{
  "best_global_step": 581,
  "best_metric": 0.7325842696629213,
  "best_model_checkpoint": "checkpoints/camonet/checkpoint-581",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 581,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24096385542168675,
      "grad_norm": 111.35398864746094,
      "learning_rate": 1.417910447761194e-05,
      "loss": 4.043815612792969,
      "step": 20
    },
    {
      "epoch": 0.4819277108433735,
      "grad_norm": 71.05435943603516,
      "learning_rate": 2.91044776119403e-05,
      "loss": 3.5163707733154297,
      "step": 40
    },
    {
      "epoch": 0.7228915662650602,
      "grad_norm": 60.80617141723633,
      "learning_rate": 4.402985074626866e-05,
      "loss": 3.0165132522583007,
      "step": 60
    },
    {
      "epoch": 0.963855421686747,
      "grad_norm": 64.55865478515625,
      "learning_rate": 4.9950171333287335e-05,
      "loss": 2.4458030700683593,
      "step": 80
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.536519765853882,
      "eval_runtime": 6.8652,
      "eval_samples_per_second": 64.82,
      "eval_steps_per_second": 2.039,
      "eval_top1": 0.3325842696629214,
      "eval_top3": 0.5438202247191011,
      "step": 83
    },
    {
      "epoch": 1.2048192771084336,
      "grad_norm": 60.52080535888672,
      "learning_rate": 4.964638184169378e-05,
      "loss": 2.153371238708496,
      "step": 100
    },
    {
      "epoch": 1.4457831325301205,
      "grad_norm": 70.68125915527344,
      "learning_rate": 4.906984324751821e-05,
      "loss": 1.828770637512207,
      "step": 120
    },
    {
      "epoch": 1.6867469879518073,
      "grad_norm": 60.982425689697266,
      "learning_rate": 4.822693581319326e-05,
      "loss": 1.7083017349243164,
      "step": 140
    },
    {
      "epoch": 1.927710843373494,
      "grad_norm": 58.39655303955078,
      "learning_rate": 4.712698757103414e-05,
      "loss": 1.712203025817871,
      "step": 160
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.7305916547775269,
      "eval_runtime": 7.5482,
      "eval_samples_per_second": 58.954,
      "eval_steps_per_second": 1.855,
      "eval_top1": 0.5415730337078651,
      "eval_top3": 0.7146067415730337,
      "step": 166
    },
    {
      "epoch": 2.1686746987951806,
      "grad_norm": 60.08979415893555,
      "learning_rate": 4.5782171094600005e-05,
      "loss": 1.3887232780456542,
      "step": 180
    },
    {
      "epoch": 2.4096385542168672,
      "grad_norm": 64.29576873779297,
      "learning_rate": 4.420736879094927e-05,
      "loss": 1.3437091827392578,
      "step": 200
    },
    {
      "epoch": 2.6506024096385543,
      "grad_norm": 62.950355529785156,
      "learning_rate": 4.242000820453141e-05,
      "loss": 1.139847469329834,
      "step": 220
    },
    {
      "epoch": 2.891566265060241,
      "grad_norm": 43.11043167114258,
      "learning_rate": 4.043986915532434e-05,
      "loss": 1.2316542625427247,
      "step": 240
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.4789479970932007,
      "eval_runtime": 7.5421,
      "eval_samples_per_second": 59.002,
      "eval_steps_per_second": 1.856,
      "eval_top1": 0.6269662921348315,
      "eval_top3": 0.7842696629213484,
      "step": 249
    },
    {
      "epoch": 3.1325301204819276,
      "grad_norm": 66.58777618408203,
      "learning_rate": 3.828886484552254e-05,
      "loss": 1.033012580871582,
      "step": 260
    },
    {
      "epoch": 3.3734939759036147,
      "grad_norm": 53.87052536010742,
      "learning_rate": 3.5990799357157864e-05,
      "loss": 0.8578171730041504,
      "step": 280
    },
    {
      "epoch": 3.6144578313253013,
      "grad_norm": 72.20587921142578,
      "learning_rate": 3.357110422430506e-05,
      "loss": 0.8009697914123535,
      "step": 300
    },
    {
      "epoch": 3.855421686746988,
      "grad_norm": 47.637420654296875,
      "learning_rate": 3.105655699509458e-05,
      "loss": 0.8424944877624512,
      "step": 320
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.3673882484436035,
      "eval_runtime": 8.6387,
      "eval_samples_per_second": 51.512,
      "eval_steps_per_second": 1.621,
      "eval_top1": 0.6202247191011236,
      "eval_top3": 0.8112359550561797,
      "step": 332
    },
    {
      "epoch": 4.096385542168675,
      "grad_norm": 44.70951461791992,
      "learning_rate": 2.8474984898065406e-05,
      "loss": 0.7181183815002441,
      "step": 340
    },
    {
      "epoch": 4.337349397590361,
      "grad_norm": 40.632354736328125,
      "learning_rate": 2.5854956892233006e-05,
      "loss": 0.6570661544799805,
      "step": 360
    },
    {
      "epoch": 4.578313253012048,
      "grad_norm": 41.45029067993164,
      "learning_rate": 2.32254675087996e-05,
      "loss": 0.5137750148773194,
      "step": 380
    },
    {
      "epoch": 4.8192771084337345,
      "grad_norm": 53.68215560913086,
      "learning_rate": 2.061561598327112e-05,
      "loss": 0.5405148029327392,
      "step": 400
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.1706817150115967,
      "eval_runtime": 8.8354,
      "eval_samples_per_second": 50.365,
      "eval_steps_per_second": 1.585,
      "eval_top1": 0.6831460674157304,
      "eval_top3": 0.849438202247191,
      "step": 415
    },
    {
      "epoch": 5.0602409638554215,
      "grad_norm": 75.12175750732422,
      "learning_rate": 1.8054284228864366e-05,
      "loss": 0.49901180267333983,
      "step": 420
    },
    {
      "epoch": 5.301204819277109,
      "grad_norm": 18.631591796875,
      "learning_rate": 1.55698172149106e-05,
      "loss": 0.29165282249450686,
      "step": 440
    },
    {
      "epoch": 5.542168674698795,
      "grad_norm": 10.731932640075684,
      "learning_rate": 1.3189709287346813e-05,
      "loss": 0.3043174982070923,
      "step": 460
    },
    {
      "epoch": 5.783132530120482,
      "grad_norm": 47.11884689331055,
      "learning_rate": 1.0940299902627938e-05,
      "loss": 0.41364297866821287,
      "step": 480
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.1097288131713867,
      "eval_runtime": 8.2234,
      "eval_samples_per_second": 54.114,
      "eval_steps_per_second": 1.702,
      "eval_top1": 0.7258426966292135,
      "eval_top3": 0.8674157303370786,
      "step": 498
    },
    {
      "epoch": 6.024096385542169,
      "grad_norm": 20.367740631103516,
      "learning_rate": 8.846482142219752e-06,
      "loss": 0.34232077598571775,
      "step": 500
    },
    {
      "epoch": 6.265060240963855,
      "grad_norm": 29.076051712036133,
      "learning_rate": 6.931427233395696e-06,
      "loss": 0.18118011951446533,
      "step": 520
    },
    {
      "epoch": 6.506024096385542,
      "grad_norm": 21.859106063842773,
      "learning_rate": 5.21632812492738e-06,
      "loss": 0.204935622215271,
      "step": 540
    },
    {
      "epoch": 6.746987951807229,
      "grad_norm": 24.722841262817383,
      "learning_rate": 3.7201649553876366e-06,
      "loss": 0.21130716800689697,
      "step": 560
    },
    {
      "epoch": 6.9879518072289155,
      "grad_norm": 27.248981475830078,
      "learning_rate": 2.4594950095101513e-06,
      "loss": 0.2202404260635376,
      "step": 580
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.105672001838684,
      "eval_runtime": 8.1718,
      "eval_samples_per_second": 54.455,
      "eval_steps_per_second": 1.713,
      "eval_top1": 0.7325842696629213,
      "eval_top3": 0.8741573033707866,
      "step": 581
    }
  ],
  "logging_steps": 20,
  "max_steps": 664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.8058967165239296e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}