| { | |
| "best_metric": 3.302131175994873, | |
| "best_model_checkpoint": "car_brands_image_detection/checkpoint-13060", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 13060, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 7.455317497253418, | |
| "learning_rate": 1.9308224442736356e-06, | |
| "loss": 3.4846, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 4.010385036468506, | |
| "learning_rate": 1.8539584934665642e-06, | |
| "loss": 3.4587, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 4.901058673858643, | |
| "learning_rate": 1.7770945426594927e-06, | |
| "loss": 3.45, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 5.844404220581055, | |
| "learning_rate": 1.700230591852421e-06, | |
| "loss": 3.4339, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 5.998371124267578, | |
| "learning_rate": 1.6233666410453497e-06, | |
| "loss": 3.414, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 5.008224010467529, | |
| "learning_rate": 1.5465026902382784e-06, | |
| "loss": 3.3956, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.541320237223395, | |
| "eval_loss": 3.4782135486602783, | |
| "eval_runtime": 711.0751, | |
| "eval_samples_per_second": 97.935, | |
| "eval_steps_per_second": 12.242, | |
| "step": 3265 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 4.298598766326904, | |
| "learning_rate": 1.4696387394312066e-06, | |
| "loss": 3.3699, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 4.588273525238037, | |
| "learning_rate": 1.3927747886241352e-06, | |
| "loss": 3.3439, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 5.975885391235352, | |
| "learning_rate": 1.3159108378170639e-06, | |
| "loss": 3.3162, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 4.5131096839904785, | |
| "learning_rate": 1.239046887009992e-06, | |
| "loss": 3.306, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 4.919468879699707, | |
| "learning_rate": 1.1621829362029207e-06, | |
| "loss": 3.2967, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 5.679530143737793, | |
| "learning_rate": 1.0853189853958494e-06, | |
| "loss": 3.273, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 5.218754768371582, | |
| "learning_rate": 1.0084550345887778e-06, | |
| "loss": 3.2877, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.552879851807177, | |
| "eval_loss": 3.3825206756591797, | |
| "eval_runtime": 1330.9633, | |
| "eval_samples_per_second": 52.322, | |
| "eval_steps_per_second": 6.54, | |
| "step": 6530 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 5.154468059539795, | |
| "learning_rate": 9.315910837817063e-07, | |
| "loss": 3.222, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 4.286391735076904, | |
| "learning_rate": 8.547271329746349e-07, | |
| "loss": 3.2313, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 5.875144004821777, | |
| "learning_rate": 7.778631821675634e-07, | |
| "loss": 3.2207, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 5.433847904205322, | |
| "learning_rate": 7.009992313604919e-07, | |
| "loss": 3.2325, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 4.9686102867126465, | |
| "learning_rate": 6.241352805534205e-07, | |
| "loss": 3.2162, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 4.758098125457764, | |
| "learning_rate": 5.472713297463489e-07, | |
| "loss": 3.2061, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5644394663909591, | |
| "eval_loss": 3.3224754333496094, | |
| "eval_runtime": 709.8901, | |
| "eval_samples_per_second": 98.098, | |
| "eval_steps_per_second": 12.262, | |
| "step": 9795 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 5.5253729820251465, | |
| "learning_rate": 4.7040737893927744e-07, | |
| "loss": 3.1879, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 5.2010908126831055, | |
| "learning_rate": 3.9354342813220597e-07, | |
| "loss": 3.1831, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "grad_norm": 5.191558837890625, | |
| "learning_rate": 3.166794773251345e-07, | |
| "loss": 3.1715, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 5.484861850738525, | |
| "learning_rate": 2.3981552651806304e-07, | |
| "loss": 3.1752, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 4.993890762329102, | |
| "learning_rate": 1.6295157571099155e-07, | |
| "loss": 3.1738, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 4.897230625152588, | |
| "learning_rate": 8.608762490392006e-08, | |
| "loss": 3.171, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "grad_norm": 5.773993968963623, | |
| "learning_rate": 9.223674096848577e-09, | |
| "loss": 3.1716, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5659328824365657, | |
| "eval_loss": 3.302131175994873, | |
| "eval_runtime": 705.9576, | |
| "eval_samples_per_second": 98.645, | |
| "eval_steps_per_second": 12.331, | |
| "step": 13060 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 13060, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 3.247178702686662e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |