| { | |
| "best_metric": 1.2208830118179321, | |
| "best_model_checkpoint": "facial_age_image_detection/checkpoint-3120", | |
| "epoch": 13.0, | |
| "eval_steps": 500, | |
| "global_step": 3120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5914377871175838, | |
| "eval_loss": 1.247507929801941, | |
| "eval_runtime": 105.4075, | |
| "eval_samples_per_second": 97.061, | |
| "eval_steps_per_second": 3.036, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.599159417456749, | |
| "eval_loss": 1.2387152910232544, | |
| "eval_runtime": 105.6722, | |
| "eval_samples_per_second": 96.818, | |
| "eval_steps_per_second": 3.028, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 16.16035270690918, | |
| "learning_rate": 6.112676056338028e-07, | |
| "loss": 0.8178, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.593979083178575, | |
| "eval_loss": 1.240220069885254, | |
| "eval_runtime": 106.9319, | |
| "eval_samples_per_second": 95.678, | |
| "eval_steps_per_second": 2.993, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5982797380510214, | |
| "eval_loss": 1.2350324392318726, | |
| "eval_runtime": 105.7769, | |
| "eval_samples_per_second": 96.722, | |
| "eval_steps_per_second": 3.025, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 12.192686080932617, | |
| "learning_rate": 5.126760563380281e-07, | |
| "loss": 0.8038, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5962271527709901, | |
| "eval_loss": 1.2324706315994263, | |
| "eval_runtime": 105.7457, | |
| "eval_samples_per_second": 96.751, | |
| "eval_steps_per_second": 3.026, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5964226370833741, | |
| "eval_loss": 1.2318035364151, | |
| "eval_runtime": 105.5068, | |
| "eval_samples_per_second": 96.97, | |
| "eval_steps_per_second": 3.033, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 12.383950233459473, | |
| "learning_rate": 4.140845070422535e-07, | |
| "loss": 0.7976, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5976932851138697, | |
| "eval_loss": 1.2295589447021484, | |
| "eval_runtime": 106.676, | |
| "eval_samples_per_second": 95.907, | |
| "eval_steps_per_second": 3.0, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5974978008014857, | |
| "eval_loss": 1.2282449007034302, | |
| "eval_runtime": 105.9543, | |
| "eval_samples_per_second": 96.561, | |
| "eval_steps_per_second": 3.02, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "grad_norm": 12.568708419799805, | |
| "learning_rate": 3.1549295774647887e-07, | |
| "loss": 0.7888, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6013097448929723, | |
| "eval_loss": 1.2256407737731934, | |
| "eval_runtime": 105.7499, | |
| "eval_samples_per_second": 96.747, | |
| "eval_steps_per_second": 3.026, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6001368390186688, | |
| "eval_loss": 1.2228951454162598, | |
| "eval_runtime": 106.2182, | |
| "eval_samples_per_second": 96.321, | |
| "eval_steps_per_second": 3.013, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "grad_norm": 10.32869815826416, | |
| "learning_rate": 2.169014084507042e-07, | |
| "loss": 0.7785, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5989639331443651, | |
| "eval_loss": 1.223892331123352, | |
| "eval_runtime": 105.6089, | |
| "eval_samples_per_second": 96.876, | |
| "eval_steps_per_second": 3.03, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.599452643925325, | |
| "eval_loss": 1.2233675718307495, | |
| "eval_runtime": 105.2752, | |
| "eval_samples_per_second": 97.183, | |
| "eval_steps_per_second": 3.04, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 10.018258094787598, | |
| "learning_rate": 1.1830985915492958e-07, | |
| "loss": 0.7763, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5997458703939009, | |
| "eval_loss": 1.2208830118179321, | |
| "eval_runtime": 106.4062, | |
| "eval_samples_per_second": 96.15, | |
| "eval_steps_per_second": 3.007, | |
| "step": 3120 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "total_flos": 1.5461399088864369e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |