| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7545638945233266, | |
| "eval_steps": 31, | |
| "global_step": 93, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008113590263691683, | |
| "eval_loss": 5.708795547485352, | |
| "eval_runtime": 34.9917, | |
| "eval_samples_per_second": 5.944, | |
| "eval_steps_per_second": 0.743, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02434077079107505, | |
| "grad_norm": 27.649152755737305, | |
| "learning_rate": 3e-05, | |
| "loss": 22.4374, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0486815415821501, | |
| "grad_norm": 24.12411880493164, | |
| "learning_rate": 6e-05, | |
| "loss": 20.3477, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07302231237322515, | |
| "grad_norm": 22.957651138305664, | |
| "learning_rate": 9e-05, | |
| "loss": 13.9237, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0973630831643002, | |
| "grad_norm": 22.14281463623047, | |
| "learning_rate": 9.999238475781957e-05, | |
| "loss": 8.2192, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12170385395537525, | |
| "grad_norm": 15.811911582946777, | |
| "learning_rate": 9.99524110790929e-05, | |
| "loss": 5.4619, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1460446247464503, | |
| "grad_norm": 17.17688751220703, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.1998, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.17038539553752535, | |
| "grad_norm": 9.06302547454834, | |
| "learning_rate": 9.976980991835894e-05, | |
| "loss": 3.5102, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1947261663286004, | |
| "grad_norm": 9.071863174438477, | |
| "learning_rate": 9.962730758206611e-05, | |
| "loss": 2.9501, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.21906693711967545, | |
| "grad_norm": 10.715943336486816, | |
| "learning_rate": 9.945079316809585e-05, | |
| "loss": 2.8262, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2434077079107505, | |
| "grad_norm": 6.811004161834717, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.6217, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2515212981744422, | |
| "eval_loss": 0.6682190299034119, | |
| "eval_runtime": 35.2448, | |
| "eval_samples_per_second": 5.902, | |
| "eval_steps_per_second": 0.738, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.26774847870182555, | |
| "grad_norm": 5.427604675292969, | |
| "learning_rate": 9.899623523104149e-05, | |
| "loss": 2.4006, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2920892494929006, | |
| "grad_norm": 7.537416458129883, | |
| "learning_rate": 9.871850323926177e-05, | |
| "loss": 2.7397, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.31643002028397565, | |
| "grad_norm": 8.674805641174316, | |
| "learning_rate": 9.84073820189054e-05, | |
| "loss": 2.4828, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.3407707910750507, | |
| "grad_norm": 8.167730331420898, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.5302, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.36511156186612576, | |
| "grad_norm": 10.937932014465332, | |
| "learning_rate": 9.768584753741134e-05, | |
| "loss": 2.3383, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3894523326572008, | |
| "grad_norm": 3.8246636390686035, | |
| "learning_rate": 9.727592877996585e-05, | |
| "loss": 2.212, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.41379310344827586, | |
| "grad_norm": 4.667800426483154, | |
| "learning_rate": 9.683360946241989e-05, | |
| "loss": 2.3763, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.4381338742393509, | |
| "grad_norm": 4.990477561950684, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.3995, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.46247464503042596, | |
| "grad_norm": 9.09072494506836, | |
| "learning_rate": 9.58530037192562e-05, | |
| "loss": 2.545, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.486815415821501, | |
| "grad_norm": 5.338695049285889, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 2.3183, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5030425963488844, | |
| "eval_loss": 0.5645254850387573, | |
| "eval_runtime": 13.4155, | |
| "eval_samples_per_second": 15.504, | |
| "eval_steps_per_second": 1.938, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.5111561866125761, | |
| "grad_norm": 3.9001667499542236, | |
| "learning_rate": 9.474671808010126e-05, | |
| "loss": 2.3843, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.5354969574036511, | |
| "grad_norm": 6.379414081573486, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.2963, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.5598377281947262, | |
| "grad_norm": 4.97390079498291, | |
| "learning_rate": 9.351778479699499e-05, | |
| "loss": 2.2517, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.5841784989858012, | |
| "grad_norm": 4.063937187194824, | |
| "learning_rate": 9.285836503510562e-05, | |
| "loss": 2.2437, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.6085192697768763, | |
| "grad_norm": 3.602717161178589, | |
| "learning_rate": 9.21695722906443e-05, | |
| "loss": 2.1695, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6328600405679513, | |
| "grad_norm": 4.555128574371338, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.1648, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.6572008113590264, | |
| "grad_norm": 5.0544867515563965, | |
| "learning_rate": 9.070577591781597e-05, | |
| "loss": 2.0922, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.6815415821501014, | |
| "grad_norm": 4.92878532409668, | |
| "learning_rate": 8.993177550236464e-05, | |
| "loss": 2.141, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 3.6681647300720215, | |
| "learning_rate": 8.91304078426207e-05, | |
| "loss": 2.1638, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.7302231237322515, | |
| "grad_norm": 4.451002597808838, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.1766, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7545638945233266, | |
| "grad_norm": 5.902685165405273, | |
| "learning_rate": 8.744778603945011e-05, | |
| "loss": 2.1005, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.7545638945233266, | |
| "eval_loss": 0.5568718314170837, | |
| "eval_runtime": 13.4611, | |
| "eval_samples_per_second": 15.452, | |
| "eval_steps_per_second": 1.931, | |
| "step": 93 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 370, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 31, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2373610700629606e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |