| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 100, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12578616352201258, | |
| "grad_norm": 1.6983928680419922, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9207, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.25157232704402516, | |
| "grad_norm": 2.0057549476623535, | |
| "learning_rate": 2.25e-05, | |
| "loss": 1.8211, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.37735849056603776, | |
| "grad_norm": 2.055621385574341, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.517, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5031446540880503, | |
| "grad_norm": 1.7859697341918945, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.9972, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6289308176100629, | |
| "grad_norm": 2.006009817123413, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.6685, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7547169811320755, | |
| "grad_norm": 1.7193933725357056, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 0.521, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8805031446540881, | |
| "grad_norm": 1.2538529634475708, | |
| "learning_rate": 4.925739315689991e-05, | |
| "loss": 0.4388, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.2421389818191528, | |
| "learning_rate": 4.8637964389982926e-05, | |
| "loss": 0.4147, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1257861635220126, | |
| "grad_norm": 1.097655177116394, | |
| "learning_rate": 4.783863644106502e-05, | |
| "loss": 0.3337, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.251572327044025, | |
| "grad_norm": 1.446738839149475, | |
| "learning_rate": 4.6865492678484895e-05, | |
| "loss": 0.3674, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.3773584905660377, | |
| "grad_norm": 1.0339725017547607, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.3454, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.5031446540880502, | |
| "grad_norm": 1.1563431024551392, | |
| "learning_rate": 4.442864903642428e-05, | |
| "loss": 0.3299, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6289308176100628, | |
| "grad_norm": 1.6715773344039917, | |
| "learning_rate": 4.2983495008466276e-05, | |
| "loss": 0.3243, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.7547169811320755, | |
| "grad_norm": 4.052298545837402, | |
| "learning_rate": 4.140147572476268e-05, | |
| "loss": 0.3037, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.880503144654088, | |
| "grad_norm": 1.0090709924697876, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 0.2797, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.0944145917892456, | |
| "learning_rate": 3.787595187275136e-05, | |
| "loss": 0.2966, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.1257861635220126, | |
| "grad_norm": 0.9973664879798889, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.2336, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.251572327044025, | |
| "grad_norm": 0.9893947839736938, | |
| "learning_rate": 3.39591987386325e-05, | |
| "loss": 0.2068, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.3773584905660377, | |
| "grad_norm": 1.6538398265838623, | |
| "learning_rate": 3.1890933895424976e-05, | |
| "loss": 0.2066, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.50314465408805, | |
| "grad_norm": 1.2943580150604248, | |
| "learning_rate": 2.9770224884413623e-05, | |
| "loss": 0.2127, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.50314465408805, | |
| "eval_loss": 0.25331389904022217, | |
| "eval_runtime": 2.6526, | |
| "eval_samples_per_second": 13.572, | |
| "eval_steps_per_second": 6.786, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.6289308176100628, | |
| "grad_norm": 1.2970905303955078, | |
| "learning_rate": 2.761321158169134e-05, | |
| "loss": 0.2128, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.7547169811320753, | |
| "grad_norm": 1.2746000289916992, | |
| "learning_rate": 2.5436310160932092e-05, | |
| "loss": 0.2079, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.880503144654088, | |
| "grad_norm": 1.2576719522476196, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.222, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.5171639919281006, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 0.2071, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.1257861635220126, | |
| "grad_norm": 0.8614316582679749, | |
| "learning_rate": 1.895195261000831e-05, | |
| "loss": 0.1476, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 3.251572327044025, | |
| "grad_norm": 1.1631187200546265, | |
| "learning_rate": 1.686079613857109e-05, | |
| "loss": 0.1453, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.3773584905660377, | |
| "grad_norm": 1.5442155599594116, | |
| "learning_rate": 1.4831583923104999e-05, | |
| "loss": 0.1565, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.50314465408805, | |
| "grad_norm": 1.555525541305542, | |
| "learning_rate": 1.2879759493841575e-05, | |
| "loss": 0.1452, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.6289308176100628, | |
| "grad_norm": 2.1617038249969482, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.1494, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.7547169811320753, | |
| "grad_norm": 1.2040760517120361, | |
| "learning_rate": 9.266990223754069e-06, | |
| "loss": 0.1674, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.880503144654088, | |
| "grad_norm": 1.2778819799423218, | |
| "learning_rate": 7.633540738525066e-06, | |
| "loss": 0.1466, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.4193259477615356, | |
| "learning_rate": 6.1322604944307e-06, | |
| "loss": 0.1431, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.1257861635220126, | |
| "grad_norm": 0.9114992618560791, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 0.1318, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 4.251572327044025, | |
| "grad_norm": 0.9443201422691345, | |
| "learning_rate": 3.5708174824471947e-06, | |
| "loss": 0.1157, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.377358490566038, | |
| "grad_norm": 0.9356920719146729, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.1248, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 4.50314465408805, | |
| "grad_norm": 0.7772902846336365, | |
| "learning_rate": 1.6604893375699594e-06, | |
| "loss": 0.1201, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.628930817610063, | |
| "grad_norm": 1.1060445308685303, | |
| "learning_rate": 9.684576015420278e-07, | |
| "loss": 0.1147, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 4.754716981132075, | |
| "grad_norm": 0.6479562520980835, | |
| "learning_rate": 4.5932041380840065e-07, | |
| "loss": 0.1335, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.880503144654088, | |
| "grad_norm": 0.9411827325820923, | |
| "learning_rate": 1.3695261579316777e-07, | |
| "loss": 0.1151, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.4004632234573364, | |
| "learning_rate": 3.807621090218261e-09, | |
| "loss": 0.1242, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.22299104928970337, | |
| "eval_runtime": 2.1223, | |
| "eval_samples_per_second": 16.962, | |
| "eval_steps_per_second": 8.481, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 200, | |
| "total_flos": 8601980193079296.0, | |
| "train_loss": 0.369250630736351, | |
| "train_runtime": 349.7562, | |
| "train_samples_per_second": 4.546, | |
| "train_steps_per_second": 0.572 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8601980193079296.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |