| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 111, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 18.24598685517398, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7167, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 12.725400800060369, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5798, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.40540540540540543, | |
| "grad_norm": 10.09031183252554, | |
| "learning_rate": 9.939652198703783e-07, | |
| "loss": 1.2993, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 4.301913372512592, | |
| "learning_rate": 9.76006553766365e-07, | |
| "loss": 1.1471, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "grad_norm": 4.127159967721974, | |
| "learning_rate": 9.465575080933957e-07, | |
| "loss": 1.0671, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 3.7189876826316546, | |
| "learning_rate": 9.063289569141251e-07, | |
| "loss": 1.0305, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9459459459459459, | |
| "grad_norm": 3.6454444504894727, | |
| "learning_rate": 8.562919820737535e-07, | |
| "loss": 1.0007, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 3.406922222433217, | |
| "learning_rate": 7.97654432138333e-07, | |
| "loss": 0.946, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 3.392252157242342, | |
| "learning_rate": 7.318317659926636e-07, | |
| "loss": 0.9231, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 3.6984654109661985, | |
| "learning_rate": 6.604128849076838e-07, | |
| "loss": 0.9036, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "eval_loss": 0.9227461814880371, | |
| "eval_runtime": 18.9023, | |
| "eval_samples_per_second": 55.549, | |
| "eval_steps_per_second": 0.899, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4864864864864864, | |
| "grad_norm": 3.3956890906900323, | |
| "learning_rate": 5.851217778611993e-07, | |
| "loss": 0.8885, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 3.3729167181306634, | |
| "learning_rate": 5.077759059601755e-07, | |
| "loss": 0.8794, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.7567567567567568, | |
| "grad_norm": 3.4187464621946004, | |
| "learning_rate": 4.302423305280385e-07, | |
| "loss": 0.879, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 3.4454278618775174, | |
| "learning_rate": 3.54392643886374e-07, | |
| "loss": 0.8643, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 3.9861061655783137, | |
| "learning_rate": 2.8205779076231446e-07, | |
| "loss": 0.8416, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 3.318571264456712, | |
| "learning_rate": 2.1498387089310865e-07, | |
| "loss": 0.8208, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.2972972972972974, | |
| "grad_norm": 3.517096902631208, | |
| "learning_rate": 1.5478998971412666e-07, | |
| "loss": 0.7978, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 3.471438197905033, | |
| "learning_rate": 1.0292917457762323e-07, | |
| "loss": 0.8326, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5675675675675675, | |
| "grad_norm": 3.4399400874271566, | |
| "learning_rate": 6.065329995036572e-08, | |
| "loss": 0.8153, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 3.7525998804223675, | |
| "learning_rate": 2.8982868265005454e-08, | |
| "loss": 0.7997, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "eval_loss": 0.8954470157623291, | |
| "eval_runtime": 18.4984, | |
| "eval_samples_per_second": 56.762, | |
| "eval_steps_per_second": 0.919, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.8378378378378377, | |
| "grad_norm": 3.3167347885160554, | |
| "learning_rate": 8.682375888868166e-09, | |
| "loss": 0.7965, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 3.93684166254045, | |
| "learning_rate": 2.418588540059607e-10, | |
| "loss": 0.8273, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 111, | |
| "total_flos": 654272366641152.0, | |
| "train_loss": 0.9825769034591881, | |
| "train_runtime": 1528.2927, | |
| "train_samples_per_second": 18.55, | |
| "train_steps_per_second": 0.073 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 111, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 654272366641152.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |