trainer_state.json

{
 "best_metric": 34.8229,
 "best_model_checkpoint": "bin/liputan6-pt-pl50/checkpoint-315",
 "epoch": 5.0,
 "eval_steps": 500,
 "global_step": 315,
 "is_hyper_param_search": false,
 "is_local_process_zero": true,
 "is_world_process_zero": true,
 "log_history": [
 {
 "epoch": 1.0,
 "grad_norm": 0.9081947207450867,
 "learning_rate": 0.0008,
 "loss": 4.2782,
 "step": 63
 },
 {
 "epoch": 1.0,
 "eval_gen_len": 38.037,
 "eval_loss": 3.2600271701812744,
 "eval_rouge1": 25.0139,
 "eval_rouge2": 13.1669,
 "eval_rougeL": 22.4852,
 "eval_rougeLsum": 23.5026,
 "eval_runtime": 1079.1281,
 "eval_samples_per_second": 0.927,
 "eval_steps_per_second": 0.03,
 "step": 63
 },
 {
 "epoch": 2.0,
 "grad_norm": 0.9287102818489075,
 "learning_rate": 0.0006,
 "loss": 3.3831,
 "step": 126
 },
 {
 "epoch": 2.0,
 "eval_gen_len": 51.621,
 "eval_loss": 3.011786460876465,
 "eval_rouge1": 28.0005,
 "eval_rouge2": 15.5199,
 "eval_rougeL": 25.1006,
 "eval_rougeLsum": 26.3175,
 "eval_runtime": 1436.7107,
 "eval_samples_per_second": 0.696,
 "eval_steps_per_second": 0.022,
 "step": 126
 },
 {
 "epoch": 3.0,
 "grad_norm": 0.8845603466033936,
 "learning_rate": 0.0004,
 "loss": 3.0732,
 "step": 189
 },
 {
 "epoch": 3.0,
 "eval_gen_len": 51.938,
 "eval_loss": 2.822599172592163,
 "eval_rouge1": 31.6641,
 "eval_rouge2": 18.1569,
 "eval_rougeL": 27.8004,
 "eval_rougeLsum": 29.8463,
 "eval_runtime": 1387.7431,
 "eval_samples_per_second": 0.721,
 "eval_steps_per_second": 0.023,
 "step": 189
 },
 {
 "epoch": 4.0,
 "grad_norm": 0.9532793164253235,
 "learning_rate": 0.0002,
 "loss": 2.83,
 "step": 252
 },
 {
 "epoch": 4.0,
 "eval_gen_len": 51.327,
 "eval_loss": 2.718118667602539,
 "eval_rouge1": 34.3328,
 "eval_rouge2": 21.5065,
 "eval_rougeL": 30.323,
 "eval_rougeLsum": 32.3623,
 "eval_runtime": 1087.9011,
 "eval_samples_per_second": 0.919,
 "eval_steps_per_second": 0.029,
 "step": 252
 },
 {
 "epoch": 5.0,
 "grad_norm": 0.9524036049842834,
 "learning_rate": 0.0,
 "loss": 2.6441,
 "step": 315
 },
 {
 "epoch": 5.0,
 "eval_gen_len": 52.623,
 "eval_loss": 2.672222137451172,
 "eval_rouge1": 34.8229,
 "eval_rouge2": 22.044,
 "eval_rougeL": 30.8324,
 "eval_rougeLsum": 33.0138,
 "eval_runtime": 1171.6265,
 "eval_samples_per_second": 0.854,
 "eval_steps_per_second": 0.027,
 "step": 315
 },
 {
 "epoch": 5.0,
 "step": 315,
 "total_flos": 3877644533760000.0,
 "train_loss": 3.241740926106771,
 "train_runtime": 6571.5037,
 "train_samples_per_second": 0.761,
 "train_steps_per_second": 0.048
 }
 ],
 "logging_steps": 500,
 "max_steps": 315,
 "num_input_tokens_seen": 0,
 "num_train_epochs": 5,
 "save_steps": 500,
 "total_flos": 3877644533760000.0,
 "train_batch_size": 16,
 "trial_name": null,
 "trial_params": null
}