| { | |
| "best_metric": 0.9064, | |
| "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k-finetuning/checkpoint-351", | |
| "epoch": 0.997867803837953, | |
| "eval_steps": 500, | |
| "global_step": 351, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 12.392043113708496, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 2.38, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 9.366357803344727, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 2.0468, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 10.620584487915039, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.5116, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 9.78640079498291, | |
| "learning_rate": 4.936507936507937e-05, | |
| "loss": 1.0056, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 7.943943977355957, | |
| "learning_rate": 4.7777777777777784e-05, | |
| "loss": 0.7888, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 9.149613380432129, | |
| "learning_rate": 4.6190476190476194e-05, | |
| "loss": 0.6486, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 7.764841079711914, | |
| "learning_rate": 4.460317460317461e-05, | |
| "loss": 0.6232, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 8.434979438781738, | |
| "learning_rate": 4.301587301587302e-05, | |
| "loss": 0.5555, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 8.733748435974121, | |
| "learning_rate": 4.1428571428571437e-05, | |
| "loss": 0.4872, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 7.663728713989258, | |
| "learning_rate": 3.984126984126984e-05, | |
| "loss": 0.5385, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 6.9408698081970215, | |
| "learning_rate": 3.8253968253968256e-05, | |
| "loss": 0.4591, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 6.367518424987793, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.4449, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 6.589195728302002, | |
| "learning_rate": 3.5079365079365075e-05, | |
| "loss": 0.4365, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 8.5034818649292, | |
| "learning_rate": 3.349206349206349e-05, | |
| "loss": 0.4538, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 6.432806491851807, | |
| "learning_rate": 3.19047619047619e-05, | |
| "loss": 0.4182, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 5.926751613616943, | |
| "learning_rate": 3.0317460317460318e-05, | |
| "loss": 0.3683, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 7.516120433807373, | |
| "learning_rate": 2.8730158730158728e-05, | |
| "loss": 0.4315, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 7.042306423187256, | |
| "learning_rate": 2.714285714285714e-05, | |
| "loss": 0.3883, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 6.7353620529174805, | |
| "learning_rate": 2.5555555555555554e-05, | |
| "loss": 0.4158, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 6.9553914070129395, | |
| "learning_rate": 2.396825396825397e-05, | |
| "loss": 0.3893, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 6.948991298675537, | |
| "learning_rate": 2.2380952380952384e-05, | |
| "loss": 0.4288, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 6.90835428237915, | |
| "learning_rate": 2.0793650793650797e-05, | |
| "loss": 0.368, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 7.47681999206543, | |
| "learning_rate": 1.920634920634921e-05, | |
| "loss": 0.4212, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 5.4326887130737305, | |
| "learning_rate": 1.761904761904762e-05, | |
| "loss": 0.4101, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 5.068437576293945, | |
| "learning_rate": 1.6031746031746033e-05, | |
| "loss": 0.3544, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 5.686431884765625, | |
| "learning_rate": 1.4444444444444444e-05, | |
| "loss": 0.3252, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 5.882190227508545, | |
| "learning_rate": 1.2857142857142857e-05, | |
| "loss": 0.3475, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 6.3953447341918945, | |
| "learning_rate": 1.126984126984127e-05, | |
| "loss": 0.3164, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 7.278485298156738, | |
| "learning_rate": 9.682539682539683e-06, | |
| "loss": 0.328, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 6.7763166427612305, | |
| "learning_rate": 8.095238095238097e-06, | |
| "loss": 0.3463, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 7.21019172668457, | |
| "learning_rate": 6.507936507936509e-06, | |
| "loss": 0.3762, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 8.0834379196167, | |
| "learning_rate": 4.920634920634921e-06, | |
| "loss": 0.3538, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 5.969318389892578, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.3159, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 7.134821891784668, | |
| "learning_rate": 1.7460317460317462e-06, | |
| "loss": 0.3528, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.126035213470459, | |
| "learning_rate": 1.5873015873015874e-07, | |
| "loss": 0.2928, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9064, | |
| "eval_loss": 0.27925798296928406, | |
| "eval_runtime": 77.4382, | |
| "eval_samples_per_second": 64.568, | |
| "eval_steps_per_second": 2.027, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 351, | |
| "total_flos": 3.4803216831306793e+18, | |
| "train_loss": 0.5742807560666674, | |
| "train_runtime": 2256.218, | |
| "train_samples_per_second": 19.945, | |
| "train_steps_per_second": 0.156 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 351, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 3.4803216831306793e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |