| { | |
| "best_metric": 0.9080459770114943, | |
| "best_model_checkpoint": "dinov2-small-imagenet1k-1-layer-finetuned-noh\\checkpoint-46", | |
| "epoch": 9.577777777777778, | |
| "eval_steps": 500, | |
| "global_step": 220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 467.71881103515625, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.7354, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 400.1885986328125, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.6331, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7651888341543513, | |
| "eval_loss": 0.5416210889816284, | |
| "eval_runtime": 69.3532, | |
| "eval_samples_per_second": 8.781, | |
| "eval_steps_per_second": 0.562, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.3111111111111111, | |
| "grad_norm": 38.50715255737305, | |
| "learning_rate": 4.797979797979798e-05, | |
| "loss": 0.5552, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.7555555555555555, | |
| "grad_norm": 288.6549072265625, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.4913, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9080459770114943, | |
| "eval_loss": 0.37552881240844727, | |
| "eval_runtime": 62.5531, | |
| "eval_samples_per_second": 9.736, | |
| "eval_steps_per_second": 0.623, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.1777777777777776, | |
| "grad_norm": 73.34674835205078, | |
| "learning_rate": 4.292929292929293e-05, | |
| "loss": 0.4369, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.6222222222222222, | |
| "grad_norm": 61.16302490234375, | |
| "learning_rate": 4.0404040404040405e-05, | |
| "loss": 0.4642, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6141215106732348, | |
| "eval_loss": 0.7140535116195679, | |
| "eval_runtime": 68.2847, | |
| "eval_samples_per_second": 8.919, | |
| "eval_steps_per_second": 0.571, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0444444444444443, | |
| "grad_norm": 245.7981414794922, | |
| "learning_rate": 3.787878787878788e-05, | |
| "loss": 0.4971, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.488888888888889, | |
| "grad_norm": 59.85107421875, | |
| "learning_rate": 3.535353535353535e-05, | |
| "loss": 0.4248, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.9333333333333336, | |
| "grad_norm": 26.801841735839844, | |
| "learning_rate": 3.282828282828283e-05, | |
| "loss": 0.4451, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8045977011494253, | |
| "eval_loss": 0.4347776770591736, | |
| "eval_runtime": 63.7065, | |
| "eval_samples_per_second": 9.559, | |
| "eval_steps_per_second": 0.612, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.355555555555555, | |
| "grad_norm": 125.90353393554688, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.3966, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 27.96786880493164, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4095, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8029556650246306, | |
| "eval_loss": 0.5060437917709351, | |
| "eval_runtime": 66.7998, | |
| "eval_samples_per_second": 9.117, | |
| "eval_steps_per_second": 0.584, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.222222222222222, | |
| "grad_norm": 28.08486557006836, | |
| "learning_rate": 2.5252525252525256e-05, | |
| "loss": 0.3595, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.666666666666667, | |
| "grad_norm": 44.42402267456055, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.3399, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7372742200328407, | |
| "eval_loss": 0.546423614025116, | |
| "eval_runtime": 63.3896, | |
| "eval_samples_per_second": 9.607, | |
| "eval_steps_per_second": 0.615, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.088888888888889, | |
| "grad_norm": 93.54395294189453, | |
| "learning_rate": 2.0202020202020203e-05, | |
| "loss": 0.333, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.533333333333333, | |
| "grad_norm": 56.091121673583984, | |
| "learning_rate": 1.7676767676767676e-05, | |
| "loss": 0.434, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.977777777777778, | |
| "grad_norm": 24.830839157104492, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.3304, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8883415435139573, | |
| "eval_loss": 0.32744449377059937, | |
| "eval_runtime": 63.7561, | |
| "eval_samples_per_second": 9.552, | |
| "eval_steps_per_second": 0.612, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "grad_norm": 59.196083068847656, | |
| "learning_rate": 1.2626262626262628e-05, | |
| "loss": 0.2963, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.844444444444444, | |
| "grad_norm": 22.917722702026367, | |
| "learning_rate": 1.0101010101010101e-05, | |
| "loss": 0.3539, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8604269293924466, | |
| "eval_loss": 0.38926875591278076, | |
| "eval_runtime": 63.7316, | |
| "eval_samples_per_second": 9.556, | |
| "eval_steps_per_second": 0.612, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.266666666666667, | |
| "grad_norm": 87.52684783935547, | |
| "learning_rate": 7.5757575757575764e-06, | |
| "loss": 0.2673, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.71111111111111, | |
| "grad_norm": 21.081283569335938, | |
| "learning_rate": 5.050505050505051e-06, | |
| "loss": 0.2849, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8637110016420362, | |
| "eval_loss": 0.3758099675178528, | |
| "eval_runtime": 62.4924, | |
| "eval_samples_per_second": 9.745, | |
| "eval_steps_per_second": 0.624, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.133333333333333, | |
| "grad_norm": 63.325809478759766, | |
| "learning_rate": 2.5252525252525253e-06, | |
| "loss": 0.265, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "grad_norm": 38.6427116394043, | |
| "learning_rate": 0.0, | |
| "loss": 0.2605, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "eval_accuracy": 0.8489326765188834, | |
| "eval_loss": 0.3968973755836487, | |
| "eval_runtime": 63.9436, | |
| "eval_samples_per_second": 9.524, | |
| "eval_steps_per_second": 0.61, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "step": 220, | |
| "total_flos": 2.7458744422511e+17, | |
| "train_loss": 0.4097213279117237, | |
| "train_runtime": 2363.2087, | |
| "train_samples_per_second": 6.089, | |
| "train_steps_per_second": 0.093 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 220, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 2.7458744422511e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |