| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "global_step": 1185, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9e-07, | |
| "loss": 2.5593, | |
| "perplexity": 12.926765411958069, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9e-07, | |
| "loss": 2.5731, | |
| "perplexity": 13.1063913429122, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.03165881737310309, | |
| "eval_loss": 2.611328125, | |
| "eval_perplexity": 13.617124090346826, | |
| "eval_runtime": 4.8498, | |
| "eval_samples_per_second": 16.083, | |
| "eval_steps_per_second": 0.619, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9e-07, | |
| "loss": 2.206, | |
| "perplexity": 9.079326356062138, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.032757718472004184, | |
| "eval_loss": 2.48046875, | |
| "eval_perplexity": 11.946863197656471, | |
| "eval_runtime": 5.428, | |
| "eval_samples_per_second": 14.37, | |
| "eval_steps_per_second": 0.553, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9e-07, | |
| "loss": 1.9105, | |
| "perplexity": 6.7564661872073595, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.03333333333333333, | |
| "eval_loss": 2.451171875, | |
| "eval_perplexity": 11.601934773230344, | |
| "eval_runtime": 5.5551, | |
| "eval_samples_per_second": 14.041, | |
| "eval_steps_per_second": 0.54, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9e-07, | |
| "loss": 1.6301, | |
| "perplexity": 5.1043851315288045, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.03453689167974882, | |
| "eval_loss": 2.5078125, | |
| "eval_perplexity": 12.278042445054236, | |
| "eval_runtime": 5.3439, | |
| "eval_samples_per_second": 14.596, | |
| "eval_steps_per_second": 0.561, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 9e-07, | |
| "loss": 1.3733, | |
| "perplexity": 3.948358804340304, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.034170591313448455, | |
| "eval_loss": 2.681640625, | |
| "eval_perplexity": 14.609041604294383, | |
| "eval_runtime": 5.3493, | |
| "eval_samples_per_second": 14.581, | |
| "eval_steps_per_second": 0.561, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 9e-07, | |
| "loss": 1.1337, | |
| "perplexity": 3.107131645102156, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.033019361590790164, | |
| "eval_loss": 3.0078125, | |
| "eval_perplexity": 20.24306974171993, | |
| "eval_runtime": 5.4602, | |
| "eval_samples_per_second": 14.285, | |
| "eval_steps_per_second": 0.549, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.9619, | |
| "perplexity": 2.616663413821779, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.033019361590790164, | |
| "eval_loss": 3.177734375, | |
| "eval_perplexity": 23.992334296139543, | |
| "eval_runtime": 5.3719, | |
| "eval_samples_per_second": 14.52, | |
| "eval_steps_per_second": 0.558, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.798, | |
| "perplexity": 2.221094294751435, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.033019361590790164, | |
| "eval_loss": 3.255859375, | |
| "eval_perplexity": 25.94189877662824, | |
| "eval_runtime": 5.4474, | |
| "eval_samples_per_second": 14.319, | |
| "eval_steps_per_second": 0.551, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.6653, | |
| "perplexity": 1.9450739560040564, | |
| "step": 711 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.033124018838304556, | |
| "eval_loss": 3.427734375, | |
| "eval_perplexity": 30.806767041912167, | |
| "eval_runtime": 5.5668, | |
| "eval_samples_per_second": 14.012, | |
| "eval_steps_per_second": 0.539, | |
| "step": 711 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.552, | |
| "perplexity": 1.736722992721326, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.03333333333333333, | |
| "eval_loss": 3.556640625, | |
| "eval_perplexity": 35.04526897307925, | |
| "eval_runtime": 5.4469, | |
| "eval_samples_per_second": 14.32, | |
| "eval_steps_per_second": 0.551, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.4568, | |
| "perplexity": 1.579013050057063, | |
| "step": 869 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.032443746729461015, | |
| "eval_loss": 3.732421875, | |
| "eval_perplexity": 41.780172086809024, | |
| "eval_runtime": 5.3482, | |
| "eval_samples_per_second": 14.584, | |
| "eval_steps_per_second": 0.561, | |
| "step": 869 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.3756, | |
| "perplexity": 1.4558646714178145, | |
| "step": 948 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.032757718472004184, | |
| "eval_loss": 3.818359375, | |
| "eval_perplexity": 45.529450258314064, | |
| "eval_runtime": 5.6642, | |
| "eval_samples_per_second": 13.771, | |
| "eval_steps_per_second": 0.53, | |
| "step": 948 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.3119, | |
| "perplexity": 1.3660180843907228, | |
| "step": 1027 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.033124018838304556, | |
| "eval_loss": 3.84765625, | |
| "eval_perplexity": 46.88305220915621, | |
| "eval_runtime": 5.4663, | |
| "eval_samples_per_second": 14.269, | |
| "eval_steps_per_second": 0.549, | |
| "step": 1027 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.2448, | |
| "perplexity": 1.2773658144929685, | |
| "step": 1106 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.03291470434327577, | |
| "eval_loss": 3.90625, | |
| "eval_perplexity": 49.71218131735948, | |
| "eval_runtime": 5.3413, | |
| "eval_samples_per_second": 14.603, | |
| "eval_steps_per_second": 0.562, | |
| "step": 1106 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.1986, | |
| "perplexity": 1.219693990715056, | |
| "step": 1185 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.032862375719518576, | |
| "eval_loss": 3.943359375, | |
| "eval_perplexity": 51.591626129191596, | |
| "eval_runtime": 5.3588, | |
| "eval_samples_per_second": 14.555, | |
| "eval_steps_per_second": 0.56, | |
| "step": 1185 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 1185, | |
| "total_flos": 39797194489856.0, | |
| "train_loss": 1.0260957291357629, | |
| "train_runtime": 68855.6328, | |
| "train_samples_per_second": 0.068, | |
| "train_steps_per_second": 0.017 | |
| } | |
| ], | |
| "max_steps": 1185, | |
| "num_train_epochs": 15, | |
| "total_flos": 39797194489856.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |