| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9200410630774494, | |
| "eval_steps": 300, | |
| "global_step": 2400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 2.97e-05, | |
| "loss": 11.4962, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_cer": 0.7285700557370748, | |
| "eval_loss": 3.9011037349700928, | |
| "eval_runtime": 45.9839, | |
| "eval_samples_per_second": 48.3, | |
| "eval_steps_per_second": 6.046, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.752929190015283e-05, | |
| "loss": 1.9358, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_cer": 0.3138093407649433, | |
| "eval_loss": 1.294021487236023, | |
| "eval_runtime": 39.5216, | |
| "eval_samples_per_second": 56.197, | |
| "eval_steps_per_second": 7.034, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.9938869077941925e-05, | |
| "loss": 1.1506, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_cer": 0.33158754564674225, | |
| "eval_loss": 1.1284617185592651, | |
| "eval_runtime": 39.8848, | |
| "eval_samples_per_second": 55.685, | |
| "eval_steps_per_second": 6.97, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 3.234844625573103e-05, | |
| "loss": 0.9373, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_cer": 0.27407265039400347, | |
| "eval_loss": 0.9184823632240295, | |
| "eval_runtime": 39.7128, | |
| "eval_samples_per_second": 55.927, | |
| "eval_steps_per_second": 7.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 2.470708099847173e-05, | |
| "loss": 0.8532, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_cer": 0.28651739381126273, | |
| "eval_loss": 0.8993165493011475, | |
| "eval_runtime": 39.7812, | |
| "eval_samples_per_second": 55.83, | |
| "eval_steps_per_second": 6.988, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.7065715741212433e-05, | |
| "loss": 0.7554, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_cer": 0.2535075917739765, | |
| "eval_loss": 0.8039466738700867, | |
| "eval_runtime": 39.8849, | |
| "eval_samples_per_second": 55.685, | |
| "eval_steps_per_second": 6.97, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 9.424350483953132e-06, | |
| "loss": 0.6747, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_cer": 0.24255237363059773, | |
| "eval_loss": 0.7663877606391907, | |
| "eval_runtime": 39.7233, | |
| "eval_samples_per_second": 55.912, | |
| "eval_steps_per_second": 6.998, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.7829852266938361e-06, | |
| "loss": 0.6314, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_cer": 0.24130309436863348, | |
| "eval_loss": 0.7546484470367432, | |
| "eval_runtime": 39.7215, | |
| "eval_samples_per_second": 55.914, | |
| "eval_steps_per_second": 6.999, | |
| "step": 2400 | |
| } | |
| ], | |
| "logging_steps": 300, | |
| "max_steps": 2463, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 600, | |
| "total_flos": 3.771491125288823e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |