| { | |
| "best_metric": 0.8981937602627258, | |
| "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": 349.2164306640625, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.611, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 174.1605682373047, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.4924, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8325123152709359, | |
| "eval_loss": 0.5212485194206238, | |
| "eval_runtime": 65.7326, | |
| "eval_samples_per_second": 9.265, | |
| "eval_steps_per_second": 0.593, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.3111111111111111, | |
| "grad_norm": 74.73553466796875, | |
| "learning_rate": 4.797979797979798e-05, | |
| "loss": 0.5101, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.7555555555555555, | |
| "grad_norm": 69.5820083618164, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.5732, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8981937602627258, | |
| "eval_loss": 0.33657345175743103, | |
| "eval_runtime": 69.7181, | |
| "eval_samples_per_second": 8.735, | |
| "eval_steps_per_second": 0.559, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.1777777777777776, | |
| "grad_norm": 36.10708999633789, | |
| "learning_rate": 4.292929292929293e-05, | |
| "loss": 0.3947, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.6222222222222222, | |
| "grad_norm": 220.30311584472656, | |
| "learning_rate": 4.0404040404040405e-05, | |
| "loss": 0.5639, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8489326765188834, | |
| "eval_loss": 0.3906978666782379, | |
| "eval_runtime": 70.3418, | |
| "eval_samples_per_second": 8.658, | |
| "eval_steps_per_second": 0.554, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0444444444444443, | |
| "grad_norm": 58.00751495361328, | |
| "learning_rate": 3.787878787878788e-05, | |
| "loss": 0.4467, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.488888888888889, | |
| "grad_norm": 61.74748992919922, | |
| "learning_rate": 3.535353535353535e-05, | |
| "loss": 0.521, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.9333333333333336, | |
| "grad_norm": 52.76782989501953, | |
| "learning_rate": 3.282828282828283e-05, | |
| "loss": 0.4759, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8817733990147784, | |
| "eval_loss": 0.3481961786746979, | |
| "eval_runtime": 69.182, | |
| "eval_samples_per_second": 8.803, | |
| "eval_steps_per_second": 0.564, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.355555555555555, | |
| "grad_norm": 83.23902893066406, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.3662, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 40.003013610839844, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3757, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.392085462808609, | |
| "eval_runtime": 67.7229, | |
| "eval_samples_per_second": 8.993, | |
| "eval_steps_per_second": 0.576, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.222222222222222, | |
| "grad_norm": 54.831241607666016, | |
| "learning_rate": 2.5252525252525256e-05, | |
| "loss": 0.3124, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.666666666666667, | |
| "grad_norm": 45.63344192504883, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.3356, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.896551724137931, | |
| "eval_loss": 0.3183760344982147, | |
| "eval_runtime": 67.9276, | |
| "eval_samples_per_second": 8.965, | |
| "eval_steps_per_second": 0.574, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.088888888888889, | |
| "grad_norm": 52.133689880371094, | |
| "learning_rate": 2.0202020202020203e-05, | |
| "loss": 0.2883, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.533333333333333, | |
| "grad_norm": 35.971134185791016, | |
| "learning_rate": 1.7676767676767676e-05, | |
| "loss": 0.3177, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.977777777777778, | |
| "grad_norm": 57.66413879394531, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.2521, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.39921438694000244, | |
| "eval_runtime": 65.0226, | |
| "eval_samples_per_second": 9.366, | |
| "eval_steps_per_second": 0.6, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "grad_norm": 74.5230484008789, | |
| "learning_rate": 1.2626262626262628e-05, | |
| "loss": 0.2193, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.844444444444444, | |
| "grad_norm": 80.1707534790039, | |
| "learning_rate": 1.0101010101010101e-05, | |
| "loss": 0.2981, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8702791461412152, | |
| "eval_loss": 0.3903738856315613, | |
| "eval_runtime": 67.5642, | |
| "eval_samples_per_second": 9.014, | |
| "eval_steps_per_second": 0.577, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.266666666666667, | |
| "grad_norm": 82.75043487548828, | |
| "learning_rate": 7.5757575757575764e-06, | |
| "loss": 0.1812, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.71111111111111, | |
| "grad_norm": 54.377445220947266, | |
| "learning_rate": 5.050505050505051e-06, | |
| "loss": 0.2302, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8719211822660099, | |
| "eval_loss": 0.3987477421760559, | |
| "eval_runtime": 67.5557, | |
| "eval_samples_per_second": 9.015, | |
| "eval_steps_per_second": 0.577, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.133333333333333, | |
| "grad_norm": 65.00289916992188, | |
| "learning_rate": 2.5252525252525253e-06, | |
| "loss": 0.2021, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "grad_norm": 52.96971130371094, | |
| "learning_rate": 0.0, | |
| "loss": 0.1979, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "eval_accuracy": 0.8604269293924466, | |
| "eval_loss": 0.4129030406475067, | |
| "eval_runtime": 66.9132, | |
| "eval_samples_per_second": 9.101, | |
| "eval_steps_per_second": 0.583, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "step": 220, | |
| "total_flos": 2.7458744422511e+17, | |
| "train_loss": 0.37115615768866106, | |
| "train_runtime": 2504.6096, | |
| "train_samples_per_second": 5.745, | |
| "train_steps_per_second": 0.088 | |
| } | |
| ], | |
| "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 | |
| } | |