afriqa_afroxlmr_squad_v2 / trainer_state.json
ToluClassics's picture
End of training
3250715
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"global_step": 10420,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.48,
"learning_rate": 2.7120921305182343e-05,
"loss": 1.6629,
"step": 1000
},
{
"epoch": 0.48,
"eval_HasAns_exact": 62.432523616734144,
"eval_HasAns_f1": 68.73496905793459,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 73.23801513877208,
"eval_NoAns_f1": 73.23801513877208,
"eval_NoAns_total": 5945,
"eval_best_exact": 67.84300513770741,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 70.98971587428946,
"eval_best_f1_thresh": 0.0,
"eval_exact": 67.84300513770741,
"eval_f1": 70.98971587428946,
"eval_total": 11873,
"step": 1000
},
{
"epoch": 0.96,
"learning_rate": 2.4241842610364685e-05,
"loss": 1.1484,
"step": 2000
},
{
"epoch": 0.96,
"eval_HasAns_exact": 69.16329284750337,
"eval_HasAns_f1": 75.09571117953085,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 76.50126156433979,
"eval_NoAns_f1": 76.50126156433979,
"eval_NoAns_total": 5945,
"eval_best_exact": 72.83753053145793,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 75.79949261957877,
"eval_best_f1_thresh": 0.0,
"eval_exact": 72.83753053145793,
"eval_f1": 75.79949261957887,
"eval_total": 11873,
"step": 2000
},
{
"epoch": 1.44,
"learning_rate": 2.1362763915547027e-05,
"loss": 0.9368,
"step": 3000
},
{
"epoch": 1.44,
"eval_HasAns_exact": 70.1417004048583,
"eval_HasAns_f1": 76.29593503528035,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 77.96467619848612,
"eval_NoAns_f1": 77.96467619848612,
"eval_NoAns_total": 5945,
"eval_best_exact": 74.05878884864819,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 77.13150028544948,
"eval_best_f1_thresh": 0.0,
"eval_exact": 74.05878884864819,
"eval_f1": 77.13150028544946,
"eval_total": 11873,
"step": 3000
},
{
"epoch": 1.92,
"learning_rate": 1.8483685220729366e-05,
"loss": 0.8956,
"step": 4000
},
{
"epoch": 1.92,
"eval_HasAns_exact": 71.37314439946019,
"eval_HasAns_f1": 77.92118181028806,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 77.47687132043734,
"eval_NoAns_f1": 77.47687132043734,
"eval_NoAns_total": 5945,
"eval_best_exact": 74.42937757938179,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 77.6987084790184,
"eval_best_f1_thresh": 0.0,
"eval_exact": 74.42937757938179,
"eval_f1": 77.69870847901858,
"eval_total": 11873,
"step": 4000
},
{
"epoch": 2.4,
"learning_rate": 1.5604606525911708e-05,
"loss": 0.7561,
"step": 5000
},
{
"epoch": 2.4,
"eval_HasAns_exact": 72.73954116059379,
"eval_HasAns_f1": 79.1460959770044,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 76.63582842724979,
"eval_NoAns_f1": 76.63582842724979,
"eval_NoAns_total": 5945,
"eval_best_exact": 74.69047418512592,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 77.88916507636502,
"eval_best_f1_thresh": 0.0,
"eval_exact": 74.69047418512592,
"eval_f1": 77.889165076365,
"eval_total": 11873,
"step": 5000
},
{
"epoch": 2.88,
"learning_rate": 1.272552783109405e-05,
"loss": 0.7174,
"step": 6000
},
{
"epoch": 2.88,
"eval_HasAns_exact": 74.2914979757085,
"eval_HasAns_f1": 81.01202096109864,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 74.48275862068965,
"eval_NoAns_f1": 74.48275862068965,
"eval_NoAns_total": 5945,
"eval_best_exact": 74.38726522361661,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 77.74271542637852,
"eval_best_f1_thresh": 0.0,
"eval_exact": 74.38726522361661,
"eval_f1": 77.74271542637855,
"eval_total": 11873,
"step": 6000
},
{
"epoch": 3.36,
"learning_rate": 9.846449136276392e-06,
"loss": 0.6354,
"step": 7000
},
{
"epoch": 3.36,
"eval_HasAns_exact": 71.9298245614035,
"eval_HasAns_f1": 78.42645795015271,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 78.4861227922624,
"eval_NoAns_f1": 78.4861227922624,
"eval_NoAns_total": 5945,
"eval_best_exact": 75.21266739661417,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 78.45633308586743,
"eval_best_f1_thresh": 0.0,
"eval_exact": 75.21266739661417,
"eval_f1": 78.45633308586743,
"eval_total": 11873,
"step": 7000
},
{
"epoch": 3.84,
"learning_rate": 6.967370441458734e-06,
"loss": 0.603,
"step": 8000
},
{
"epoch": 3.84,
"eval_HasAns_exact": 69.68623481781377,
"eval_HasAns_f1": 75.68905120730992,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 82.45584524810765,
"eval_NoAns_f1": 82.45584524810765,
"eval_NoAns_total": 5945,
"eval_best_exact": 76.0801819253769,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 79.07729264355531,
"eval_best_f1_thresh": 0.0,
"eval_exact": 76.0801819253769,
"eval_f1": 79.07729264355528,
"eval_total": 11873,
"step": 8000
},
{
"epoch": 4.32,
"learning_rate": 4.088291746641075e-06,
"loss": 0.5511,
"step": 9000
},
{
"epoch": 4.32,
"eval_HasAns_exact": 72.08164642375169,
"eval_HasAns_f1": 78.78904517769465,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 79.1757779646762,
"eval_NoAns_f1": 79.1757779646762,
"eval_NoAns_total": 5945,
"eval_best_exact": 75.63379095426598,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 78.98268843707342,
"eval_best_f1_thresh": 0.0,
"eval_exact": 75.63379095426598,
"eval_f1": 78.98268843707345,
"eval_total": 11873,
"step": 9000
},
{
"epoch": 4.8,
"learning_rate": 1.2092130518234166e-06,
"loss": 0.5237,
"step": 10000
},
{
"epoch": 4.8,
"eval_HasAns_exact": 71.9298245614035,
"eval_HasAns_f1": 78.5833413462209,
"eval_HasAns_total": 5928,
"eval_NoAns_exact": 79.42809083263246,
"eval_NoAns_f1": 79.42809083263246,
"eval_NoAns_total": 5945,
"eval_best_exact": 75.6843257811842,
"eval_best_exact_thresh": 0.0,
"eval_best_f1": 79.00632085407202,
"eval_best_f1_thresh": 0.0,
"eval_exact": 75.6843257811842,
"eval_f1": 79.00632085407196,
"eval_total": 11873,
"step": 10000
},
{
"epoch": 5.0,
"step": 10420,
"total_flos": 1.3063237394164992e+17,
"train_loss": 0.8300563907440244,
"train_runtime": 12062.7684,
"train_samples_per_second": 55.26,
"train_steps_per_second": 0.864
}
],
"max_steps": 10420,
"num_train_epochs": 5,
"total_flos": 1.3063237394164992e+17,
"trial_name": null,
"trial_params": null
}