{ "best_metric": 0.6287625418060201, "best_model_checkpoint": "dinov2-base-finetuned-eurosat/checkpoint-126", "epoch": 2.9881422924901186, "eval_steps": 500, "global_step": 189, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.15810276679841898, "grad_norm": 70.1077651977539, "learning_rate": 2.6315789473684212e-05, "loss": 0.76, "step": 10 }, { "epoch": 0.31620553359683795, "grad_norm": 48.27754211425781, "learning_rate": 4.970588235294118e-05, "loss": 1.2682, "step": 20 }, { "epoch": 0.4743083003952569, "grad_norm": 15.019556045532227, "learning_rate": 4.6764705882352944e-05, "loss": 0.7174, "step": 30 }, { "epoch": 0.6324110671936759, "grad_norm": 1.4938839673995972, "learning_rate": 4.382352941176471e-05, "loss": 0.649, "step": 40 }, { "epoch": 0.7905138339920948, "grad_norm": 6.551772117614746, "learning_rate": 4.0882352941176474e-05, "loss": 0.6611, "step": 50 }, { "epoch": 0.9486166007905138, "grad_norm": 12.39152717590332, "learning_rate": 3.794117647058824e-05, "loss": 0.6861, "step": 60 }, { "epoch": 0.9960474308300395, "eval_accuracy": 0.6276477146042363, "eval_loss": 0.6515693068504333, "eval_runtime": 20.1373, "eval_samples_per_second": 44.544, "eval_steps_per_second": 1.44, "step": 63 }, { "epoch": 1.1067193675889329, "grad_norm": 1.726658821105957, "learning_rate": 3.5e-05, "loss": 0.6682, "step": 70 }, { "epoch": 1.2648221343873518, "grad_norm": 7.709726810455322, "learning_rate": 3.205882352941177e-05, "loss": 0.6531, "step": 80 }, { "epoch": 1.4229249011857708, "grad_norm": 4.282989978790283, "learning_rate": 2.9117647058823534e-05, "loss": 0.6574, "step": 90 }, { "epoch": 1.5810276679841897, "grad_norm": 8.803805351257324, "learning_rate": 2.6176470588235295e-05, "loss": 0.6842, "step": 100 }, { "epoch": 1.7391304347826086, "grad_norm": 1.4123204946517944, "learning_rate": 2.323529411764706e-05, "loss": 0.6511, "step": 110 }, { "epoch": 1.8972332015810278, "grad_norm": 2.3040549755096436, "learning_rate": 2.0294117647058825e-05, "loss": 0.6401, "step": 120 }, { "epoch": 1.992094861660079, "eval_accuracy": 0.6287625418060201, "eval_loss": 0.6582574844360352, "eval_runtime": 20.0609, "eval_samples_per_second": 44.714, "eval_steps_per_second": 1.446, "step": 126 }, { "epoch": 2.0553359683794468, "grad_norm": 5.009319305419922, "learning_rate": 1.735294117647059e-05, "loss": 0.6274, "step": 130 }, { "epoch": 2.2134387351778657, "grad_norm": 4.530085563659668, "learning_rate": 1.4411764705882352e-05, "loss": 0.6532, "step": 140 }, { "epoch": 2.3715415019762847, "grad_norm": 0.5754580497741699, "learning_rate": 1.1470588235294118e-05, "loss": 0.6457, "step": 150 }, { "epoch": 2.5296442687747036, "grad_norm": 6.618398189544678, "learning_rate": 8.529411764705883e-06, "loss": 0.6514, "step": 160 }, { "epoch": 2.6877470355731226, "grad_norm": 3.6953773498535156, "learning_rate": 5.588235294117647e-06, "loss": 0.6178, "step": 170 }, { "epoch": 2.8458498023715415, "grad_norm": 6.594776153564453, "learning_rate": 2.647058823529412e-06, "loss": 0.6342, "step": 180 }, { "epoch": 2.9881422924901186, "eval_accuracy": 0.6287625418060201, "eval_loss": 0.6409574747085571, "eval_runtime": 20.3183, "eval_samples_per_second": 44.147, "eval_steps_per_second": 1.427, "step": 189 }, { "epoch": 2.9881422924901186, "step": 189, "total_flos": 1.8869538215666074e+18, "train_loss": 0.693489463241012, "train_runtime": 1138.9978, "train_samples_per_second": 21.242, "train_steps_per_second": 0.166 } ], "logging_steps": 10, "max_steps": 189, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "total_flos": 1.8869538215666074e+18, "train_batch_size": 32, "trial_name": null, "trial_params": null }