| { | |
| "best_metric": 0.9066910147666931, | |
| "best_model_checkpoint": "/home/iais_marenpielka/Bouthaina/res_nw_eg_1.5/checkpoint-21315", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 56840, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.173611044883728, | |
| "learning_rate": 4.766772598870057e-05, | |
| "loss": 1.2517, | |
| "step": 7105 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.27630483908384024, | |
| "eval_loss": 1.0203286409378052, | |
| "eval_rouge1": 0.45177772565881835, | |
| "eval_rouge2": 0.09405681309859196, | |
| "eval_rougeL": 0.4439172371201736, | |
| "eval_runtime": 399.4609, | |
| "eval_samples_per_second": 35.57, | |
| "eval_steps_per_second": 4.448, | |
| "step": 7105 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3228689432144165, | |
| "learning_rate": 4.515889830508475e-05, | |
| "loss": 0.8548, | |
| "step": 14210 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.3013703850585307, | |
| "eval_loss": 0.9441202282905579, | |
| "eval_rouge1": 0.5003553784801738, | |
| "eval_rouge2": 0.1420292743132085, | |
| "eval_rougeL": 0.49367336720870203, | |
| "eval_runtime": 384.7586, | |
| "eval_samples_per_second": 36.93, | |
| "eval_steps_per_second": 4.618, | |
| "step": 14210 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.1944307088851929, | |
| "learning_rate": 4.265007062146893e-05, | |
| "loss": 0.7533, | |
| "step": 21315 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 0.310650314692537, | |
| "eval_loss": 0.9066910147666931, | |
| "eval_rouge1": 0.5230238843217684, | |
| "eval_rouge2": 0.16671937404400886, | |
| "eval_rougeL": 0.5166126640861954, | |
| "eval_runtime": 1037.5655, | |
| "eval_samples_per_second": 13.695, | |
| "eval_steps_per_second": 1.713, | |
| "step": 21315 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.6159381866455078, | |
| "learning_rate": 4.014124293785311e-05, | |
| "loss": 0.6781, | |
| "step": 28420 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 0.32048113085357516, | |
| "eval_loss": 0.909532904624939, | |
| "eval_rouge1": 0.5602679605982237, | |
| "eval_rouge2": 0.28777006827963697, | |
| "eval_rougeL": 0.5559392353971754, | |
| "eval_runtime": 377.7695, | |
| "eval_samples_per_second": 37.613, | |
| "eval_steps_per_second": 4.704, | |
| "step": 28420 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.188673973083496, | |
| "learning_rate": 3.763241525423729e-05, | |
| "loss": 0.6134, | |
| "step": 35525 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 0.3250047891084671, | |
| "eval_loss": 0.9152291417121887, | |
| "eval_rouge1": 0.5456119185913646, | |
| "eval_rouge2": 0.19532259739828473, | |
| "eval_rougeL": 0.5403228217764222, | |
| "eval_runtime": 377.7325, | |
| "eval_samples_per_second": 37.617, | |
| "eval_steps_per_second": 4.704, | |
| "step": 35525 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.2344160079956055, | |
| "learning_rate": 3.5123587570621466e-05, | |
| "loss": 0.5542, | |
| "step": 42630 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 0.3266066839569369, | |
| "eval_loss": 0.9293861985206604, | |
| "eval_rouge1": 0.5719694809249145, | |
| "eval_rouge2": 0.3028610424917365, | |
| "eval_rougeL": 0.5675532772123946, | |
| "eval_runtime": 380.2832, | |
| "eval_samples_per_second": 37.364, | |
| "eval_steps_per_second": 4.673, | |
| "step": 42630 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.5062270164489746, | |
| "learning_rate": 3.261475988700565e-05, | |
| "loss": 0.4993, | |
| "step": 49735 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 0.3252361071861535, | |
| "eval_loss": 0.9462244510650635, | |
| "eval_rouge1": 0.5502763386552358, | |
| "eval_rouge2": 0.20139303396164254, | |
| "eval_rougeL": 0.5447611551880709, | |
| "eval_runtime": 378.2074, | |
| "eval_samples_per_second": 37.569, | |
| "eval_steps_per_second": 4.698, | |
| "step": 49735 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.5913320779800415, | |
| "learning_rate": 3.010593220338983e-05, | |
| "loss": 0.4482, | |
| "step": 56840 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 0.32453596360754455, | |
| "eval_loss": 0.9838415384292603, | |
| "eval_rouge1": 0.5485197319407531, | |
| "eval_rouge2": 0.20067000334446508, | |
| "eval_rougeL": 0.542963955198015, | |
| "eval_runtime": 377.6879, | |
| "eval_samples_per_second": 37.621, | |
| "eval_steps_per_second": 4.705, | |
| "step": 56840 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 56840, | |
| "total_flos": 5.152666773356544e+17, | |
| "train_loss": 0.7066081702751144, | |
| "train_runtime": 59748.056, | |
| "train_samples_per_second": 19.025, | |
| "train_steps_per_second": 2.378 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 142100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.152666773356544e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |