| { | |
| "best_metric": 0.0358840748667717, | |
| "best_model_checkpoint": "man_woman_face_image_detection/checkpoint-9945", | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 9945, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7541478129713424, | |
| "grad_norm": 0.12809374928474426, | |
| "learning_rate": 9.54522486104093e-07, | |
| "loss": 0.0535, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9918687690023333, | |
| "eval_loss": 0.03707238286733627, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 139.741, | |
| "eval_samples_per_second": 101.209, | |
| "eval_steps_per_second": 12.652, | |
| "step": 663 | |
| }, | |
| { | |
| "epoch": 1.5082956259426847, | |
| "grad_norm": 0.07325731962919235, | |
| "learning_rate": 9.039919151086407e-07, | |
| "loss": 0.0514, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9917273562893304, | |
| "eval_loss": 0.037097617983818054, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 145.1641, | |
| "eval_samples_per_second": 97.428, | |
| "eval_steps_per_second": 12.179, | |
| "step": 1326 | |
| }, | |
| { | |
| "epoch": 2.262443438914027, | |
| "grad_norm": 0.12023092061281204, | |
| "learning_rate": 8.534613441131885e-07, | |
| "loss": 0.0536, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9915859435763276, | |
| "eval_loss": 0.03732353821396828, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 142.5995, | |
| "eval_samples_per_second": 99.18, | |
| "eval_steps_per_second": 12.398, | |
| "step": 1989 | |
| }, | |
| { | |
| "epoch": 3.0165912518853695, | |
| "grad_norm": 0.08819910883903503, | |
| "learning_rate": 8.029307731177362e-07, | |
| "loss": 0.0458, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.770739064856712, | |
| "grad_norm": 1.7117729187011719, | |
| "learning_rate": 7.524002021222839e-07, | |
| "loss": 0.0434, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9917980626458318, | |
| "eval_loss": 0.03692413866519928, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 137.5145, | |
| "eval_samples_per_second": 102.847, | |
| "eval_steps_per_second": 12.857, | |
| "step": 2652 | |
| }, | |
| { | |
| "epoch": 4.524886877828054, | |
| "grad_norm": 0.07585655897855759, | |
| "learning_rate": 7.018696311268317e-07, | |
| "loss": 0.0441, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9917980626458318, | |
| "eval_loss": 0.03679042309522629, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 136.8062, | |
| "eval_samples_per_second": 103.38, | |
| "eval_steps_per_second": 12.923, | |
| "step": 3315 | |
| }, | |
| { | |
| "epoch": 5.279034690799397, | |
| "grad_norm": 0.27443283796310425, | |
| "learning_rate": 6.513390601313794e-07, | |
| "loss": 0.0438, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9915859435763276, | |
| "eval_loss": 0.03690684959292412, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 136.7631, | |
| "eval_samples_per_second": 103.412, | |
| "eval_steps_per_second": 12.927, | |
| "step": 3978 | |
| }, | |
| { | |
| "epoch": 6.033182503770739, | |
| "grad_norm": 0.3576681315898895, | |
| "learning_rate": 6.008084891359272e-07, | |
| "loss": 0.0451, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.787330316742081, | |
| "grad_norm": 0.7324354648590088, | |
| "learning_rate": 5.50277918140475e-07, | |
| "loss": 0.0428, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9915859435763276, | |
| "eval_loss": 0.036498043686151505, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 136.3477, | |
| "eval_samples_per_second": 103.727, | |
| "eval_steps_per_second": 12.967, | |
| "step": 4641 | |
| }, | |
| { | |
| "epoch": 7.541478129713424, | |
| "grad_norm": 10.407703399658203, | |
| "learning_rate": 4.997473471450227e-07, | |
| "loss": 0.0377, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9915859435763276, | |
| "eval_loss": 0.03636253997683525, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 136.8932, | |
| "eval_samples_per_second": 103.314, | |
| "eval_steps_per_second": 12.915, | |
| "step": 5304 | |
| }, | |
| { | |
| "epoch": 8.295625942684767, | |
| "grad_norm": 0.776454508304596, | |
| "learning_rate": 4.492167761495705e-07, | |
| "loss": 0.0415, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9917273562893304, | |
| "eval_loss": 0.036238111555576324, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 140.2201, | |
| "eval_samples_per_second": 100.863, | |
| "eval_steps_per_second": 12.609, | |
| "step": 5967 | |
| }, | |
| { | |
| "epoch": 9.049773755656108, | |
| "grad_norm": 0.08934183418750763, | |
| "learning_rate": 3.986862051541182e-07, | |
| "loss": 0.0377, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.803921568627452, | |
| "grad_norm": 0.10604743659496307, | |
| "learning_rate": 3.48155634158666e-07, | |
| "loss": 0.0397, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9920101817153362, | |
| "eval_loss": 0.036092888563871384, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 138.0603, | |
| "eval_samples_per_second": 102.441, | |
| "eval_steps_per_second": 12.806, | |
| "step": 6630 | |
| }, | |
| { | |
| "epoch": 10.558069381598793, | |
| "grad_norm": 0.11123999208211899, | |
| "learning_rate": 2.976250631632137e-07, | |
| "loss": 0.0389, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9918687690023333, | |
| "eval_loss": 0.035989925265312195, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 137.8038, | |
| "eval_samples_per_second": 102.631, | |
| "eval_steps_per_second": 12.83, | |
| "step": 7293 | |
| }, | |
| { | |
| "epoch": 11.312217194570136, | |
| "grad_norm": 2.75396990776062, | |
| "learning_rate": 2.470944921677615e-07, | |
| "loss": 0.0397, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9919394753588348, | |
| "eval_loss": 0.03610728308558464, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 135.6518, | |
| "eval_samples_per_second": 104.26, | |
| "eval_steps_per_second": 13.033, | |
| "step": 7956 | |
| }, | |
| { | |
| "epoch": 12.066365007541478, | |
| "grad_norm": 0.13482791185379028, | |
| "learning_rate": 1.9656392117230926e-07, | |
| "loss": 0.0338, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 12.820512820512821, | |
| "grad_norm": 0.09956225007772446, | |
| "learning_rate": 1.46033350176857e-07, | |
| "loss": 0.0387, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9917980626458318, | |
| "eval_loss": 0.036081310361623764, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 135.9098, | |
| "eval_samples_per_second": 104.062, | |
| "eval_steps_per_second": 13.009, | |
| "step": 8619 | |
| }, | |
| { | |
| "epoch": 13.574660633484163, | |
| "grad_norm": 2.5569167137145996, | |
| "learning_rate": 9.550277918140474e-08, | |
| "loss": 0.0308, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9920101817153362, | |
| "eval_loss": 0.03591620549559593, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 138.98, | |
| "eval_samples_per_second": 101.763, | |
| "eval_steps_per_second": 12.721, | |
| "step": 9282 | |
| }, | |
| { | |
| "epoch": 14.328808446455506, | |
| "grad_norm": 0.17602261900901794, | |
| "learning_rate": 4.49722081859525e-08, | |
| "loss": 0.0355, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9920101817153362, | |
| "eval_loss": 0.0358840748667717, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 140.3453, | |
| "eval_samples_per_second": 100.773, | |
| "eval_steps_per_second": 12.597, | |
| "step": 9945 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9945, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 2.46575636354151e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |