| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 120.0, | |
| "eval_steps": 17748, | |
| "global_step": 177480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 22.951730728149414, | |
| "learning_rate": 9.000676132521975e-06, | |
| "loss": 2.3088, | |
| "step": 17748 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 3.187467336654663, | |
| "eval_runtime": 33.5588, | |
| "eval_samples_per_second": 289.492, | |
| "eval_steps_per_second": 6.585, | |
| "step": 17748 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 2.020862579345703, | |
| "learning_rate": 8.001126887536624e-06, | |
| "loss": 2.0566, | |
| "step": 35496 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 3.3423843383789062, | |
| "eval_runtime": 31.6561, | |
| "eval_samples_per_second": 306.892, | |
| "eval_steps_per_second": 6.981, | |
| "step": 35496 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 8.087705612182617, | |
| "learning_rate": 7.0016339869281044e-06, | |
| "loss": 1.9815, | |
| "step": 53244 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 3.454538345336914, | |
| "eval_runtime": 31.7681, | |
| "eval_samples_per_second": 305.81, | |
| "eval_steps_per_second": 6.957, | |
| "step": 53244 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 7.905551433563232, | |
| "learning_rate": 6.0020847419427546e-06, | |
| "loss": 1.9566, | |
| "step": 70992 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 3.566826343536377, | |
| "eval_runtime": 31.9459, | |
| "eval_samples_per_second": 304.108, | |
| "eval_steps_per_second": 6.918, | |
| "step": 70992 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 1.7085500955581665, | |
| "learning_rate": 5.0025918413342355e-06, | |
| "loss": 1.9488, | |
| "step": 88740 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 3.5228500366210938, | |
| "eval_runtime": 31.684, | |
| "eval_samples_per_second": 306.622, | |
| "eval_steps_per_second": 6.975, | |
| "step": 88740 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "grad_norm": 0.28708475828170776, | |
| "learning_rate": 4.003098940725716e-06, | |
| "loss": 1.9424, | |
| "step": 106488 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_loss": 3.677123785018921, | |
| "eval_runtime": 33.4796, | |
| "eval_samples_per_second": 290.177, | |
| "eval_steps_per_second": 6.601, | |
| "step": 106488 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "grad_norm": 0.5243535041809082, | |
| "learning_rate": 3.0035496957403653e-06, | |
| "loss": 1.9411, | |
| "step": 124236 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_loss": 3.786760091781616, | |
| "eval_runtime": 31.6666, | |
| "eval_samples_per_second": 306.79, | |
| "eval_steps_per_second": 6.979, | |
| "step": 124236 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "grad_norm": 0.4514042139053345, | |
| "learning_rate": 2.004056795131846e-06, | |
| "loss": 1.9388, | |
| "step": 141984 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_loss": 3.7067413330078125, | |
| "eval_runtime": 33.5223, | |
| "eval_samples_per_second": 289.807, | |
| "eval_steps_per_second": 6.593, | |
| "step": 141984 | |
| }, | |
| { | |
| "epoch": 108.0, | |
| "grad_norm": 0.2749279737472534, | |
| "learning_rate": 1.0045638945233266e-06, | |
| "loss": 1.9352, | |
| "step": 159732 | |
| }, | |
| { | |
| "epoch": 108.0, | |
| "eval_loss": 3.850806951522827, | |
| "eval_runtime": 31.5804, | |
| "eval_samples_per_second": 307.628, | |
| "eval_steps_per_second": 6.998, | |
| "step": 159732 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "grad_norm": 0.5212497115135193, | |
| "learning_rate": 5.01464953797611e-09, | |
| "loss": 1.9318, | |
| "step": 177480 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "eval_loss": 3.8198835849761963, | |
| "eval_runtime": 33.3012, | |
| "eval_samples_per_second": 291.731, | |
| "eval_steps_per_second": 6.636, | |
| "step": 177480 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "step": 177480, | |
| "total_flos": 2.803155588736635e+18, | |
| "train_loss": 1.9941566479321613, | |
| "train_runtime": 61251.1099, | |
| "train_samples_per_second": 115.842, | |
| "train_steps_per_second": 2.898 | |
| } | |
| ], | |
| "logging_steps": 17748, | |
| "max_steps": 177480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 120, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.803155588736635e+18, | |
| "train_batch_size": 40, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |