{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.9998833002684093, "eval_steps": 100, "global_step": 2142, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.09335978527249388, "grad_norm": 5.347364902496338, "learning_rate": 9.897551417342852e-05, "loss": 4.6888, "step": 100 }, { "epoch": 0.09335978527249388, "eval_loss": 0.3825870752334595, "eval_runtime": 23.2535, "eval_samples_per_second": 3.741, "eval_steps_per_second": 3.741, "step": 100 }, { "epoch": 0.18671957054498775, "grad_norm": 5.412732124328613, "learning_rate": 9.374809314115647e-05, "loss": 2.3316, "step": 200 }, { "epoch": 0.18671957054498775, "eval_loss": 0.25006037950515747, "eval_runtime": 23.4002, "eval_samples_per_second": 3.718, "eval_steps_per_second": 3.718, "step": 200 }, { "epoch": 0.2800793558174816, "grad_norm": 2.9451260566711426, "learning_rate": 8.454375586546417e-05, "loss": 1.5881, "step": 300 }, { "epoch": 0.2800793558174816, "eval_loss": 0.1890459656715393, "eval_runtime": 23.2549, "eval_samples_per_second": 3.741, "eval_steps_per_second": 3.741, "step": 300 }, { "epoch": 0.3734391410899755, "grad_norm": 5.892703533172607, "learning_rate": 7.219922175040683e-05, "loss": 1.2221, "step": 400 }, { "epoch": 0.3734391410899755, "eval_loss": 0.14931949973106384, "eval_runtime": 23.3305, "eval_samples_per_second": 3.729, "eval_steps_per_second": 3.729, "step": 400 }, { "epoch": 0.4667989263624694, "grad_norm": 3.341991424560547, "learning_rate": 5.783666949486448e-05, "loss": 1.038, "step": 500 }, { "epoch": 0.4667989263624694, "eval_loss": 0.13619396090507507, "eval_runtime": 23.3865, "eval_samples_per_second": 3.72, "eval_steps_per_second": 3.72, "step": 500 }, { "epoch": 0.5601587116349632, "grad_norm": 2.6955649852752686, "learning_rate": 4.276172554471998e-05, "loss": 0.9644, "step": 600 }, { "epoch": 0.5601587116349632, "eval_loss": 0.1279439479112625, "eval_runtime": 23.3534, "eval_samples_per_second": 3.725, "eval_steps_per_second": 3.725, "step": 600 }, { "epoch": 0.6535184969074571, "grad_norm": 3.1168127059936523, "learning_rate": 2.834477624505443e-05, "loss": 0.8806, "step": 700 }, { "epoch": 0.6535184969074571, "eval_loss": 0.12419606000185013, "eval_runtime": 23.2939, "eval_samples_per_second": 3.735, "eval_steps_per_second": 3.735, "step": 700 }, { "epoch": 0.746878282179951, "grad_norm": 2.7166848182678223, "learning_rate": 1.5896392999869652e-05, "loss": 0.8333, "step": 800 }, { "epoch": 0.746878282179951, "eval_loss": 0.1232759952545166, "eval_runtime": 23.4216, "eval_samples_per_second": 3.715, "eval_steps_per_second": 3.715, "step": 800 }, { "epoch": 0.8402380674524449, "grad_norm": 2.7149593830108643, "learning_rate": 6.548194903336408e-06, "loss": 0.8307, "step": 900 }, { "epoch": 0.8402380674524449, "eval_loss": 0.12131741642951965, "eval_runtime": 23.2806, "eval_samples_per_second": 3.737, "eval_steps_per_second": 3.737, "step": 900 }, { "epoch": 0.9335978527249388, "grad_norm": 2.5703158378601074, "learning_rate": 1.149979013737007e-06, "loss": 0.8138, "step": 1000 }, { "epoch": 0.9335978527249388, "eval_loss": 0.12097638100385666, "eval_runtime": 23.3058, "eval_samples_per_second": 3.733, "eval_steps_per_second": 3.733, "step": 1000 }, { "epoch": 1.0270743377290232, "grad_norm": 1.5639070272445679, "learning_rate": 5.02646972298292e-05, "loss": 0.4182, "step": 1100 }, { "epoch": 1.0270743377290232, "eval_loss": 0.12452684342861176, "eval_runtime": 11.3284, "eval_samples_per_second": 7.68, "eval_steps_per_second": 7.68, "step": 1100 }, { "epoch": 1.1204341230015171, "grad_norm": 1.3367255926132202, "learning_rate": 4.2727771933821015e-05, "loss": 0.807, "step": 1200 }, { "epoch": 1.1204341230015171, "eval_loss": 0.12348207086324692, "eval_runtime": 11.3327, "eval_samples_per_second": 7.677, "eval_steps_per_second": 7.677, "step": 1200 }, { "epoch": 1.213793908274011, "grad_norm": 1.3575553894042969, "learning_rate": 3.535690704015874e-05, "loss": 0.7986, "step": 1300 }, { "epoch": 1.213793908274011, "eval_loss": 0.12156575173139572, "eval_runtime": 11.3167, "eval_samples_per_second": 7.688, "eval_steps_per_second": 7.688, "step": 1300 }, { "epoch": 1.307153693546505, "grad_norm": 1.1323909759521484, "learning_rate": 2.8320415310590165e-05, "loss": 0.769, "step": 1400 }, { "epoch": 1.307153693546505, "eval_loss": 0.11894583702087402, "eval_runtime": 11.3324, "eval_samples_per_second": 7.677, "eval_steps_per_second": 7.677, "step": 1400 }, { "epoch": 1.4005134788189988, "grad_norm": 0.9585688710212708, "learning_rate": 2.1778974138217168e-05, "loss": 0.7456, "step": 1500 }, { "epoch": 1.4005134788189988, "eval_loss": 0.11771341413259506, "eval_runtime": 11.3495, "eval_samples_per_second": 7.666, "eval_steps_per_second": 7.666, "step": 1500 }, { "epoch": 1.4938732640914925, "grad_norm": 1.1619073152542114, "learning_rate": 1.58819564996737e-05, "loss": 0.7581, "step": 1600 }, { "epoch": 1.4938732640914925, "eval_loss": 0.11598959565162659, "eval_runtime": 11.3352, "eval_samples_per_second": 7.675, "eval_steps_per_second": 7.675, "step": 1600 }, { "epoch": 1.5872330493639866, "grad_norm": 0.9958471655845642, "learning_rate": 1.0764020042215306e-05, "loss": 0.7593, "step": 1700 }, { "epoch": 1.5872330493639866, "eval_loss": 0.1154765784740448, "eval_runtime": 11.381, "eval_samples_per_second": 7.644, "eval_steps_per_second": 7.644, "step": 1700 }, { "epoch": 1.6805928346364802, "grad_norm": 1.3259460926055908, "learning_rate": 6.54203219348205e-06, "loss": 0.7625, "step": 1800 }, { "epoch": 1.6805928346364802, "eval_loss": 0.11468129605054855, "eval_runtime": 11.3548, "eval_samples_per_second": 7.662, "eval_steps_per_second": 7.662, "step": 1800 }, { "epoch": 1.7739526199089743, "grad_norm": 1.9127427339553833, "learning_rate": 3.3124015086622216e-06, "loss": 0.7506, "step": 1900 }, { "epoch": 1.7739526199089743, "eval_loss": 0.11443212628364563, "eval_runtime": 11.3509, "eval_samples_per_second": 7.665, "eval_steps_per_second": 7.665, "step": 1900 }, { "epoch": 1.867312405181468, "grad_norm": 1.113871693611145, "learning_rate": 1.1488761934046633e-06, "loss": 0.7525, "step": 2000 }, { "epoch": 1.867312405181468, "eval_loss": 0.11432737857103348, "eval_runtime": 11.3318, "eval_samples_per_second": 7.678, "eval_steps_per_second": 7.678, "step": 2000 }, { "epoch": 1.960672190453962, "grad_norm": 1.360288381576538, "learning_rate": 1.008600729287823e-07, "loss": 0.7349, "step": 2100 }, { "epoch": 1.960672190453962, "eval_loss": 0.11436222493648529, "eval_runtime": 11.2327, "eval_samples_per_second": 7.745, "eval_steps_per_second": 7.745, "step": 2100 } ], "logging_steps": 100, "max_steps": 2142, "num_input_tokens_seen": 0, "num_train_epochs": 2, "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": 1.6817613284788224e+17, "train_batch_size": 1, "trial_name": null, "trial_params": null }