alephbertgimmel_parashoot / trainer_state.json
imvladikon's picture
End of training
aaa18e7
{
"best_metric": 51.70745349157301,
"best_model_checkpoint": "./alephbertgimmel_parashoot/checkpoint-1844",
"epoch": 5.0,
"global_step": 2305,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_exact_match": 24.43438914027149,
"eval_f1": 45.67166307580162,
"eval_runtime": 6.5471,
"eval_samples_per_second": 36.81,
"eval_steps_per_second": 4.735,
"step": 461
},
{
"epoch": 1.08,
"learning_rate": 2.349240780911063e-05,
"loss": 3.387,
"step": 500
},
{
"epoch": 2.0,
"eval_exact_match": 29.41176470588235,
"eval_f1": 50.15387293791816,
"eval_runtime": 6.6624,
"eval_samples_per_second": 36.173,
"eval_steps_per_second": 4.653,
"step": 922
},
{
"epoch": 2.17,
"learning_rate": 1.698481561822126e-05,
"loss": 2.1624,
"step": 1000
},
{
"epoch": 3.0,
"eval_exact_match": 28.959276018099548,
"eval_f1": 50.438687898467485,
"eval_runtime": 6.6135,
"eval_samples_per_second": 36.44,
"eval_steps_per_second": 4.687,
"step": 1383
},
{
"epoch": 3.25,
"learning_rate": 1.0477223427331888e-05,
"loss": 1.5594,
"step": 1500
},
{
"epoch": 4.0,
"eval_exact_match": 29.41176470588235,
"eval_f1": 51.70745349157301,
"eval_runtime": 6.6431,
"eval_samples_per_second": 36.278,
"eval_steps_per_second": 4.666,
"step": 1844
},
{
"epoch": 4.34,
"learning_rate": 3.9696312364425166e-06,
"loss": 1.1811,
"step": 2000
},
{
"epoch": 5.0,
"eval_exact_match": 28.054298642533936,
"eval_f1": 51.328669460394764,
"eval_runtime": 6.6203,
"eval_samples_per_second": 36.403,
"eval_steps_per_second": 4.683,
"step": 2305
},
{
"epoch": 5.0,
"step": 2305,
"total_flos": 1803927484316160.0,
"train_loss": 1.930801941050366,
"train_runtime": 1080.0867,
"train_samples_per_second": 8.522,
"train_steps_per_second": 2.134
}
],
"max_steps": 2305,
"num_train_epochs": 5,
"total_flos": 1803927484316160.0,
"trial_name": null,
"trial_params": null
}