| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 200.0, | |
| "eval_steps": 780, | |
| "global_step": 7800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 15.68354606628418, | |
| "learning_rate": 9e-06, | |
| "loss": 2.479, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 2.5237555503845215, | |
| "eval_runtime": 9.7181, | |
| "eval_samples_per_second": 256.223, | |
| "eval_steps_per_second": 2.573, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 7.548585891723633, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.5415, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 1.9512752294540405, | |
| "eval_runtime": 9.4805, | |
| "eval_samples_per_second": 262.644, | |
| "eval_steps_per_second": 2.637, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 4.004195213317871, | |
| "learning_rate": 7e-06, | |
| "loss": 0.1937, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 1.675230622291565, | |
| "eval_runtime": 9.4427, | |
| "eval_samples_per_second": 263.695, | |
| "eval_steps_per_second": 2.648, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.42832350730896, | |
| "learning_rate": 6e-06, | |
| "loss": 0.1072, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_loss": 1.5575916767120361, | |
| "eval_runtime": 9.4213, | |
| "eval_samples_per_second": 264.296, | |
| "eval_steps_per_second": 2.654, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 4.648789405822754, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0722, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 1.487786054611206, | |
| "eval_runtime": 9.4183, | |
| "eval_samples_per_second": 264.378, | |
| "eval_steps_per_second": 2.654, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "grad_norm": 1.3144129514694214, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0542, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "eval_loss": 1.418711543083191, | |
| "eval_runtime": 9.7326, | |
| "eval_samples_per_second": 255.84, | |
| "eval_steps_per_second": 2.569, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 140.0, | |
| "grad_norm": 1.629385232925415, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0433, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 140.0, | |
| "eval_loss": 1.3937839269638062, | |
| "eval_runtime": 9.4439, | |
| "eval_samples_per_second": 263.662, | |
| "eval_steps_per_second": 2.647, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 160.0, | |
| "grad_norm": 1.3994438648223877, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.0376, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 160.0, | |
| "eval_loss": 1.3543996810913086, | |
| "eval_runtime": 9.4766, | |
| "eval_samples_per_second": 262.751, | |
| "eval_steps_per_second": 2.638, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 180.0, | |
| "grad_norm": 1.0756237506866455, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.0333, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 180.0, | |
| "eval_loss": 1.3324673175811768, | |
| "eval_runtime": 9.5119, | |
| "eval_samples_per_second": 261.777, | |
| "eval_steps_per_second": 2.628, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "grad_norm": 0.9140580892562866, | |
| "learning_rate": 0.0, | |
| "loss": 0.0311, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "eval_loss": 1.3237212896347046, | |
| "eval_runtime": 9.4689, | |
| "eval_samples_per_second": 262.967, | |
| "eval_steps_per_second": 2.64, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "step": 7800, | |
| "total_flos": 1.3175247232512e+17, | |
| "train_loss": 0.35931170732547074, | |
| "train_runtime": 4504.8295, | |
| "train_samples_per_second": 110.548, | |
| "train_steps_per_second": 1.731 | |
| } | |
| ], | |
| "logging_steps": 780, | |
| "max_steps": 7800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 200, | |
| "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": 1.3175247232512e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |