roberta-base-description2genre / trainer_state.json
pszemraj's picture
End of training
1a4d61f
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 6.0,
"eval_steps": 500,
"global_step": 372,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.16,
"learning_rate": 2.6666666666666667e-05,
"loss": 0.6871,
"step": 10
},
{
"epoch": 0.32,
"learning_rate": 3.943977591036415e-05,
"loss": 0.506,
"step": 20
},
{
"epoch": 0.48,
"learning_rate": 3.8319327731092444e-05,
"loss": 0.386,
"step": 30
},
{
"epoch": 0.65,
"learning_rate": 3.719887955182073e-05,
"loss": 0.3467,
"step": 40
},
{
"epoch": 0.81,
"learning_rate": 3.6078431372549025e-05,
"loss": 0.3235,
"step": 50
},
{
"epoch": 0.97,
"learning_rate": 3.495798319327731e-05,
"loss": 0.3118,
"step": 60
},
{
"epoch": 1.0,
"eval_f1": 0.336166194523135,
"eval_loss": 0.2885044813156128,
"eval_runtime": 3.3771,
"eval_samples_per_second": 292.857,
"eval_steps_per_second": 4.738,
"step": 62
},
{
"epoch": 1.13,
"learning_rate": 3.383753501400561e-05,
"loss": 0.3046,
"step": 70
},
{
"epoch": 1.29,
"learning_rate": 3.2717086834733894e-05,
"loss": 0.2906,
"step": 80
},
{
"epoch": 1.45,
"learning_rate": 3.159663865546219e-05,
"loss": 0.2896,
"step": 90
},
{
"epoch": 1.61,
"learning_rate": 3.047619047619048e-05,
"loss": 0.2826,
"step": 100
},
{
"epoch": 1.77,
"learning_rate": 2.935574229691877e-05,
"loss": 0.2749,
"step": 110
},
{
"epoch": 1.94,
"learning_rate": 2.8235294117647063e-05,
"loss": 0.2676,
"step": 120
},
{
"epoch": 2.0,
"eval_f1": 0.4882154882154882,
"eval_loss": 0.25112977623939514,
"eval_runtime": 3.3888,
"eval_samples_per_second": 291.842,
"eval_steps_per_second": 4.721,
"step": 124
},
{
"epoch": 2.1,
"learning_rate": 2.7114845938375354e-05,
"loss": 0.2534,
"step": 130
},
{
"epoch": 2.26,
"learning_rate": 2.5994397759103644e-05,
"loss": 0.2491,
"step": 140
},
{
"epoch": 2.42,
"learning_rate": 2.4873949579831935e-05,
"loss": 0.2452,
"step": 150
},
{
"epoch": 2.58,
"learning_rate": 2.3753501400560226e-05,
"loss": 0.2373,
"step": 160
},
{
"epoch": 2.74,
"learning_rate": 2.2633053221288516e-05,
"loss": 0.2355,
"step": 170
},
{
"epoch": 2.9,
"learning_rate": 2.1512605042016807e-05,
"loss": 0.2325,
"step": 180
},
{
"epoch": 3.0,
"eval_f1": 0.6093467596178673,
"eval_loss": 0.22724518179893494,
"eval_runtime": 3.3851,
"eval_samples_per_second": 292.16,
"eval_steps_per_second": 4.727,
"step": 186
},
{
"epoch": 3.06,
"learning_rate": 2.0392156862745097e-05,
"loss": 0.231,
"step": 190
},
{
"epoch": 3.23,
"learning_rate": 1.927170868347339e-05,
"loss": 0.2195,
"step": 200
},
{
"epoch": 3.39,
"learning_rate": 1.8151260504201682e-05,
"loss": 0.2156,
"step": 210
},
{
"epoch": 3.55,
"learning_rate": 1.7030812324929973e-05,
"loss": 0.2184,
"step": 220
},
{
"epoch": 3.71,
"learning_rate": 1.5910364145658263e-05,
"loss": 0.2109,
"step": 230
},
{
"epoch": 3.87,
"learning_rate": 1.4789915966386557e-05,
"loss": 0.2127,
"step": 240
},
{
"epoch": 4.0,
"eval_f1": 0.6591346153846154,
"eval_loss": 0.21806256473064423,
"eval_runtime": 3.3784,
"eval_samples_per_second": 292.738,
"eval_steps_per_second": 4.736,
"step": 248
},
{
"epoch": 4.03,
"learning_rate": 1.3669467787114848e-05,
"loss": 0.2067,
"step": 250
},
{
"epoch": 4.19,
"learning_rate": 1.2549019607843138e-05,
"loss": 0.1984,
"step": 260
},
{
"epoch": 4.35,
"learning_rate": 1.1428571428571429e-05,
"loss": 0.1957,
"step": 270
},
{
"epoch": 4.52,
"learning_rate": 1.030812324929972e-05,
"loss": 0.1967,
"step": 280
},
{
"epoch": 4.68,
"learning_rate": 9.187675070028012e-06,
"loss": 0.1975,
"step": 290
},
{
"epoch": 4.84,
"learning_rate": 8.067226890756303e-06,
"loss": 0.194,
"step": 300
},
{
"epoch": 5.0,
"learning_rate": 6.946778711484594e-06,
"loss": 0.1978,
"step": 310
},
{
"epoch": 5.0,
"eval_f1": 0.6685687113647171,
"eval_loss": 0.2140355408191681,
"eval_runtime": 3.3785,
"eval_samples_per_second": 292.736,
"eval_steps_per_second": 4.736,
"step": 310
},
{
"epoch": 5.16,
"learning_rate": 5.826330532212886e-06,
"loss": 0.1877,
"step": 320
},
{
"epoch": 5.32,
"learning_rate": 4.705882352941177e-06,
"loss": 0.1877,
"step": 330
},
{
"epoch": 5.48,
"learning_rate": 3.585434173669468e-06,
"loss": 0.1803,
"step": 340
},
{
"epoch": 5.65,
"learning_rate": 2.4649859943977594e-06,
"loss": 0.1874,
"step": 350
},
{
"epoch": 5.81,
"learning_rate": 1.3445378151260504e-06,
"loss": 0.1911,
"step": 360
},
{
"epoch": 5.97,
"learning_rate": 2.2408963585434175e-07,
"loss": 0.1817,
"step": 370
},
{
"epoch": 6.0,
"eval_f1": 0.6717231571462432,
"eval_loss": 0.21304987370967865,
"eval_runtime": 3.3805,
"eval_samples_per_second": 292.556,
"eval_steps_per_second": 4.733,
"step": 372
},
{
"epoch": 6.0,
"step": 372,
"total_flos": 1.2495360147628032e+16,
"train_loss": 0.2572957956662742,
"train_runtime": 515.2049,
"train_samples_per_second": 92.165,
"train_steps_per_second": 0.722
}
],
"logging_steps": 10,
"max_steps": 372,
"num_train_epochs": 6,
"save_steps": 500,
"total_flos": 1.2495360147628032e+16,
"trial_name": null,
"trial_params": null
}