{
  "best_metric": 0.0003763487620744854,
  "best_model_checkpoint": "./vit-base-fruit-punch/checkpoint-1000",
  "epoch": 8.0,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 100187.125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 1.127,
      "step": 10
    },
    {
      "epoch": 0.16,
      "grad_norm": 96563.390625,
      "learning_rate": 4.9e-05,
      "loss": 0.7007,
      "step": 20
    },
    {
      "epoch": 0.24,
      "grad_norm": 64723.85546875,
      "learning_rate": 4.85e-05,
      "loss": 0.4167,
      "step": 30
    },
    {
      "epoch": 0.32,
      "grad_norm": 40675.4140625,
      "learning_rate": 4.8e-05,
      "loss": 0.2602,
      "step": 40
    },
    {
      "epoch": 0.4,
      "grad_norm": 28872.85546875,
      "learning_rate": 4.75e-05,
      "loss": 0.16,
      "step": 50
    },
    {
      "epoch": 0.48,
      "grad_norm": 26935.421875,
      "learning_rate": 4.7e-05,
      "loss": 0.1115,
      "step": 60
    },
    {
      "epoch": 0.56,
      "grad_norm": 19970.119140625,
      "learning_rate": 4.6500000000000005e-05,
      "loss": 0.0944,
      "step": 70
    },
    {
      "epoch": 0.64,
      "grad_norm": 19774.392578125,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.0761,
      "step": 80
    },
    {
      "epoch": 0.72,
      "grad_norm": 14233.1318359375,
      "learning_rate": 4.55e-05,
      "loss": 0.0588,
      "step": 90
    },
    {
      "epoch": 0.8,
      "grad_norm": 12306.6767578125,
      "learning_rate": 4.5e-05,
      "loss": 0.0488,
      "step": 100
    },
    {
      "epoch": 0.8,
      "eval_accuracy": 1.0,
      "eval_loss": 0.04701722040772438,
      "eval_runtime": 12.891,
      "eval_samples_per_second": 77.574,
      "eval_steps_per_second": 4.887,
      "step": 100
    },
    {
      "epoch": 0.88,
      "grad_norm": 11777.6708984375,
      "learning_rate": 4.4500000000000004e-05,
      "loss": 0.0429,
      "step": 110
    },
    {
      "epoch": 0.96,
      "grad_norm": 10813.1064453125,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0458,
      "step": 120
    },
    {
      "epoch": 1.04,
      "grad_norm": 8880.1279296875,
      "learning_rate": 4.35e-05,
      "loss": 0.0384,
      "step": 130
    },
    {
      "epoch": 1.12,
      "grad_norm": 8182.60107421875,
      "learning_rate": 4.3e-05,
      "loss": 0.0309,
      "step": 140
    },
    {
      "epoch": 1.2,
      "grad_norm": 7128.07275390625,
      "learning_rate": 4.25e-05,
      "loss": 0.0281,
      "step": 150
    },
    {
      "epoch": 1.28,
      "grad_norm": 6803.498046875,
      "learning_rate": 4.2e-05,
      "loss": 0.0254,
      "step": 160
    },
    {
      "epoch": 1.3599999999999999,
      "grad_norm": 6198.88037109375,
      "learning_rate": 4.15e-05,
      "loss": 0.0232,
      "step": 170
    },
    {
      "epoch": 1.44,
      "grad_norm": 5394.99072265625,
      "learning_rate": 4.1e-05,
      "loss": 0.021,
      "step": 180
    },
    {
      "epoch": 1.52,
      "grad_norm": 5170.45458984375,
      "learning_rate": 4.05e-05,
      "loss": 0.0191,
      "step": 190
    },
    {
      "epoch": 1.6,
      "grad_norm": 4661.20263671875,
      "learning_rate": 4e-05,
      "loss": 0.0174,
      "step": 200
    },
    {
      "epoch": 1.6,
      "eval_accuracy": 1.0,
      "eval_loss": 0.017339378595352173,
      "eval_runtime": 13.5483,
      "eval_samples_per_second": 73.81,
      "eval_steps_per_second": 4.65,
      "step": 200
    },
    {
      "epoch": 1.6800000000000002,
      "grad_norm": 4417.46337890625,
      "learning_rate": 3.9500000000000005e-05,
      "loss": 0.016,
      "step": 210
    },
    {
      "epoch": 1.76,
      "grad_norm": 3967.432861328125,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 0.0146,
      "step": 220
    },
    {
      "epoch": 1.8399999999999999,
      "grad_norm": 3568.837646484375,
      "learning_rate": 3.85e-05,
      "loss": 0.0134,
      "step": 230
    },
    {
      "epoch": 1.92,
      "grad_norm": 3472.84716796875,
      "learning_rate": 3.8e-05,
      "loss": 0.0123,
      "step": 240
    },
    {
      "epoch": 2.0,
      "grad_norm": 3190.490966796875,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.0113,
      "step": 250
    },
    {
      "epoch": 2.08,
      "grad_norm": 2825.7802734375,
      "learning_rate": 3.7e-05,
      "loss": 0.0104,
      "step": 260
    },
    {
      "epoch": 2.16,
      "grad_norm": 2605.496337890625,
      "learning_rate": 3.65e-05,
      "loss": 0.0095,
      "step": 270
    },
    {
      "epoch": 2.24,
      "grad_norm": 2393.7314453125,
      "learning_rate": 3.6e-05,
      "loss": 0.0088,
      "step": 280
    },
    {
      "epoch": 2.32,
      "grad_norm": 2354.353515625,
      "learning_rate": 3.55e-05,
      "loss": 0.0081,
      "step": 290
    },
    {
      "epoch": 2.4,
      "grad_norm": 2027.639404296875,
      "learning_rate": 3.5e-05,
      "loss": 0.0074,
      "step": 300
    },
    {
      "epoch": 2.4,
      "eval_accuracy": 1.0,
      "eval_loss": 0.007476483471691608,
      "eval_runtime": 13.6387,
      "eval_samples_per_second": 73.321,
      "eval_steps_per_second": 4.619,
      "step": 300
    },
    {
      "epoch": 2.48,
      "grad_norm": 1890.090087890625,
      "learning_rate": 3.45e-05,
      "loss": 0.0068,
      "step": 310
    },
    {
      "epoch": 2.56,
      "grad_norm": 1713.1453857421875,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 0.0063,
      "step": 320
    },
    {
      "epoch": 2.64,
      "grad_norm": 1611.7166748046875,
      "learning_rate": 3.35e-05,
      "loss": 0.0059,
      "step": 330
    },
    {
      "epoch": 2.7199999999999998,
      "grad_norm": 1491.3282470703125,
      "learning_rate": 3.3e-05,
      "loss": 0.0054,
      "step": 340
    },
    {
      "epoch": 2.8,
      "grad_norm": 1385.7913818359375,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 0.005,
      "step": 350
    },
    {
      "epoch": 2.88,
      "grad_norm": 1317.3277587890625,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.0047,
      "step": 360
    },
    {
      "epoch": 2.96,
      "grad_norm": 1197.3973388671875,
      "learning_rate": 3.15e-05,
      "loss": 0.0043,
      "step": 370
    },
    {
      "epoch": 3.04,
      "grad_norm": 1110.10693359375,
      "learning_rate": 3.1e-05,
      "loss": 0.004,
      "step": 380
    },
    {
      "epoch": 3.12,
      "grad_norm": 1046.7801513671875,
      "learning_rate": 3.05e-05,
      "loss": 0.0037,
      "step": 390
    },
    {
      "epoch": 3.2,
      "grad_norm": 958.0781860351562,
      "learning_rate": 3e-05,
      "loss": 0.0035,
      "step": 400
    },
    {
      "epoch": 3.2,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0034720886033028364,
      "eval_runtime": 13.6477,
      "eval_samples_per_second": 73.272,
      "eval_steps_per_second": 4.616,
      "step": 400
    },
    {
      "epoch": 3.2800000000000002,
      "grad_norm": 986.9517822265625,
      "learning_rate": 2.95e-05,
      "loss": 0.0032,
      "step": 410
    },
    {
      "epoch": 3.36,
      "grad_norm": 841.5371704101562,
      "learning_rate": 2.9e-05,
      "loss": 0.003,
      "step": 420
    },
    {
      "epoch": 3.44,
      "grad_norm": 797.5939331054688,
      "learning_rate": 2.8499999999999998e-05,
      "loss": 0.0028,
      "step": 430
    },
    {
      "epoch": 3.52,
      "grad_norm": 735.5321655273438,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 0.0026,
      "step": 440
    },
    {
      "epoch": 3.6,
      "grad_norm": 701.638427734375,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 0.0024,
      "step": 450
    },
    {
      "epoch": 3.68,
      "grad_norm": 647.36279296875,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 0.0023,
      "step": 460
    },
    {
      "epoch": 3.76,
      "grad_norm": 599.30126953125,
      "learning_rate": 2.6500000000000004e-05,
      "loss": 0.0021,
      "step": 470
    },
    {
      "epoch": 3.84,
      "grad_norm": 590.8321533203125,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 0.002,
      "step": 480
    },
    {
      "epoch": 3.92,
      "grad_norm": 546.0530395507812,
      "learning_rate": 2.5500000000000003e-05,
      "loss": 0.0019,
      "step": 490
    },
    {
      "epoch": 4.0,
      "grad_norm": 500.25738525390625,
      "learning_rate": 2.5e-05,
      "loss": 0.0018,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0022775332909077406,
      "eval_runtime": 13.3322,
      "eval_samples_per_second": 75.006,
      "eval_steps_per_second": 4.725,
      "step": 500
    },
    {
      "epoch": 4.08,
      "grad_norm": 494.0545349121094,
      "learning_rate": 2.45e-05,
      "loss": 0.0017,
      "step": 510
    },
    {
      "epoch": 4.16,
      "grad_norm": 452.84375,
      "learning_rate": 2.4e-05,
      "loss": 0.0016,
      "step": 520
    },
    {
      "epoch": 4.24,
      "grad_norm": 434.6565246582031,
      "learning_rate": 2.35e-05,
      "loss": 0.0015,
      "step": 530
    },
    {
      "epoch": 4.32,
      "grad_norm": 404.3072204589844,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 0.0014,
      "step": 540
    },
    {
      "epoch": 4.4,
      "grad_norm": 389.670166015625,
      "learning_rate": 2.25e-05,
      "loss": 0.0013,
      "step": 550
    },
    {
      "epoch": 4.48,
      "grad_norm": 387.8797607421875,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.0012,
      "step": 560
    },
    {
      "epoch": 4.5600000000000005,
      "grad_norm": 357.6133728027344,
      "learning_rate": 2.15e-05,
      "loss": 0.0012,
      "step": 570
    },
    {
      "epoch": 4.64,
      "grad_norm": 320.7620544433594,
      "learning_rate": 2.1e-05,
      "loss": 0.0011,
      "step": 580
    },
    {
      "epoch": 4.72,
      "grad_norm": 309.20062255859375,
      "learning_rate": 2.05e-05,
      "loss": 0.001,
      "step": 590
    },
    {
      "epoch": 4.8,
      "grad_norm": 292.0805358886719,
      "learning_rate": 2e-05,
      "loss": 0.001,
      "step": 600
    },
    {
      "epoch": 4.8,
      "eval_accuracy": 1.0,
      "eval_loss": 0.001073041232302785,
      "eval_runtime": 13.7046,
      "eval_samples_per_second": 72.968,
      "eval_steps_per_second": 4.597,
      "step": 600
    },
    {
      "epoch": 4.88,
      "grad_norm": 283.4959411621094,
      "learning_rate": 1.9500000000000003e-05,
      "loss": 0.0009,
      "step": 610
    },
    {
      "epoch": 4.96,
      "grad_norm": 261.8572998046875,
      "learning_rate": 1.9e-05,
      "loss": 0.0009,
      "step": 620
    },
    {
      "epoch": 5.04,
      "grad_norm": 252.91981506347656,
      "learning_rate": 1.85e-05,
      "loss": 0.0009,
      "step": 630
    },
    {
      "epoch": 5.12,
      "grad_norm": 239.09896850585938,
      "learning_rate": 1.8e-05,
      "loss": 0.0008,
      "step": 640
    },
    {
      "epoch": 5.2,
      "grad_norm": 232.7013397216797,
      "learning_rate": 1.75e-05,
      "loss": 0.0008,
      "step": 650
    },
    {
      "epoch": 5.28,
      "grad_norm": 220.06301879882812,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 0.0007,
      "step": 660
    },
    {
      "epoch": 5.36,
      "grad_norm": 219.54986572265625,
      "learning_rate": 1.65e-05,
      "loss": 0.0007,
      "step": 670
    },
    {
      "epoch": 5.44,
      "grad_norm": 206.40716552734375,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0007,
      "step": 680
    },
    {
      "epoch": 5.52,
      "grad_norm": 196.642578125,
      "learning_rate": 1.55e-05,
      "loss": 0.0007,
      "step": 690
    },
    {
      "epoch": 5.6,
      "grad_norm": 187.69554138183594,
      "learning_rate": 1.5e-05,
      "loss": 0.0006,
      "step": 700
    },
    {
      "epoch": 5.6,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0006605549133382738,
      "eval_runtime": 13.6953,
      "eval_samples_per_second": 73.018,
      "eval_steps_per_second": 4.6,
      "step": 700
    },
    {
      "epoch": 5.68,
      "grad_norm": 183.08045959472656,
      "learning_rate": 1.45e-05,
      "loss": 0.0006,
      "step": 710
    },
    {
      "epoch": 5.76,
      "grad_norm": 174.93222045898438,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.0006,
      "step": 720
    },
    {
      "epoch": 5.84,
      "grad_norm": 177.07530212402344,
      "learning_rate": 1.3500000000000001e-05,
      "loss": 0.0006,
      "step": 730
    },
    {
      "epoch": 5.92,
      "grad_norm": 166.14947509765625,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.0005,
      "step": 740
    },
    {
      "epoch": 6.0,
      "grad_norm": 165.67318725585938,
      "learning_rate": 1.25e-05,
      "loss": 0.0005,
      "step": 750
    },
    {
      "epoch": 6.08,
      "grad_norm": 158.77545166015625,
      "learning_rate": 1.2e-05,
      "loss": 0.0005,
      "step": 760
    },
    {
      "epoch": 6.16,
      "grad_norm": 149.71511840820312,
      "learning_rate": 1.1500000000000002e-05,
      "loss": 0.0005,
      "step": 770
    },
    {
      "epoch": 6.24,
      "grad_norm": 144.20770263671875,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.0005,
      "step": 780
    },
    {
      "epoch": 6.32,
      "grad_norm": 148.6312255859375,
      "learning_rate": 1.05e-05,
      "loss": 0.0005,
      "step": 790
    },
    {
      "epoch": 6.4,
      "grad_norm": 135.4142303466797,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "step": 800
    },
    {
      "epoch": 6.4,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0004884201916866004,
      "eval_runtime": 13.5521,
      "eval_samples_per_second": 73.79,
      "eval_steps_per_second": 4.649,
      "step": 800
    },
    {
      "epoch": 6.48,
      "grad_norm": 133.83642578125,
      "learning_rate": 9.5e-06,
      "loss": 0.0004,
      "step": 810
    },
    {
      "epoch": 6.5600000000000005,
      "grad_norm": 138.82203674316406,
      "learning_rate": 9e-06,
      "loss": 0.0004,
      "step": 820
    },
    {
      "epoch": 6.64,
      "grad_norm": 127.43915557861328,
      "learning_rate": 8.500000000000002e-06,
      "loss": 0.0004,
      "step": 830
    },
    {
      "epoch": 6.72,
      "grad_norm": 126.12251281738281,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0004,
      "step": 840
    },
    {
      "epoch": 6.8,
      "grad_norm": 121.66053771972656,
      "learning_rate": 7.5e-06,
      "loss": 0.0004,
      "step": 850
    },
    {
      "epoch": 6.88,
      "grad_norm": 121.25574493408203,
      "learning_rate": 7.000000000000001e-06,
      "loss": 0.0004,
      "step": 860
    },
    {
      "epoch": 6.96,
      "grad_norm": 126.3290023803711,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.0004,
      "step": 870
    },
    {
      "epoch": 7.04,
      "grad_norm": 117.99575805664062,
      "learning_rate": 6e-06,
      "loss": 0.0004,
      "step": 880
    },
    {
      "epoch": 7.12,
      "grad_norm": 116.10645294189453,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.0004,
      "step": 890
    },
    {
      "epoch": 7.2,
      "grad_norm": 113.11275482177734,
      "learning_rate": 5e-06,
      "loss": 0.0004,
      "step": 900
    },
    {
      "epoch": 7.2,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0003921452153008431,
      "eval_runtime": 13.7013,
      "eval_samples_per_second": 72.986,
      "eval_steps_per_second": 4.598,
      "step": 900
    },
    {
      "epoch": 7.28,
      "grad_norm": 114.9211196899414,
      "learning_rate": 4.5e-06,
      "loss": 0.0004,
      "step": 910
    },
    {
      "epoch": 7.36,
      "grad_norm": 110.5498046875,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0004,
      "step": 920
    },
    {
      "epoch": 7.44,
      "grad_norm": 110.70841979980469,
      "learning_rate": 3.5000000000000004e-06,
      "loss": 0.0004,
      "step": 930
    },
    {
      "epoch": 7.52,
      "grad_norm": 115.6305160522461,
      "learning_rate": 3e-06,
      "loss": 0.0004,
      "step": 940
    },
    {
      "epoch": 7.6,
      "grad_norm": 106.5681381225586,
      "learning_rate": 2.5e-06,
      "loss": 0.0004,
      "step": 950
    },
    {
      "epoch": 7.68,
      "grad_norm": 109.81066131591797,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0003,
      "step": 960
    },
    {
      "epoch": 7.76,
      "grad_norm": 107.74824523925781,
      "learning_rate": 1.5e-06,
      "loss": 0.0003,
      "step": 970
    },
    {
      "epoch": 7.84,
      "grad_norm": 108.34854888916016,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.0003,
      "step": 980
    },
    {
      "epoch": 7.92,
      "grad_norm": 107.01416778564453,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.0003,
      "step": 990
    },
    {
      "epoch": 8.0,
      "grad_norm": 111.06143188476562,
      "learning_rate": 0.0,
      "loss": 0.0003,
      "step": 1000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0003763487620744854,
      "eval_runtime": 13.1171,
      "eval_samples_per_second": 76.236,
      "eval_steps_per_second": 4.803,
      "step": 1000
    },
    {
      "epoch": 8.0,
      "step": 1000,
      "total_flos": 2.479168170953736e+18,
      "train_loss": 0.03565365221118554,
      "train_runtime": 1044.6113,
      "train_samples_per_second": 30.626,
      "train_steps_per_second": 0.957
    }
  ],
  "logging_steps": 10,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.479168170953736e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}