de-slothbug's picture
End of full fine-tuning with BART
c76ff12 verified
Raw
History Blame Contribute Delete
2.26 kB
{
"best_global_step": 500,
"best_metric": 0.23719677329063416,
"best_model_checkpoint": "bart-pristine-finetune/checkpoint-500",
"epoch": 0.1738828029907842,
"eval_steps": 500,
"global_step": 500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03477656059815684,
"grad_norm": 18.65494728088379,
"learning_rate": 9.900000000000002e-06,
"loss": 6.5893,
"step": 100
},
{
"epoch": 0.06955312119631368,
"grad_norm": 15.10452938079834,
"learning_rate": 1.9900000000000003e-05,
"loss": 1.9811,
"step": 200
},
{
"epoch": 0.10432968179447052,
"grad_norm": 1.449442744255066,
"learning_rate": 2.9900000000000002e-05,
"loss": 0.3884,
"step": 300
},
{
"epoch": 0.13910624239262737,
"grad_norm": 0.8391104340553284,
"learning_rate": 3.99e-05,
"loss": 0.1479,
"step": 400
},
{
"epoch": 0.1738828029907842,
"grad_norm": 0.8194042444229126,
"learning_rate": 4.99e-05,
"loss": 0.1068,
"step": 500
},
{
"epoch": 0.1738828029907842,
"eval_gen_len": 20.974,
"eval_loss": 0.23719677329063416,
"eval_rouge1": 0.4651,
"eval_rouge2": 0.3753,
"eval_rougeL": 0.4611,
"eval_rougeLsum": 0.4615,
"eval_runtime": 131.8304,
"eval_samples_per_second": 36.737,
"eval_steps_per_second": 2.298,
"step": 500
}
],
"logging_steps": 100,
"max_steps": 8628,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 3,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 609736458240000.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}