| { | |
| "best_metric": 1.3211654424667358, | |
| "best_model_checkpoint": "facial_age_image_detection/checkpoint-6400", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 6400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5492572322126661, | |
| "eval_loss": 1.3843398094177246, | |
| "eval_runtime": 53.0532, | |
| "eval_samples_per_second": 96.431, | |
| "eval_steps_per_second": 3.016, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 9.20909595489502, | |
| "learning_rate": 9.291338582677165e-07, | |
| "loss": 1.0144, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5471071149335418, | |
| "eval_loss": 1.3768093585968018, | |
| "eval_runtime": 52.7424, | |
| "eval_samples_per_second": 97.0, | |
| "eval_steps_per_second": 3.034, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5433932759968726, | |
| "eval_loss": 1.3764420747756958, | |
| "eval_runtime": 51.5402, | |
| "eval_samples_per_second": 99.262, | |
| "eval_steps_per_second": 3.104, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 8.877883911132812, | |
| "learning_rate": 8.503937007874016e-07, | |
| "loss": 1.0053, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.547302580140735, | |
| "eval_loss": 1.366408348083496, | |
| "eval_runtime": 52.1841, | |
| "eval_samples_per_second": 98.038, | |
| "eval_steps_per_second": 3.066, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "grad_norm": 12.386143684387207, | |
| "learning_rate": 7.716535433070865e-07, | |
| "loss": 0.9744, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5517982799061767, | |
| "eval_loss": 1.3610535860061646, | |
| "eval_runtime": 52.1993, | |
| "eval_samples_per_second": 98.009, | |
| "eval_steps_per_second": 3.065, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5521892103205629, | |
| "eval_loss": 1.3569880723953247, | |
| "eval_runtime": 51.8422, | |
| "eval_samples_per_second": 98.684, | |
| "eval_steps_per_second": 3.086, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 12.370375633239746, | |
| "learning_rate": 6.929133858267716e-07, | |
| "loss": 0.9667, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5494526974198592, | |
| "eval_loss": 1.3512985706329346, | |
| "eval_runtime": 52.2761, | |
| "eval_samples_per_second": 97.865, | |
| "eval_steps_per_second": 3.061, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "grad_norm": 11.062934875488281, | |
| "learning_rate": 6.141732283464567e-07, | |
| "loss": 0.9536, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5492572322126661, | |
| "eval_loss": 1.3439780473709106, | |
| "eval_runtime": 51.9554, | |
| "eval_samples_per_second": 98.469, | |
| "eval_steps_per_second": 3.08, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5484753713838937, | |
| "eval_loss": 1.3428295850753784, | |
| "eval_runtime": 52.082, | |
| "eval_samples_per_second": 98.23, | |
| "eval_steps_per_second": 3.072, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "grad_norm": 12.75069808959961, | |
| "learning_rate": 5.354330708661418e-07, | |
| "loss": 0.9437, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5484753713838937, | |
| "eval_loss": 1.3394215106964111, | |
| "eval_runtime": 51.1719, | |
| "eval_samples_per_second": 99.977, | |
| "eval_steps_per_second": 3.127, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.94, | |
| "grad_norm": 11.664592742919922, | |
| "learning_rate": 4.5669291338582676e-07, | |
| "loss": 0.9279, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5471071149335418, | |
| "eval_loss": 1.3347781896591187, | |
| "eval_runtime": 51.4976, | |
| "eval_samples_per_second": 99.344, | |
| "eval_steps_per_second": 3.107, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5541438623924941, | |
| "eval_loss": 1.3312886953353882, | |
| "eval_runtime": 51.2447, | |
| "eval_samples_per_second": 99.835, | |
| "eval_steps_per_second": 3.122, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 8.319609642028809, | |
| "learning_rate": 3.779527559055118e-07, | |
| "loss": 0.921, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5547302580140735, | |
| "eval_loss": 1.3314929008483887, | |
| "eval_runtime": 51.5959, | |
| "eval_samples_per_second": 99.155, | |
| "eval_steps_per_second": 3.101, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.5508209538702111, | |
| "eval_loss": 1.3308985233306885, | |
| "eval_runtime": 51.327, | |
| "eval_samples_per_second": 99.675, | |
| "eval_steps_per_second": 3.117, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "grad_norm": 6.9851226806640625, | |
| "learning_rate": 2.9921259842519687e-07, | |
| "loss": 0.9223, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.5564894448788116, | |
| "eval_loss": 1.3276044130325317, | |
| "eval_runtime": 51.4978, | |
| "eval_samples_per_second": 99.344, | |
| "eval_steps_per_second": 3.107, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "grad_norm": 6.1127119064331055, | |
| "learning_rate": 2.2047244094488187e-07, | |
| "loss": 0.9054, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5545347928068803, | |
| "eval_loss": 1.3215641975402832, | |
| "eval_runtime": 51.1737, | |
| "eval_samples_per_second": 99.973, | |
| "eval_steps_per_second": 3.127, | |
| "step": 5120 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.5568803752931978, | |
| "eval_loss": 1.324193000793457, | |
| "eval_runtime": 51.5617, | |
| "eval_samples_per_second": 99.221, | |
| "eval_steps_per_second": 3.103, | |
| "step": 5440 | |
| }, | |
| { | |
| "epoch": 17.19, | |
| "grad_norm": 7.823403358459473, | |
| "learning_rate": 1.4173228346456693e-07, | |
| "loss": 0.9092, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.5551211884284597, | |
| "eval_loss": 1.3241610527038574, | |
| "eval_runtime": 52.1852, | |
| "eval_samples_per_second": 98.035, | |
| "eval_steps_per_second": 3.066, | |
| "step": 5760 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "grad_norm": 8.171327590942383, | |
| "learning_rate": 6.299212598425197e-08, | |
| "loss": 0.9017, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.5576622361219703, | |
| "eval_loss": 1.3216114044189453, | |
| "eval_runtime": 51.9014, | |
| "eval_samples_per_second": 98.572, | |
| "eval_steps_per_second": 3.083, | |
| "step": 6080 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5586395621579359, | |
| "eval_loss": 1.3211654424667358, | |
| "eval_runtime": 51.4519, | |
| "eval_samples_per_second": 99.433, | |
| "eval_steps_per_second": 3.11, | |
| "step": 6400 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 3.171569043869614e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |