| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01345985597954102, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005383942391816407, | |
| "grad_norm": 6.684871673583984, | |
| "learning_rate": 2e-05, | |
| "loss": 6.0997, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005383942391816407, | |
| "eval_loss": 6.004375457763672, | |
| "eval_runtime": 185.4714, | |
| "eval_samples_per_second": 8.438, | |
| "eval_steps_per_second": 4.222, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010767884783632815, | |
| "grad_norm": 6.894040107727051, | |
| "learning_rate": 4e-05, | |
| "loss": 6.1851, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0016151827175449223, | |
| "grad_norm": 6.502943992614746, | |
| "learning_rate": 6e-05, | |
| "loss": 6.03, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0016151827175449223, | |
| "eval_loss": 5.705997467041016, | |
| "eval_runtime": 185.4313, | |
| "eval_samples_per_second": 8.44, | |
| "eval_steps_per_second": 4.223, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002153576956726563, | |
| "grad_norm": 6.02913761138916, | |
| "learning_rate": 8e-05, | |
| "loss": 5.7463, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0026919711959082038, | |
| "grad_norm": 5.216821193695068, | |
| "learning_rate": 0.0001, | |
| "loss": 5.2279, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0032303654350898446, | |
| "grad_norm": 3.4631779193878174, | |
| "learning_rate": 0.00012, | |
| "loss": 4.6907, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0032303654350898446, | |
| "eval_loss": 4.778470039367676, | |
| "eval_runtime": 185.4262, | |
| "eval_samples_per_second": 8.44, | |
| "eval_steps_per_second": 4.223, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0037687596742714854, | |
| "grad_norm": 8.573506355285645, | |
| "learning_rate": 0.00014, | |
| "loss": 4.8765, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.004307153913453126, | |
| "grad_norm": 10.554330825805664, | |
| "learning_rate": 0.00016, | |
| "loss": 5.0392, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.004845548152634767, | |
| "grad_norm": 6.853347301483154, | |
| "learning_rate": 0.00018, | |
| "loss": 4.7749, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004845548152634767, | |
| "eval_loss": 4.601557731628418, | |
| "eval_runtime": 185.5481, | |
| "eval_samples_per_second": 8.434, | |
| "eval_steps_per_second": 4.22, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0053839423918164075, | |
| "grad_norm": 3.9383599758148193, | |
| "learning_rate": 0.0002, | |
| "loss": 4.51, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005922336630998048, | |
| "grad_norm": 3.0932722091674805, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 4.3751, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.006460730870179689, | |
| "grad_norm": 2.336444854736328, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 4.2279, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.006460730870179689, | |
| "eval_loss": 4.283242225646973, | |
| "eval_runtime": 185.6479, | |
| "eval_samples_per_second": 8.43, | |
| "eval_steps_per_second": 4.218, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00699912510936133, | |
| "grad_norm": 2.7305572032928467, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 4.1567, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.007537519348542971, | |
| "grad_norm": 2.3469622135162354, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 4.1394, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.00807591358772461, | |
| "grad_norm": 2.188025712966919, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 4.1941, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00807591358772461, | |
| "eval_loss": 4.151458740234375, | |
| "eval_runtime": 186.8151, | |
| "eval_samples_per_second": 8.377, | |
| "eval_steps_per_second": 4.191, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.008614307826906252, | |
| "grad_norm": 3.1134698390960693, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 4.0946, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.009152702066087893, | |
| "grad_norm": 2.663564682006836, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 3.9865, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.009691096305269533, | |
| "grad_norm": 1.9932688474655151, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 4.158, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.009691096305269533, | |
| "eval_loss": 4.052786827087402, | |
| "eval_runtime": 185.5821, | |
| "eval_samples_per_second": 8.433, | |
| "eval_steps_per_second": 4.219, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.010229490544451174, | |
| "grad_norm": 2.1913344860076904, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 4.1199, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.010767884783632815, | |
| "grad_norm": 2.3027989864349365, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 3.9268, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011306279022814456, | |
| "grad_norm": 2.2296042442321777, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 3.8072, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.011306279022814456, | |
| "eval_loss": 4.0160722732543945, | |
| "eval_runtime": 185.5603, | |
| "eval_samples_per_second": 8.434, | |
| "eval_steps_per_second": 4.22, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.011844673261996097, | |
| "grad_norm": 1.9645236730575562, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 4.1506, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.012383067501177738, | |
| "grad_norm": 2.151200294494629, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 4.0872, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.012921461740359378, | |
| "grad_norm": 1.6876325607299805, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 4.0389, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.012921461740359378, | |
| "eval_loss": 3.9968693256378174, | |
| "eval_runtime": 185.331, | |
| "eval_samples_per_second": 8.444, | |
| "eval_steps_per_second": 4.225, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01345985597954102, | |
| "grad_norm": 1.7987089157104492, | |
| "learning_rate": 0.0, | |
| "loss": 3.944, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.83447180771328e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |