| { | |
| "best_metric": 0.465503990650177, | |
| "best_model_checkpoint": "/home/iais_marenpielka/Bouthaina/res_nw_yem/checkpoint-918", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 1683, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 41.373477935791016, | |
| "learning_rate": 1.53e-05, | |
| "loss": 7.4125, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.013470228801119116, | |
| "eval_loss": 2.3318939208984375, | |
| "eval_rouge1": 0.4042810146176442, | |
| "eval_rouge2": 0.1765421862019638, | |
| "eval_rougeL": 0.40385731535628305, | |
| "eval_runtime": 1.2315, | |
| "eval_samples_per_second": 246.862, | |
| "eval_steps_per_second": 30.858, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.7725179195404053, | |
| "learning_rate": 3.06e-05, | |
| "loss": 1.0077, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.3375849927940314, | |
| "eval_loss": 0.5519431233406067, | |
| "eval_rouge1": 0.47471630348624405, | |
| "eval_rouge2": 0.18392227454023974, | |
| "eval_rougeL": 0.4733153598700648, | |
| "eval_runtime": 1.1576, | |
| "eval_samples_per_second": 262.603, | |
| "eval_steps_per_second": 32.825, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.682711124420166, | |
| "learning_rate": 4.5900000000000004e-05, | |
| "loss": 0.483, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 0.35824220855964284, | |
| "eval_loss": 0.4972667694091797, | |
| "eval_rouge1": 0.5145197705108664, | |
| "eval_rouge2": 0.2200386914249186, | |
| "eval_rougeL": 0.5134621462419481, | |
| "eval_runtime": 1.2102, | |
| "eval_samples_per_second": 251.205, | |
| "eval_steps_per_second": 31.401, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.890738606452942, | |
| "learning_rate": 4.7812500000000003e-05, | |
| "loss": 0.3795, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 0.3717354832918087, | |
| "eval_loss": 0.4739198088645935, | |
| "eval_rouge1": 0.5397330658204715, | |
| "eval_rouge2": 0.2524237751117604, | |
| "eval_rougeL": 0.5377765269792509, | |
| "eval_runtime": 1.2779, | |
| "eval_samples_per_second": 237.89, | |
| "eval_steps_per_second": 29.736, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.3704463243484497, | |
| "learning_rate": 4.482421875e-05, | |
| "loss": 0.2901, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 0.37954739911448565, | |
| "eval_loss": 0.46911630034446716, | |
| "eval_rouge1": 0.57101655669895, | |
| "eval_rouge2": 0.29268127822000256, | |
| "eval_rougeL": 0.567665717772471, | |
| "eval_runtime": 1.1189, | |
| "eval_samples_per_second": 271.684, | |
| "eval_steps_per_second": 33.96, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.911377429962158, | |
| "learning_rate": 4.18359375e-05, | |
| "loss": 0.2239, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 0.39471976173966783, | |
| "eval_loss": 0.465503990650177, | |
| "eval_rouge1": 0.5909012601210781, | |
| "eval_rouge2": 0.32323469643447167, | |
| "eval_rougeL": 0.5882112866531675, | |
| "eval_runtime": 5.248, | |
| "eval_samples_per_second": 57.927, | |
| "eval_steps_per_second": 7.241, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.639866590499878, | |
| "learning_rate": 3.884765625e-05, | |
| "loss": 0.1763, | |
| "step": 1071 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 0.4011122295711346, | |
| "eval_loss": 0.46726807951927185, | |
| "eval_rouge1": 0.6046871760857919, | |
| "eval_rouge2": 0.3378049073324742, | |
| "eval_rougeL": 0.6018660127815343, | |
| "eval_runtime": 14.6215, | |
| "eval_samples_per_second": 20.791, | |
| "eval_steps_per_second": 2.599, | |
| "step": 1071 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.1884260177612305, | |
| "learning_rate": 3.5859375e-05, | |
| "loss": 0.1429, | |
| "step": 1224 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 0.4084847317717048, | |
| "eval_loss": 0.46664053201675415, | |
| "eval_rouge1": 0.619977821151152, | |
| "eval_rouge2": 0.3652452050559121, | |
| "eval_rougeL": 0.6168287532556173, | |
| "eval_runtime": 13.3071, | |
| "eval_samples_per_second": 22.845, | |
| "eval_steps_per_second": 2.856, | |
| "step": 1224 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.133341908454895, | |
| "learning_rate": 3.287109375e-05, | |
| "loss": 0.1206, | |
| "step": 1377 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 0.412982833438841, | |
| "eval_loss": 0.47646617889404297, | |
| "eval_rouge1": 0.6230350047425831, | |
| "eval_rouge2": 0.37065368099082896, | |
| "eval_rougeL": 0.6207890023241163, | |
| "eval_runtime": 4.1195, | |
| "eval_samples_per_second": 73.796, | |
| "eval_steps_per_second": 9.225, | |
| "step": 1377 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.0680757761001587, | |
| "learning_rate": 2.9882812500000002e-05, | |
| "loss": 0.1052, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 0.41481558555871684, | |
| "eval_loss": 0.47586962580680847, | |
| "eval_rouge1": 0.6232860618818599, | |
| "eval_rouge2": 0.3762342947707073, | |
| "eval_rougeL": 0.6204668025380538, | |
| "eval_runtime": 1.7191, | |
| "eval_samples_per_second": 176.836, | |
| "eval_steps_per_second": 22.105, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.2003881931304932, | |
| "learning_rate": 2.689453125e-05, | |
| "loss": 0.0955, | |
| "step": 1683 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bleu": 0.4179443502693027, | |
| "eval_loss": 0.47885996103286743, | |
| "eval_rouge1": 0.6342804965319174, | |
| "eval_rouge2": 0.3809817493272587, | |
| "eval_rougeL": 0.6306668222386218, | |
| "eval_runtime": 1.9634, | |
| "eval_samples_per_second": 154.831, | |
| "eval_steps_per_second": 19.354, | |
| "step": 1683 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 1683, | |
| "total_flos": 875916214272000.0, | |
| "train_loss": 0.9488435417714746, | |
| "train_runtime": 1368.0254, | |
| "train_samples_per_second": 17.821, | |
| "train_steps_per_second": 2.237 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3060, | |
| "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": 875916214272000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |