electra-small-mnli / trainer_state.json
howey's picture
Update from ubuntu
9e8a6bd
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"global_step": 36816,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 1.9728378965667103e-05,
"loss": 0.9635,
"step": 500
},
{
"epoch": 0.08,
"learning_rate": 1.9456757931334205e-05,
"loss": 0.7888,
"step": 1000
},
{
"epoch": 0.12,
"learning_rate": 1.9185136897001307e-05,
"loss": 0.7178,
"step": 1500
},
{
"epoch": 0.16,
"learning_rate": 1.8913515862668405e-05,
"loss": 0.6949,
"step": 2000
},
{
"epoch": 0.2,
"learning_rate": 1.8641894828335507e-05,
"loss": 0.6726,
"step": 2500
},
{
"epoch": 0.24,
"learning_rate": 1.837027379400261e-05,
"loss": 0.654,
"step": 3000
},
{
"epoch": 0.29,
"learning_rate": 1.809865275966971e-05,
"loss": 0.6358,
"step": 3500
},
{
"epoch": 0.33,
"learning_rate": 1.7827031725336812e-05,
"loss": 0.636,
"step": 4000
},
{
"epoch": 0.37,
"learning_rate": 1.7555410691003914e-05,
"loss": 0.6217,
"step": 4500
},
{
"epoch": 0.41,
"learning_rate": 1.7283789656671015e-05,
"loss": 0.612,
"step": 5000
},
{
"epoch": 0.45,
"learning_rate": 1.7012168622338114e-05,
"loss": 0.5993,
"step": 5500
},
{
"epoch": 0.49,
"learning_rate": 1.6740547588005215e-05,
"loss": 0.5968,
"step": 6000
},
{
"epoch": 0.53,
"learning_rate": 1.6468926553672317e-05,
"loss": 0.599,
"step": 6500
},
{
"epoch": 0.57,
"learning_rate": 1.619730551933942e-05,
"loss": 0.5795,
"step": 7000
},
{
"epoch": 0.61,
"learning_rate": 1.592568448500652e-05,
"loss": 0.5939,
"step": 7500
},
{
"epoch": 0.65,
"learning_rate": 1.5654063450673622e-05,
"loss": 0.5669,
"step": 8000
},
{
"epoch": 0.69,
"learning_rate": 1.5382442416340724e-05,
"loss": 0.5799,
"step": 8500
},
{
"epoch": 0.73,
"learning_rate": 1.5110821382007822e-05,
"loss": 0.5666,
"step": 9000
},
{
"epoch": 0.77,
"learning_rate": 1.4839200347674924e-05,
"loss": 0.5683,
"step": 9500
},
{
"epoch": 0.81,
"learning_rate": 1.4567579313342026e-05,
"loss": 0.5629,
"step": 10000
},
{
"epoch": 0.86,
"learning_rate": 1.4295958279009128e-05,
"loss": 0.5619,
"step": 10500
},
{
"epoch": 0.9,
"learning_rate": 1.402433724467623e-05,
"loss": 0.5454,
"step": 11000
},
{
"epoch": 0.94,
"learning_rate": 1.3752716210343331e-05,
"loss": 0.556,
"step": 11500
},
{
"epoch": 0.98,
"learning_rate": 1.3481095176010431e-05,
"loss": 0.5548,
"step": 12000
},
{
"epoch": 1.02,
"learning_rate": 1.3209474141677531e-05,
"loss": 0.5355,
"step": 12500
},
{
"epoch": 1.06,
"learning_rate": 1.2937853107344633e-05,
"loss": 0.5076,
"step": 13000
},
{
"epoch": 1.1,
"learning_rate": 1.2666232073011735e-05,
"loss": 0.5003,
"step": 13500
},
{
"epoch": 1.14,
"learning_rate": 1.2394611038678836e-05,
"loss": 0.501,
"step": 14000
},
{
"epoch": 1.18,
"learning_rate": 1.2122990004345938e-05,
"loss": 0.5115,
"step": 14500
},
{
"epoch": 1.22,
"learning_rate": 1.185136897001304e-05,
"loss": 0.5065,
"step": 15000
},
{
"epoch": 1.26,
"learning_rate": 1.157974793568014e-05,
"loss": 0.5079,
"step": 15500
},
{
"epoch": 1.3,
"learning_rate": 1.130812690134724e-05,
"loss": 0.5127,
"step": 16000
},
{
"epoch": 1.34,
"learning_rate": 1.1036505867014341e-05,
"loss": 0.4956,
"step": 16500
},
{
"epoch": 1.39,
"learning_rate": 1.0764884832681443e-05,
"loss": 0.4939,
"step": 17000
},
{
"epoch": 1.43,
"learning_rate": 1.0493263798348545e-05,
"loss": 0.496,
"step": 17500
},
{
"epoch": 1.47,
"learning_rate": 1.0221642764015647e-05,
"loss": 0.4861,
"step": 18000
},
{
"epoch": 1.51,
"learning_rate": 9.950021729682747e-06,
"loss": 0.4964,
"step": 18500
},
{
"epoch": 1.55,
"learning_rate": 9.678400695349848e-06,
"loss": 0.5023,
"step": 19000
},
{
"epoch": 1.59,
"learning_rate": 9.40677966101695e-06,
"loss": 0.5021,
"step": 19500
},
{
"epoch": 1.63,
"learning_rate": 9.13515862668405e-06,
"loss": 0.4965,
"step": 20000
},
{
"epoch": 1.67,
"learning_rate": 8.863537592351152e-06,
"loss": 0.4902,
"step": 20500
},
{
"epoch": 1.71,
"learning_rate": 8.591916558018254e-06,
"loss": 0.4969,
"step": 21000
},
{
"epoch": 1.75,
"learning_rate": 8.320295523685355e-06,
"loss": 0.4814,
"step": 21500
},
{
"epoch": 1.79,
"learning_rate": 8.048674489352455e-06,
"loss": 0.4934,
"step": 22000
},
{
"epoch": 1.83,
"learning_rate": 7.777053455019557e-06,
"loss": 0.5016,
"step": 22500
},
{
"epoch": 1.87,
"learning_rate": 7.505432420686659e-06,
"loss": 0.5007,
"step": 23000
},
{
"epoch": 1.91,
"learning_rate": 7.23381138635376e-06,
"loss": 0.4894,
"step": 23500
},
{
"epoch": 1.96,
"learning_rate": 6.962190352020861e-06,
"loss": 0.4905,
"step": 24000
},
{
"epoch": 2.0,
"learning_rate": 6.690569317687963e-06,
"loss": 0.4804,
"step": 24500
},
{
"epoch": 2.04,
"learning_rate": 6.418948283355063e-06,
"loss": 0.4529,
"step": 25000
},
{
"epoch": 2.08,
"learning_rate": 6.147327249022165e-06,
"loss": 0.4558,
"step": 25500
},
{
"epoch": 2.12,
"learning_rate": 5.8757062146892665e-06,
"loss": 0.4558,
"step": 26000
},
{
"epoch": 2.16,
"learning_rate": 5.6040851803563665e-06,
"loss": 0.4478,
"step": 26500
},
{
"epoch": 2.2,
"learning_rate": 5.332464146023468e-06,
"loss": 0.4515,
"step": 27000
},
{
"epoch": 2.24,
"learning_rate": 5.06084311169057e-06,
"loss": 0.4492,
"step": 27500
},
{
"epoch": 2.28,
"learning_rate": 4.789222077357671e-06,
"loss": 0.4502,
"step": 28000
},
{
"epoch": 2.32,
"learning_rate": 4.5176010430247726e-06,
"loss": 0.4564,
"step": 28500
},
{
"epoch": 2.36,
"learning_rate": 4.2459800086918734e-06,
"loss": 0.4501,
"step": 29000
},
{
"epoch": 2.4,
"learning_rate": 3.974358974358974e-06,
"loss": 0.4576,
"step": 29500
},
{
"epoch": 2.44,
"learning_rate": 3.702737940026076e-06,
"loss": 0.4586,
"step": 30000
},
{
"epoch": 2.49,
"learning_rate": 3.431116905693177e-06,
"loss": 0.4425,
"step": 30500
},
{
"epoch": 2.53,
"learning_rate": 3.1594958713602786e-06,
"loss": 0.4559,
"step": 31000
},
{
"epoch": 2.57,
"learning_rate": 2.8878748370273795e-06,
"loss": 0.4429,
"step": 31500
},
{
"epoch": 2.61,
"learning_rate": 2.6162538026944812e-06,
"loss": 0.4494,
"step": 32000
},
{
"epoch": 2.65,
"learning_rate": 2.344632768361582e-06,
"loss": 0.4465,
"step": 32500
},
{
"epoch": 2.69,
"learning_rate": 2.0730117340286834e-06,
"loss": 0.45,
"step": 33000
},
{
"epoch": 2.73,
"learning_rate": 1.8013906996957845e-06,
"loss": 0.4547,
"step": 33500
},
{
"epoch": 2.77,
"learning_rate": 1.5297696653628858e-06,
"loss": 0.4386,
"step": 34000
},
{
"epoch": 2.81,
"learning_rate": 1.258148631029987e-06,
"loss": 0.4519,
"step": 34500
},
{
"epoch": 2.85,
"learning_rate": 9.865275966970884e-07,
"loss": 0.4395,
"step": 35000
},
{
"epoch": 2.89,
"learning_rate": 7.149065623641896e-07,
"loss": 0.4439,
"step": 35500
},
{
"epoch": 2.93,
"learning_rate": 4.432855280312908e-07,
"loss": 0.4513,
"step": 36000
},
{
"epoch": 2.97,
"learning_rate": 1.7166449369839202e-07,
"loss": 0.4531,
"step": 36500
},
{
"epoch": 3.0,
"step": 36816,
"total_flos": 1.2259454125459968e+16,
"train_runtime": 3279.0742,
"train_samples_per_second": 11.228
}
],
"max_steps": 36816,
"num_train_epochs": 3,
"total_flos": 1.2259454125459968e+16,
"trial_name": null,
"trial_params": null
}