Aditya3107's picture
End of training
b11a2b9
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 17.999072356215212,
"global_step": 6462,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.56,
"eval_loss": 2.8832294940948486,
"eval_runtime": 56.6037,
"eval_samples_per_second": 9.045,
"eval_steps_per_second": 1.131,
"eval_wer": 1.0,
"step": 200
},
{
"epoch": 1.11,
"eval_loss": 1.1704888343811035,
"eval_runtime": 56.4819,
"eval_samples_per_second": 9.065,
"eval_steps_per_second": 1.133,
"eval_wer": 0.7788489628980426,
"step": 400
},
{
"epoch": 1.39,
"learning_rate": 0.00029939999999999996,
"loss": 3.3987,
"step": 500
},
{
"epoch": 1.67,
"eval_loss": 0.7739295959472656,
"eval_runtime": 57.0638,
"eval_samples_per_second": 8.972,
"eval_steps_per_second": 1.122,
"eval_wer": 0.5895413380075957,
"step": 600
},
{
"epoch": 2.23,
"eval_loss": 0.6045235395431519,
"eval_runtime": 56.2609,
"eval_samples_per_second": 9.1,
"eval_steps_per_second": 1.138,
"eval_wer": 0.4902132632193982,
"step": 800
},
{
"epoch": 2.78,
"learning_rate": 0.0002748909761824891,
"loss": 0.8313,
"step": 1000
},
{
"epoch": 2.78,
"eval_loss": 0.5234911441802979,
"eval_runtime": 56.3102,
"eval_samples_per_second": 9.092,
"eval_steps_per_second": 1.137,
"eval_wer": 0.43938066023955596,
"step": 1000
},
{
"epoch": 3.34,
"eval_loss": 0.48238545656204224,
"eval_runtime": 56.3832,
"eval_samples_per_second": 9.081,
"eval_steps_per_second": 1.135,
"eval_wer": 0.4002337131171487,
"step": 1200
},
{
"epoch": 3.9,
"eval_loss": 0.4378102719783783,
"eval_runtime": 56.2821,
"eval_samples_per_second": 9.097,
"eval_steps_per_second": 1.137,
"eval_wer": 0.37540169442009935,
"step": 1400
},
{
"epoch": 4.18,
"learning_rate": 0.00024973163367997313,
"loss": 0.5342,
"step": 1500
},
{
"epoch": 4.46,
"eval_loss": 0.44333964586257935,
"eval_runtime": 56.3381,
"eval_samples_per_second": 9.088,
"eval_steps_per_second": 1.136,
"eval_wer": 0.36342389716622847,
"step": 1600
},
{
"epoch": 5.01,
"eval_loss": 0.41031020879745483,
"eval_runtime": 56.4477,
"eval_samples_per_second": 9.07,
"eval_steps_per_second": 1.134,
"eval_wer": 0.34852468594799885,
"step": 1800
},
{
"epoch": 5.57,
"learning_rate": 0.00022457229117745723,
"loss": 0.3792,
"step": 2000
},
{
"epoch": 5.57,
"eval_loss": 0.3816453218460083,
"eval_runtime": 56.4551,
"eval_samples_per_second": 9.069,
"eval_steps_per_second": 1.134,
"eval_wer": 0.33099620216184633,
"step": 2000
},
{
"epoch": 6.13,
"eval_loss": 0.3952919840812683,
"eval_runtime": 56.4457,
"eval_samples_per_second": 9.071,
"eval_steps_per_second": 1.134,
"eval_wer": 0.32252410166520595,
"step": 2200
},
{
"epoch": 6.68,
"eval_loss": 0.39945441484451294,
"eval_runtime": 56.6634,
"eval_samples_per_second": 9.036,
"eval_steps_per_second": 1.129,
"eval_wer": 0.31317557697925796,
"step": 2400
},
{
"epoch": 6.96,
"learning_rate": 0.00019941294867494128,
"loss": 0.2924,
"step": 2500
},
{
"epoch": 7.24,
"eval_loss": 0.3906857967376709,
"eval_runtime": 56.3831,
"eval_samples_per_second": 9.081,
"eval_steps_per_second": 1.135,
"eval_wer": 0.2930178206251826,
"step": 2600
},
{
"epoch": 7.8,
"eval_loss": 0.35171157121658325,
"eval_runtime": 56.7082,
"eval_samples_per_second": 9.029,
"eval_steps_per_second": 1.129,
"eval_wer": 0.2740286298568507,
"step": 2800
},
{
"epoch": 8.36,
"learning_rate": 0.00017425360617242535,
"loss": 0.2217,
"step": 3000
},
{
"epoch": 8.36,
"eval_loss": 0.33607447147369385,
"eval_runtime": 56.5616,
"eval_samples_per_second": 9.052,
"eval_steps_per_second": 1.132,
"eval_wer": 0.2591294186386211,
"step": 3000
},
{
"epoch": 8.91,
"eval_loss": 0.3340049088001251,
"eval_runtime": 56.5741,
"eval_samples_per_second": 9.05,
"eval_steps_per_second": 1.131,
"eval_wer": 0.2451066316096991,
"step": 3200
},
{
"epoch": 9.47,
"eval_loss": 0.3125685453414917,
"eval_runtime": 56.4726,
"eval_samples_per_second": 9.066,
"eval_steps_per_second": 1.133,
"eval_wer": 0.2448144902132632,
"step": 3400
},
{
"epoch": 9.75,
"learning_rate": 0.00014909426366990943,
"loss": 0.1714,
"step": 3500
},
{
"epoch": 10.03,
"eval_loss": 0.34412676095962524,
"eval_runtime": 56.5983,
"eval_samples_per_second": 9.046,
"eval_steps_per_second": 1.131,
"eval_wer": 0.2556237218813906,
"step": 3600
},
{
"epoch": 10.58,
"eval_loss": 0.3404456079006195,
"eval_runtime": 56.5608,
"eval_samples_per_second": 9.052,
"eval_steps_per_second": 1.132,
"eval_wer": 0.2521180251241601,
"step": 3800
},
{
"epoch": 11.14,
"learning_rate": 0.0001239349211673935,
"loss": 0.1395,
"step": 4000
},
{
"epoch": 11.14,
"eval_loss": 0.3728441894054413,
"eval_runtime": 56.7823,
"eval_samples_per_second": 9.017,
"eval_steps_per_second": 1.127,
"eval_wer": 0.25182588372772424,
"step": 4000
},
{
"epoch": 11.7,
"eval_loss": 0.3828706741333008,
"eval_runtime": 56.4473,
"eval_samples_per_second": 9.07,
"eval_steps_per_second": 1.134,
"eval_wer": 0.23955594507741748,
"step": 4200
},
{
"epoch": 12.26,
"eval_loss": 0.3465881943702698,
"eval_runtime": 56.3386,
"eval_samples_per_second": 9.088,
"eval_steps_per_second": 1.136,
"eval_wer": 0.23605024832018698,
"step": 4400
},
{
"epoch": 12.53,
"learning_rate": 9.877557866487755e-05,
"loss": 0.1069,
"step": 4500
},
{
"epoch": 12.81,
"eval_loss": 0.3187991976737976,
"eval_runtime": 56.3638,
"eval_samples_per_second": 9.084,
"eval_steps_per_second": 1.135,
"eval_wer": 0.2240724510663161,
"step": 4600
},
{
"epoch": 13.37,
"eval_loss": 0.3395535349845886,
"eval_runtime": 56.3257,
"eval_samples_per_second": 9.09,
"eval_steps_per_second": 1.136,
"eval_wer": 0.21969033011977798,
"step": 4800
},
{
"epoch": 13.93,
"learning_rate": 7.361623616236162e-05,
"loss": 0.0845,
"step": 5000
},
{
"epoch": 13.93,
"eval_loss": 0.3364916741847992,
"eval_runtime": 56.321,
"eval_samples_per_second": 9.091,
"eval_steps_per_second": 1.136,
"eval_wer": 0.2205667543090856,
"step": 5000
},
{
"epoch": 14.48,
"eval_loss": 0.3458584249019623,
"eval_runtime": 56.3485,
"eval_samples_per_second": 9.086,
"eval_steps_per_second": 1.136,
"eval_wer": 0.22085889570552147,
"step": 5200
},
{
"epoch": 15.04,
"eval_loss": 0.3429270386695862,
"eval_runtime": 57.9716,
"eval_samples_per_second": 8.832,
"eval_steps_per_second": 1.104,
"eval_wer": 0.2193981887233421,
"step": 5400
},
{
"epoch": 15.32,
"learning_rate": 4.845689365984569e-05,
"loss": 0.0675,
"step": 5500
},
{
"epoch": 15.6,
"eval_loss": 0.3433798849582672,
"eval_runtime": 56.3488,
"eval_samples_per_second": 9.086,
"eval_steps_per_second": 1.136,
"eval_wer": 0.2182296231375986,
"step": 5600
},
{
"epoch": 16.16,
"eval_loss": 0.3434172570705414,
"eval_runtime": 56.3181,
"eval_samples_per_second": 9.091,
"eval_steps_per_second": 1.136,
"eval_wer": 0.20829681565877886,
"step": 5800
},
{
"epoch": 16.71,
"learning_rate": 2.329755115732975e-05,
"loss": 0.0561,
"step": 6000
},
{
"epoch": 16.71,
"eval_loss": 0.33747875690460205,
"eval_runtime": 56.3763,
"eval_samples_per_second": 9.082,
"eval_steps_per_second": 1.135,
"eval_wer": 0.20362255331580484,
"step": 6000
},
{
"epoch": 17.27,
"eval_loss": 0.3445747494697571,
"eval_runtime": 56.6382,
"eval_samples_per_second": 9.04,
"eval_steps_per_second": 1.13,
"eval_wer": 0.19865614957639496,
"step": 6200
},
{
"epoch": 17.83,
"eval_loss": 0.33622780442237854,
"eval_runtime": 56.3321,
"eval_samples_per_second": 9.089,
"eval_steps_per_second": 1.136,
"eval_wer": 0.19777972538708735,
"step": 6400
},
{
"epoch": 18.0,
"step": 6462,
"total_flos": 3.4883692787272507e+19,
"train_loss": 0.4896983008235647,
"train_runtime": 62213.9785,
"train_samples_per_second": 2.494,
"train_steps_per_second": 0.104
}
],
"max_steps": 6462,
"num_train_epochs": 18,
"total_flos": 3.4883692787272507e+19,
"trial_name": null,
"trial_params": null
}