| { | |
| "best_metric": 0.6265950798988342, | |
| "best_model_checkpoint": "/home/iais_marenpielka/Bouthaina/res_nw_dj/checkpoint-13515", | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 40545, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.3912957906723022, | |
| "learning_rate": 3.552372190826653e-05, | |
| "loss": 1.2513, | |
| "step": 2703 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.22005527068442493, | |
| "eval_loss": 0.7111806869506836, | |
| "eval_rouge1": 0.5171822713118965, | |
| "eval_rouge2": 0.253317924025756, | |
| "eval_rougeL": 0.5148202154962768, | |
| "eval_runtime": 21.8652, | |
| "eval_samples_per_second": 247.197, | |
| "eval_steps_per_second": 30.917, | |
| "step": 2703 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.4323362112045288, | |
| "learning_rate": 1.7761860954133264e-05, | |
| "loss": 0.6462, | |
| "step": 5406 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.24508343939723337, | |
| "eval_loss": 0.6569304466247559, | |
| "eval_rouge1": 0.557883823201639, | |
| "eval_rouge2": 0.2973836473978787, | |
| "eval_rougeL": 0.5553146565104614, | |
| "eval_runtime": 150.8243, | |
| "eval_samples_per_second": 35.836, | |
| "eval_steps_per_second": 4.482, | |
| "step": 5406 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2555302381515503, | |
| "learning_rate": 0.0, | |
| "loss": 0.5673, | |
| "step": 8109 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 0.2518224574721449, | |
| "eval_loss": 0.6498554348945618, | |
| "eval_rouge1": 0.57015304568835, | |
| "eval_rouge2": 0.3118394078609683, | |
| "eval_rougeL": 0.5676740384593599, | |
| "eval_runtime": 172.9357, | |
| "eval_samples_per_second": 31.254, | |
| "eval_steps_per_second": 3.909, | |
| "step": 8109 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.4183156490325928, | |
| "learning_rate": 4.037341299477222e-05, | |
| "loss": 0.556, | |
| "step": 10812 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 0.25983253831381714, | |
| "eval_loss": 0.63295978307724, | |
| "eval_rouge1": 0.5829310998998971, | |
| "eval_rouge2": 0.326193735265751, | |
| "eval_rougeL": 0.5807349807527458, | |
| "eval_runtime": 40.5509, | |
| "eval_samples_per_second": 133.289, | |
| "eval_steps_per_second": 16.67, | |
| "step": 10812 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.186664342880249, | |
| "learning_rate": 3.785007468259896e-05, | |
| "loss": 0.4786, | |
| "step": 13515 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 0.2678870499231116, | |
| "eval_loss": 0.6265950798988342, | |
| "eval_rouge1": 0.5977012354572853, | |
| "eval_rouge2": 0.34430833134800065, | |
| "eval_rougeL": 0.5958973349618409, | |
| "eval_runtime": 36.0053, | |
| "eval_samples_per_second": 150.117, | |
| "eval_steps_per_second": 18.775, | |
| "step": 13515 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.1597645282745361, | |
| "learning_rate": 3.5326736370425696e-05, | |
| "loss": 0.4123, | |
| "step": 16218 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 0.2707063976336222, | |
| "eval_loss": 0.630312979221344, | |
| "eval_rouge1": 0.604436172840783, | |
| "eval_rouge2": 0.35482389915380186, | |
| "eval_rougeL": 0.6026792061487433, | |
| "eval_runtime": 161.1219, | |
| "eval_samples_per_second": 33.546, | |
| "eval_steps_per_second": 4.196, | |
| "step": 16218 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.27555251121521, | |
| "learning_rate": 3.280339805825243e-05, | |
| "loss": 0.3573, | |
| "step": 18921 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 0.27393786087649125, | |
| "eval_loss": 0.6372247338294983, | |
| "eval_rouge1": 0.6108088318784769, | |
| "eval_rouge2": 0.36307215008060617, | |
| "eval_rougeL": 0.6088126794948809, | |
| "eval_runtime": 159.7397, | |
| "eval_samples_per_second": 33.836, | |
| "eval_steps_per_second": 4.232, | |
| "step": 18921 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.6310491561889648, | |
| "learning_rate": 3.0280059746079166e-05, | |
| "loss": 0.3108, | |
| "step": 21624 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 0.2759919170184585, | |
| "eval_loss": 0.646577000617981, | |
| "eval_rouge1": 0.6130373826214599, | |
| "eval_rouge2": 0.36776670773707465, | |
| "eval_rougeL": 0.6111699095881062, | |
| "eval_runtime": 150.1086, | |
| "eval_samples_per_second": 36.007, | |
| "eval_steps_per_second": 4.503, | |
| "step": 21624 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.6629250049591064, | |
| "learning_rate": 2.77567214339059e-05, | |
| "loss": 0.2719, | |
| "step": 24327 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 0.2801810856751608, | |
| "eval_loss": 0.6550981998443604, | |
| "eval_rouge1": 0.6177880665801061, | |
| "eval_rouge2": 0.3727206564119486, | |
| "eval_rougeL": 0.6159303776582042, | |
| "eval_runtime": 159.1145, | |
| "eval_samples_per_second": 33.969, | |
| "eval_steps_per_second": 4.249, | |
| "step": 24327 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.5184797048568726, | |
| "learning_rate": 2.523338312173264e-05, | |
| "loss": 0.2396, | |
| "step": 27030 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 0.27697504925957206, | |
| "eval_loss": 0.665830671787262, | |
| "eval_rouge1": 0.6173182845826588, | |
| "eval_rouge2": 0.37288754745117264, | |
| "eval_rougeL": 0.6152835100872982, | |
| "eval_runtime": 110.2964, | |
| "eval_samples_per_second": 49.004, | |
| "eval_steps_per_second": 6.129, | |
| "step": 27030 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.3458495140075684, | |
| "learning_rate": 2.2710044809559374e-05, | |
| "loss": 0.2129, | |
| "step": 29733 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bleu": 0.27843197191410246, | |
| "eval_loss": 0.6767598390579224, | |
| "eval_rouge1": 0.6188286228800556, | |
| "eval_rouge2": 0.3762109244532287, | |
| "eval_rougeL": 0.617052574223907, | |
| "eval_runtime": 24.0534, | |
| "eval_samples_per_second": 224.709, | |
| "eval_steps_per_second": 28.104, | |
| "step": 29733 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.636675477027893, | |
| "learning_rate": 2.018670649738611e-05, | |
| "loss": 0.191, | |
| "step": 32436 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_bleu": 0.2780108798800892, | |
| "eval_loss": 0.6870447993278503, | |
| "eval_rouge1": 0.6208121821010748, | |
| "eval_rouge2": 0.37810190638421814, | |
| "eval_rougeL": 0.6189436880506437, | |
| "eval_runtime": 42.9158, | |
| "eval_samples_per_second": 125.944, | |
| "eval_steps_per_second": 15.752, | |
| "step": 32436 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.3978990316390991, | |
| "learning_rate": 1.7663368185212848e-05, | |
| "loss": 0.1733, | |
| "step": 35139 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_bleu": 0.2799527424248887, | |
| "eval_loss": 0.696670651435852, | |
| "eval_rouge1": 0.6213835516562576, | |
| "eval_rouge2": 0.3799254363900967, | |
| "eval_rougeL": 0.6195026516671875, | |
| "eval_runtime": 17.6056, | |
| "eval_samples_per_second": 307.005, | |
| "eval_steps_per_second": 38.397, | |
| "step": 35139 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.386664628982544, | |
| "learning_rate": 1.5140029873039583e-05, | |
| "loss": 0.1593, | |
| "step": 37842 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_bleu": 0.2790160057234741, | |
| "eval_loss": 0.7063180208206177, | |
| "eval_rouge1": 0.6214879203521921, | |
| "eval_rouge2": 0.379862056883408, | |
| "eval_rougeL": 0.6194802915698101, | |
| "eval_runtime": 17.8263, | |
| "eval_samples_per_second": 303.204, | |
| "eval_steps_per_second": 37.921, | |
| "step": 37842 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.5141432285308838, | |
| "learning_rate": 1.261669156086632e-05, | |
| "loss": 0.1478, | |
| "step": 40545 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_bleu": 0.2808668290315019, | |
| "eval_loss": 0.7138631939888, | |
| "eval_rouge1": 0.6216459883291815, | |
| "eval_rouge2": 0.3804311054098596, | |
| "eval_rougeL": 0.6196983257570402, | |
| "eval_runtime": 17.6502, | |
| "eval_samples_per_second": 306.229, | |
| "eval_steps_per_second": 38.3, | |
| "step": 40545 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 40545, | |
| "total_flos": 2.118621118464e+16, | |
| "train_loss": 0.058961041791913305, | |
| "train_runtime": 1668.9384, | |
| "train_samples_per_second": 259.111, | |
| "train_steps_per_second": 32.392 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 54060, | |
| "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": 2.118621118464e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |