| { | |
| "best_metric": 0.20781023800373077, | |
| "best_model_checkpoint": "/home/iais_marenpielka/Bouthaina/res_nw_irq_aragpt2-base/checkpoint-12684", | |
| "epoch": 17.0, | |
| "eval_steps": 500, | |
| "global_step": 17969, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.21245746314525604, | |
| "learning_rate": 4.865067829457365e-05, | |
| "loss": 0.9334, | |
| "step": 1057 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.003236761532472025, | |
| "eval_loss": 0.24612107872962952, | |
| "eval_rouge1": 0.1592859430436859, | |
| "eval_rouge2": 0.01850245530694966, | |
| "eval_rougeL": 0.15327183350590745, | |
| "eval_runtime": 47.2784, | |
| "eval_samples_per_second": 44.693, | |
| "eval_steps_per_second": 5.605, | |
| "step": 1057 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.1900666505098343, | |
| "learning_rate": 4.609011627906977e-05, | |
| "loss": 0.0868, | |
| "step": 2114 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.01485892408066134, | |
| "eval_loss": 0.2332119345664978, | |
| "eval_rouge1": 0.24547054015158856, | |
| "eval_rouge2": 0.05101888171169662, | |
| "eval_rougeL": 0.23941488622001442, | |
| "eval_runtime": 106.4482, | |
| "eval_samples_per_second": 19.85, | |
| "eval_steps_per_second": 2.489, | |
| "step": 2114 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.17272518575191498, | |
| "learning_rate": 4.3529554263565894e-05, | |
| "loss": 0.0767, | |
| "step": 3171 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 0.025242358861939478, | |
| "eval_loss": 0.23418210446834564, | |
| "eval_rouge1": 0.2960794609105263, | |
| "eval_rouge2": 0.078188861596172, | |
| "eval_rougeL": 0.2910032597311343, | |
| "eval_runtime": 46.6571, | |
| "eval_samples_per_second": 45.288, | |
| "eval_steps_per_second": 5.68, | |
| "step": 3171 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.26465269923210144, | |
| "learning_rate": 4.096899224806201e-05, | |
| "loss": 0.0696, | |
| "step": 4228 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 0.040436621142289915, | |
| "eval_loss": 0.2277713567018509, | |
| "eval_rouge1": 0.3300020560602195, | |
| "eval_rouge2": 0.10496894572689558, | |
| "eval_rougeL": 0.32523455556499214, | |
| "eval_runtime": 46.6771, | |
| "eval_samples_per_second": 45.268, | |
| "eval_steps_per_second": 5.677, | |
| "step": 4228 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.19601380825042725, | |
| "learning_rate": 3.840843023255814e-05, | |
| "loss": 0.0636, | |
| "step": 5285 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 0.05166825934118283, | |
| "eval_loss": 0.22189512848854065, | |
| "eval_rouge1": 0.3535589938019066, | |
| "eval_rouge2": 0.12152558383102884, | |
| "eval_rougeL": 0.3480324766784169, | |
| "eval_runtime": 76.0841, | |
| "eval_samples_per_second": 27.772, | |
| "eval_steps_per_second": 3.483, | |
| "step": 5285 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.22665895521640778, | |
| "learning_rate": 3.5847868217054265e-05, | |
| "loss": 0.0587, | |
| "step": 6342 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 0.05899956949346335, | |
| "eval_loss": 0.22368572652339935, | |
| "eval_rouge1": 0.3654354164962346, | |
| "eval_rouge2": 0.13482081314433186, | |
| "eval_rougeL": 0.3611275649771175, | |
| "eval_runtime": 108.2192, | |
| "eval_samples_per_second": 19.525, | |
| "eval_steps_per_second": 2.449, | |
| "step": 6342 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.1510099470615387, | |
| "learning_rate": 3.328730620155039e-05, | |
| "loss": 0.0542, | |
| "step": 7399 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 0.06674675327067288, | |
| "eval_loss": 0.21935084462165833, | |
| "eval_rouge1": 0.375454032579127, | |
| "eval_rouge2": 0.1440449212655856, | |
| "eval_rougeL": 0.37123789211821234, | |
| "eval_runtime": 169.6548, | |
| "eval_samples_per_second": 12.455, | |
| "eval_steps_per_second": 1.562, | |
| "step": 7399 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.17433832585811615, | |
| "learning_rate": 3.0726744186046517e-05, | |
| "loss": 0.0502, | |
| "step": 8456 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 0.07150751079471653, | |
| "eval_loss": 0.20800183713436127, | |
| "eval_rouge1": 0.38017302561406896, | |
| "eval_rouge2": 0.1520586373563908, | |
| "eval_rougeL": 0.3761265584375465, | |
| "eval_runtime": 59.9519, | |
| "eval_samples_per_second": 35.245, | |
| "eval_steps_per_second": 4.42, | |
| "step": 8456 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.23176324367523193, | |
| "learning_rate": 2.816618217054264e-05, | |
| "loss": 0.0468, | |
| "step": 9513 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 0.07701622828998697, | |
| "eval_loss": 0.21230556070804596, | |
| "eval_rouge1": 0.39305333263185294, | |
| "eval_rouge2": 0.16160655664305132, | |
| "eval_rougeL": 0.3889099045386816, | |
| "eval_runtime": 47.0282, | |
| "eval_samples_per_second": 44.93, | |
| "eval_steps_per_second": 5.635, | |
| "step": 9513 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.22061537206172943, | |
| "learning_rate": 2.560562015503876e-05, | |
| "loss": 0.0438, | |
| "step": 10570 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 0.08116803822206035, | |
| "eval_loss": 0.21123112738132477, | |
| "eval_rouge1": 0.39211790695723037, | |
| "eval_rouge2": 0.16476729267221837, | |
| "eval_rougeL": 0.3884162333833129, | |
| "eval_runtime": 52.3744, | |
| "eval_samples_per_second": 40.344, | |
| "eval_steps_per_second": 5.06, | |
| "step": 10570 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.29454106092453003, | |
| "learning_rate": 2.3045058139534884e-05, | |
| "loss": 0.0408, | |
| "step": 11627 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bleu": 0.08163331720700502, | |
| "eval_loss": 0.21018122136592865, | |
| "eval_rouge1": 0.39673991981574686, | |
| "eval_rouge2": 0.16533579969575996, | |
| "eval_rougeL": 0.3935895133910881, | |
| "eval_runtime": 169.7591, | |
| "eval_samples_per_second": 12.447, | |
| "eval_steps_per_second": 1.561, | |
| "step": 11627 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.20639048516750336, | |
| "learning_rate": 2.0484496124031007e-05, | |
| "loss": 0.0384, | |
| "step": 12684 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_bleu": 0.08445355735141684, | |
| "eval_loss": 0.20781023800373077, | |
| "eval_rouge1": 0.40183732359428986, | |
| "eval_rouge2": 0.17106600485502385, | |
| "eval_rougeL": 0.39778146195081354, | |
| "eval_runtime": 169.7164, | |
| "eval_samples_per_second": 12.45, | |
| "eval_steps_per_second": 1.561, | |
| "step": 12684 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.206428661942482, | |
| "learning_rate": 1.7923934108527132e-05, | |
| "loss": 0.0363, | |
| "step": 13741 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_bleu": 0.08696571570490616, | |
| "eval_loss": 0.2144923061132431, | |
| "eval_rouge1": 0.4023135478695373, | |
| "eval_rouge2": 0.1719689510122407, | |
| "eval_rougeL": 0.39862674092398764, | |
| "eval_runtime": 169.681, | |
| "eval_samples_per_second": 12.453, | |
| "eval_steps_per_second": 1.562, | |
| "step": 13741 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.1995713710784912, | |
| "learning_rate": 1.5363372093023258e-05, | |
| "loss": 0.0343, | |
| "step": 14798 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_bleu": 0.08783833270174173, | |
| "eval_loss": 0.21651192009449005, | |
| "eval_rouge1": 0.406251961398795, | |
| "eval_rouge2": 0.17568144009642983, | |
| "eval_rougeL": 0.4022547036607911, | |
| "eval_runtime": 169.5372, | |
| "eval_samples_per_second": 12.463, | |
| "eval_steps_per_second": 1.563, | |
| "step": 14798 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.22589825093746185, | |
| "learning_rate": 1.280281007751938e-05, | |
| "loss": 0.0327, | |
| "step": 15855 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_bleu": 0.09203652391095736, | |
| "eval_loss": 0.21689023077487946, | |
| "eval_rouge1": 0.4049011529575125, | |
| "eval_rouge2": 0.17920037507226205, | |
| "eval_rougeL": 0.4013754949817908, | |
| "eval_runtime": 169.7215, | |
| "eval_samples_per_second": 12.45, | |
| "eval_steps_per_second": 1.561, | |
| "step": 15855 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.2390994280576706, | |
| "learning_rate": 1.0242248062015503e-05, | |
| "loss": 0.0313, | |
| "step": 16912 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_bleu": 0.09196638031992929, | |
| "eval_loss": 0.21753403544425964, | |
| "eval_rouge1": 0.4078242153735353, | |
| "eval_rouge2": 0.18209517089320632, | |
| "eval_rougeL": 0.40483345816401317, | |
| "eval_runtime": 169.6048, | |
| "eval_samples_per_second": 12.458, | |
| "eval_steps_per_second": 1.562, | |
| "step": 16912 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.31637752056121826, | |
| "learning_rate": 7.681686046511629e-06, | |
| "loss": 0.0301, | |
| "step": 17969 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_bleu": 0.09441216110381762, | |
| "eval_loss": 0.2191299945116043, | |
| "eval_rouge1": 0.41033944929551286, | |
| "eval_rouge2": 0.18390791273310297, | |
| "eval_rougeL": 0.40661041910814566, | |
| "eval_runtime": 169.743, | |
| "eval_samples_per_second": 12.448, | |
| "eval_steps_per_second": 1.561, | |
| "step": 17969 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "step": 17969, | |
| "total_flos": 7.5086968651776e+16, | |
| "train_loss": 0.10163650903314878, | |
| "train_runtime": 11373.3667, | |
| "train_samples_per_second": 14.863, | |
| "train_steps_per_second": 1.859 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 21140, | |
| "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": 7.5086968651776e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |