| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.015517408592765009, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007758704296382504, | |
| "grad_norm": 23.366365432739258, | |
| "learning_rate": 1e-05, | |
| "loss": 22.6895, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007758704296382504, | |
| "eval_loss": NaN, | |
| "eval_runtime": 111.5166, | |
| "eval_samples_per_second": 19.468, | |
| "eval_steps_per_second": 9.738, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0015517408592765009, | |
| "grad_norm": 23.651832580566406, | |
| "learning_rate": 2e-05, | |
| "loss": 21.1084, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0023276112889147513, | |
| "grad_norm": 21.718303680419922, | |
| "learning_rate": 3e-05, | |
| "loss": 21.801, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0031034817185530018, | |
| "grad_norm": 21.868783950805664, | |
| "learning_rate": 4e-05, | |
| "loss": 22.9325, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003879352148191252, | |
| "grad_norm": 20.122451782226562, | |
| "learning_rate": 5e-05, | |
| "loss": 21.3532, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.003879352148191252, | |
| "eval_loss": NaN, | |
| "eval_runtime": 96.7454, | |
| "eval_samples_per_second": 22.44, | |
| "eval_steps_per_second": 11.225, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004655222577829503, | |
| "grad_norm": 24.650022506713867, | |
| "learning_rate": 6e-05, | |
| "loss": 21.8922, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005431093007467753, | |
| "grad_norm": 30.706462860107422, | |
| "learning_rate": 7e-05, | |
| "loss": 21.2497, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0062069634371060035, | |
| "grad_norm": 26.0838565826416, | |
| "learning_rate": 8e-05, | |
| "loss": 19.2692, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0069828338667442535, | |
| "grad_norm": 21.70370101928711, | |
| "learning_rate": 9e-05, | |
| "loss": 20.4647, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007758704296382504, | |
| "grad_norm": 20.149873733520508, | |
| "learning_rate": 0.0001, | |
| "loss": 19.2082, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007758704296382504, | |
| "eval_loss": NaN, | |
| "eval_runtime": 55.6653, | |
| "eval_samples_per_second": 39.001, | |
| "eval_steps_per_second": 19.509, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008534574726020755, | |
| "grad_norm": 25.1776180267334, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 21.373, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009310445155659005, | |
| "grad_norm": 27.650684356689453, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 18.2556, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010086315585297255, | |
| "grad_norm": 21.88231658935547, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 17.9876, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.010862186014935505, | |
| "grad_norm": 18.6796817779541, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 17.8176, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.011638056444573755, | |
| "grad_norm": 17.174560546875, | |
| "learning_rate": 5e-05, | |
| "loss": 16.9347, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011638056444573755, | |
| "eval_loss": NaN, | |
| "eval_runtime": 46.4331, | |
| "eval_samples_per_second": 46.755, | |
| "eval_steps_per_second": 23.388, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012413926874212007, | |
| "grad_norm": 17.368879318237305, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 16.599, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013189797303850257, | |
| "grad_norm": 16.477272033691406, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 18.5643, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.013965667733488507, | |
| "grad_norm": 16.044984817504883, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 15.8571, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014741538163126757, | |
| "grad_norm": 16.53944969177246, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 16.9206, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.015517408592765009, | |
| "grad_norm": 15.834726333618164, | |
| "learning_rate": 0.0, | |
| "loss": 16.8041, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.015517408592765009, | |
| "eval_loss": NaN, | |
| "eval_runtime": 46.4644, | |
| "eval_samples_per_second": 46.724, | |
| "eval_steps_per_second": 23.373, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.402135828758528e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |