| { | |
| "best_metric": 0.45813843607902527, | |
| "best_model_checkpoint": "akram-convNeXt-flipkarts/checkpoint-192", | |
| "epoch": 4.0, | |
| "eval_steps": 32, | |
| "global_step": 224, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 5.786337852478027, | |
| "learning_rate": 0.00019107142857142858, | |
| "loss": 0.4817, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 3.83489727973938, | |
| "learning_rate": 0.00018214285714285714, | |
| "loss": 0.3841, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 5.209750652313232, | |
| "learning_rate": 0.00017321428571428574, | |
| "loss": 0.3777, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.5408539772033691, | |
| "eval_runtime": 43.884, | |
| "eval_samples_per_second": 3.6, | |
| "eval_steps_per_second": 0.456, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 7.569049835205078, | |
| "learning_rate": 0.00016428571428571428, | |
| "loss": 0.4523, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 3.3220114707946777, | |
| "learning_rate": 0.00015535714285714287, | |
| "loss": 0.2812, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 2.3158109188079834, | |
| "learning_rate": 0.00014642857142857141, | |
| "loss": 0.131, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_accuracy": 0.8417721518987342, | |
| "eval_loss": 0.5632613897323608, | |
| "eval_runtime": 39.0388, | |
| "eval_samples_per_second": 4.047, | |
| "eval_steps_per_second": 0.512, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.282313585281372, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.177, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 0.194750115275383, | |
| "learning_rate": 0.00012857142857142858, | |
| "loss": 0.1291, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 4.554263591766357, | |
| "learning_rate": 0.00011964285714285714, | |
| "loss": 0.1426, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_accuracy": 0.8291139240506329, | |
| "eval_loss": 0.5170620083808899, | |
| "eval_runtime": 38.7538, | |
| "eval_samples_per_second": 4.077, | |
| "eval_steps_per_second": 0.516, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 2.037961721420288, | |
| "learning_rate": 0.00011071428571428572, | |
| "loss": 0.2537, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 3.5785367488861084, | |
| "learning_rate": 0.00010178571428571428, | |
| "loss": 0.3462, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 3.6267411708831787, | |
| "learning_rate": 9.285714285714286e-05, | |
| "loss": 0.2413, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_accuracy": 0.8607594936708861, | |
| "eval_loss": 0.47749054431915283, | |
| "eval_runtime": 38.2656, | |
| "eval_samples_per_second": 4.129, | |
| "eval_steps_per_second": 0.523, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 18.910554885864258, | |
| "learning_rate": 8.392857142857144e-05, | |
| "loss": 0.2646, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 6.847780227661133, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.2463, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 5.892199993133545, | |
| "learning_rate": 6.607142857142857e-05, | |
| "loss": 0.3724, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 0.5983772277832031, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 0.1852, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_accuracy": 0.8607594936708861, | |
| "eval_loss": 0.4647924304008484, | |
| "eval_runtime": 37.8434, | |
| "eval_samples_per_second": 4.175, | |
| "eval_steps_per_second": 0.528, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 5.2921462059021, | |
| "learning_rate": 4.8214285714285716e-05, | |
| "loss": 0.3187, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "grad_norm": 6.326498985290527, | |
| "learning_rate": 3.928571428571429e-05, | |
| "loss": 0.1574, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 8.077910423278809, | |
| "learning_rate": 3.0357142857142857e-05, | |
| "loss": 0.2165, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "eval_accuracy": 0.8734177215189873, | |
| "eval_loss": 0.45813843607902527, | |
| "eval_runtime": 39.2858, | |
| "eval_samples_per_second": 4.022, | |
| "eval_steps_per_second": 0.509, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "grad_norm": 1.3805155754089355, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 0.1937, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.4734901785850525, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.134, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "grad_norm": 4.029452323913574, | |
| "learning_rate": 3.5714285714285714e-06, | |
| "loss": 0.1529, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8734177215189873, | |
| "eval_loss": 0.4617495834827423, | |
| "eval_runtime": 40.3868, | |
| "eval_samples_per_second": 3.912, | |
| "eval_steps_per_second": 0.495, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 224, | |
| "total_flos": 8.966780447204966e+16, | |
| "train_loss": 0.2551893559949739, | |
| "train_runtime": 4113.2929, | |
| "train_samples_per_second": 0.867, | |
| "train_steps_per_second": 0.054 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 224, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 32, | |
| "total_flos": 8.966780447204966e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |