.270420132224356e-06,
"loss": 0.7455,
"step": 2750
},
{
"epoch": 1.7914267434420985,
"grad_norm": 5.72037935256958,
"learning_rate": 8.057155043719343e-06,
"loss": 0.7615,
"step": 2800
},
{
"epoch": 1.8234165067178503,
"grad_norm": 5.367597579956055,
"learning_rate": 7.843889955214333e-06,
"loss": 0.7478,
"step": 2850
},
{
"epoch": 1.855406269993602,
"grad_norm": 4.492830753326416,
"learning_rate": 7.630624866709321e-06,
"loss": 0.7644,
"step": 2900
},
{
"epoch": 1.8873960332693538,
"grad_norm": 4.395884990692139,
"learning_rate": 7.4173597782043085e-06,
"loss": 0.726,
"step": 2950
},
{
"epoch": 1.9193857965451055,
"grad_norm": 5.388432025909424,
"learning_rate": 7.204094689699297e-06,
"loss": 0.7305,
"step": 3000
},
{
"epoch": 1.9513755598208573,
"grad_norm": 3.6554412841796875,
"learning_rate": 6.990829601194286e-06,
"loss": 0.7312,
"step": 3050
},
{
"epoch": 1.983365323096609,
"grad_norm": 2.9544596672058105,
"learning_rate": 6.777564512689273e-06,
"loss": 0.7535,
"step": 3100
},
{
"epoch": 2.0,
"eval_accuracy": 0.6842610364683301,
"eval_f1": 0.5914024960556081,
"eval_loss": 0.71613609790802,
"eval_runtime": 145.4665,
"eval_samples_per_second": 21.489,
"eval_steps_per_second": 1.347,
"step": 3126
},
{
"epoch": 2.015355086372361,
"grad_norm": 2.3898167610168457,
"learning_rate": 6.5642994241842614e-06,
"loss": 0.7209,
"step": 3150
},
{
"epoch": 2.0473448496481126,
"grad_norm": 3.216728448867798,
"learning_rate": 6.3510343356792505e-06,
"loss": 0.7328,
"step": 3200
},
{
"epoch": 2.0793346129238643,
"grad_norm": 5.2315673828125,
"learning_rate": 6.137769247174238e-06,
"loss": 0.6979,
"step": 3250
},
{
"epoch": 2.111324376199616,
"grad_norm": 5.086841583251953,
"learning_rate": 5.924504158669226e-06,
"loss": 0.7401,
"step": 3300
},
{
"epoch": 2.143314139475368,
"grad_norm": 3.975651502609253,
"learning_rate": 5.711239070164215e-06,
"loss": 0.7695,
"step": 3350
},
{
"epoch": 2.1753039027511196,
"grad_norm": 5.14952278137207,
"learning_rate": 5.497973981659203e-06,
"loss": 0.7259,
"step": 3400
},
{
"epoch": 2.2072936660268714,
"grad_norm": 10.398428916931152,
"learning_rate": 5.284708893154191e-06,
"loss": 0.7099,
"step": 3450
},
{
"epoch": 2.239283429302623,
"grad_norm": 5.051026821136475,
"learning_rate": 5.07144380464918e-06,
"loss": 0.7256,
"step": 3500
},
{
"epoch": 2.271273192578375,
"grad_norm": 5.900022506713867,
"learning_rate": 4.858178716144167e-06,
"loss": 0.6803,
"step": 3550
},
{
"epoch": 2.3032629558541267,
"grad_norm": 3.650153636932373,
"learning_rate": 4.644913627639156e-06,
"loss": 0.6936,
"step": 3600
},
{
"epoch": 2.3352527191298784,
"grad_norm": 3.192567825317383,
"learning_rate": 4.431648539134144e-06,
"loss": 0.7388,
"step": 3650
},
{
"epoch": 2.36724248240563,
"grad_norm": 2.792283058166504,
"learning_rate": 4.218383450629132e-06,
"loss": 0.7199,
"step": 3700
},
{
"epoch": 2.399232245681382,
"grad_norm": 4.723822593688965,
"learning_rate": 4.005118362124121e-06,
"loss": 0.7536,
"step": 3750
},
{
"epoch": 2.4312220089571337,
"grad_norm": 6.793713092803955,
"learning_rate": 3.791853273619109e-06,
"loss": 0.7043,
"step": 3800
},
{
"epoch": 2.4632117722328855,
"grad_norm": 4.855713367462158,
"learning_rate": 3.5785881851140968e-06,
"loss": 0.7587,
"step": 3850
},
{
"epoch": 2.495201535508637,
"grad_norm": 3.8809781074523926,
"learning_rate": 3.3653230966090854e-06,
"loss": 0.7007,
"step": 3900
},
{
"epoch": 2.527191298784389,
"grad_norm": 3.591956377029419,
"learning_rate": 3.1520580081040737e-06,
"loss": 0.6805,
"step": 3950
},
{
"epoch": 2.5591810620601407,
"grad_norm": 4.135153293609619,
"learning_rate": 2.9387929195990615e-06,
"loss": 0.7628,
"step": 4000
},
{
"epoch": 2.5911708253358925,
"grad_norm": 4.887539386749268,
"learning_rate": 2.72552783109405e-06,
"loss": 0.7203,
"step": 4050
},
{
"epoch": 2.6231605886116443,
"grad_norm": 5.000283241271973,
"learning_rate": 2.5122627425890384e-06,
"loss": 0.709,
"step": 4100
},
{
"epoch": 2.655150351887396,
"grad_norm": 5.802310943603516,
"learning_rate": 2.2989976540840266e-06,
"loss": 0.6938,
"step": 4150
},
{
"epoch": 2.6871401151631478,
"grad_norm": 2.909243583679199,
"learning_rate": 2.085732565579015e-06,
"loss": 0.7236,
"step": 4200
},
{
"epoch": 2.7191298784388995,
"grad_norm": 3.879182815551758,
"learning_rate": 1.872467477074003e-06,
"loss": 0.7327,
"step": 4250
},
{
"epoch": 2.7511196417146513,
"grad_norm": 5.3820295333862305,
"learning_rate": 1.6592023885689915e-06,
"loss": 0.6762,
"step": 4300
},
{
"epoch": 2.783109404990403,
"grad_norm": 4.583397388458252,
"learning_rate": 1.4459373000639796e-06,
"loss": 0.7111,
"step": 4350
},
{
"epoch": 2.815099168266155,
"grad_norm": 3.990649938583374,
"learning_rate": 1.2326722115589678e-06,
"loss": 0.7177,
"step": 4400
},
{
"epoch": 2.8470889315419066,
"grad_norm": 5.5702433586120605,
"learning_rate": 1.019407123053956e-06,
"loss": 0.7068,
"step": 4450
},
{
"epoch": 2.8790786948176583,
"grad_norm": 4.175040245056152,
"learning_rate": 8.061420345489445e-07,
"loss": 0.7092,
"step": 4500
},
{
"epoch": 2.91106845809341,
"grad_norm": 4.912069797515869,
"learning_rate": 5.928769460439326e-07,
"loss": 0.7022,
"step": 4550
},
{
"epoch": 2.943058221369162,
"grad_norm": 4.595305442810059,
"learning_rate": 3.796118575389209e-07,
"loss": 0.7228,
"step": 4600
},
{
"epoch": 2.9750479846449136,
"grad_norm": 3.835669755935669,
"learning_rate": 1.6634676903390917e-07,
"loss": 0.6946,
"step": 4650
},
{
"epoch": 3.0,
"eval_accuracy": 0.6829814459373,
"eval_f1": 0.5863851194804445,
"eval_loss": 0.7020198106765747,
"eval_runtime": 144.2948,
"eval_samples_per_second": 21.664,
"eval_steps_per_second": 1.358,
"step": 4689
}
],
"logging_steps": 50,
"max_steps": 4689,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1757912647073562.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}