| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.992, | |
| "eval_steps": 5, | |
| "global_step": 31, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 0.06547047197818756, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0571, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 0.06732479482889175, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0552, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 0.061726897954940796, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0524, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 0.05993999168276787, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0527, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.04297811910510063, | |
| "learning_rate": 9.966191788709716e-05, | |
| "loss": 0.0452, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.04343939200043678, | |
| "eval_runtime": 1.3959, | |
| "eval_samples_per_second": 5.731, | |
| "eval_steps_per_second": 1.433, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 0.03668003901839256, | |
| "learning_rate": 9.865224352899119e-05, | |
| "loss": 0.0387, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": 0.04254748299717903, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0373, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 0.03773832321166992, | |
| "learning_rate": 9.468163201617062e-05, | |
| "loss": 0.0353, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 0.03248106688261032, | |
| "learning_rate": 9.177439057064683e-05, | |
| "loss": 0.0379, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.029653361067175865, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0349, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.03697124496102333, | |
| "eval_runtime": 1.3747, | |
| "eval_samples_per_second": 5.82, | |
| "eval_steps_per_second": 1.455, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.352, | |
| "grad_norm": 0.02675388567149639, | |
| "learning_rate": 8.43120818934367e-05, | |
| "loss": 0.0324, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 0.02256765589118004, | |
| "learning_rate": 7.985792958513931e-05, | |
| "loss": 0.0334, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.416, | |
| "grad_norm": 0.021855270490050316, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0313, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 0.019800638779997826, | |
| "learning_rate": 6.980398830195785e-05, | |
| "loss": 0.0262, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.020864585414528847, | |
| "learning_rate": 6.434016163555452e-05, | |
| "loss": 0.031, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.0329812727868557, | |
| "eval_runtime": 1.3771, | |
| "eval_samples_per_second": 5.809, | |
| "eval_steps_per_second": 1.452, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 0.019576789811253548, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0305, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.544, | |
| "grad_norm": 0.020052531734108925, | |
| "learning_rate": 5.290724144552379e-05, | |
| "loss": 0.0283, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 0.019002581015229225, | |
| "learning_rate": 4.709275855447621e-05, | |
| "loss": 0.0293, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.608, | |
| "grad_norm": 0.02045363001525402, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0265, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.018427610397338867, | |
| "learning_rate": 3.5659838364445505e-05, | |
| "loss": 0.0258, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.031122902408242226, | |
| "eval_runtime": 1.38, | |
| "eval_samples_per_second": 5.797, | |
| "eval_steps_per_second": 1.449, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 0.01990111917257309, | |
| "learning_rate": 3.019601169804216e-05, | |
| "loss": 0.0304, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.704, | |
| "grad_norm": 0.019229382276535034, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0285, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.736, | |
| "grad_norm": 0.0193803608417511, | |
| "learning_rate": 2.0142070414860704e-05, | |
| "loss": 0.0306, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 0.01592605747282505, | |
| "learning_rate": 1.5687918106563326e-05, | |
| "loss": 0.027, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.01785469613969326, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0277, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.03004794754087925, | |
| "eval_runtime": 1.3771, | |
| "eval_samples_per_second": 5.81, | |
| "eval_steps_per_second": 1.452, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.832, | |
| "grad_norm": 0.0174510907381773, | |
| "learning_rate": 8.225609429353187e-06, | |
| "loss": 0.0287, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 0.016350215300917625, | |
| "learning_rate": 5.318367983829392e-06, | |
| "loss": 0.0275, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 0.015370819717645645, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0238, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.928, | |
| "grad_norm": 0.016408780589699745, | |
| "learning_rate": 1.3477564710088098e-06, | |
| "loss": 0.0283, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.016846995800733566, | |
| "learning_rate": 3.380821129028489e-07, | |
| "loss": 0.0261, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.029837772250175476, | |
| "eval_runtime": 1.3754, | |
| "eval_samples_per_second": 5.817, | |
| "eval_steps_per_second": 1.454, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "grad_norm": 0.017581390216946602, | |
| "learning_rate": 0.0, | |
| "loss": 0.0283, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "step": 31, | |
| "total_flos": 5.244736447827149e+16, | |
| "train_loss": 0.033818339808813984, | |
| "train_runtime": 413.5784, | |
| "train_samples_per_second": 2.418, | |
| "train_steps_per_second": 0.075 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 31, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.244736447827149e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |