{
  "best_metric": 0.02545962668955326,
  "best_model_checkpoint": "./outputsstar/checkpoint-740",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 740,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06756756756756757,
      "grad_norm": 1.9284253120422363,
      "learning_rate": 1.972972972972973e-05,
      "loss": 0.5718,
      "step": 10
    },
    {
      "epoch": 0.13513513513513514,
      "grad_norm": 1.4602636098861694,
      "learning_rate": 1.9459459459459463e-05,
      "loss": 0.3133,
      "step": 20
    },
    {
      "epoch": 0.20270270270270271,
      "grad_norm": 1.0061994791030884,
      "learning_rate": 1.918918918918919e-05,
      "loss": 0.1601,
      "step": 30
    },
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 0.45298728346824646,
      "learning_rate": 1.891891891891892e-05,
      "loss": 0.1146,
      "step": 40
    },
    {
      "epoch": 0.33783783783783783,
      "grad_norm": 0.5685911178588867,
      "learning_rate": 1.864864864864865e-05,
      "loss": 0.0863,
      "step": 50
    },
    {
      "epoch": 0.40540540540540543,
      "grad_norm": 0.4350438714027405,
      "learning_rate": 1.8378378378378383e-05,
      "loss": 0.0465,
      "step": 60
    },
    {
      "epoch": 0.47297297297297297,
      "grad_norm": 0.20809537172317505,
      "learning_rate": 1.8108108108108108e-05,
      "loss": 0.0345,
      "step": 70
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 0.1794459968805313,
      "learning_rate": 1.783783783783784e-05,
      "loss": 0.0447,
      "step": 80
    },
    {
      "epoch": 0.6081081081081081,
      "grad_norm": 0.17304065823554993,
      "learning_rate": 1.756756756756757e-05,
      "loss": 0.0638,
      "step": 90
    },
    {
      "epoch": 0.6756756756756757,
      "grad_norm": 0.16675682365894318,
      "learning_rate": 1.72972972972973e-05,
      "loss": 0.058,
      "step": 100
    },
    {
      "epoch": 0.7432432432432432,
      "grad_norm": 0.158709317445755,
      "learning_rate": 1.7027027027027028e-05,
      "loss": 0.0691,
      "step": 110
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 0.17359796166419983,
      "learning_rate": 1.6756756756756757e-05,
      "loss": 0.075,
      "step": 120
    },
    {
      "epoch": 0.8783783783783784,
      "grad_norm": 3.065023422241211,
      "learning_rate": 1.648648648648649e-05,
      "loss": 0.0731,
      "step": 130
    },
    {
      "epoch": 0.9459459459459459,
      "grad_norm": 0.18713080883026123,
      "learning_rate": 1.6216216216216218e-05,
      "loss": 0.0436,
      "step": 140
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9808612440191388,
      "eval_loss": 0.05818231776356697,
      "eval_runtime": 27.0314,
      "eval_samples_per_second": 7.732,
      "eval_steps_per_second": 0.999,
      "step": 148
    },
    {
      "epoch": 1.0135135135135136,
      "grad_norm": 1.1932803392410278,
      "learning_rate": 1.5945945945945947e-05,
      "loss": 0.0697,
      "step": 150
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 0.15206500887870789,
      "learning_rate": 1.5675675675675676e-05,
      "loss": 0.0186,
      "step": 160
    },
    {
      "epoch": 1.1486486486486487,
      "grad_norm": 5.430366039276123,
      "learning_rate": 1.540540540540541e-05,
      "loss": 0.0391,
      "step": 170
    },
    {
      "epoch": 1.2162162162162162,
      "grad_norm": 0.12790042161941528,
      "learning_rate": 1.5135135135135138e-05,
      "loss": 0.0149,
      "step": 180
    },
    {
      "epoch": 1.2837837837837838,
      "grad_norm": 0.09696532785892487,
      "learning_rate": 1.4864864864864865e-05,
      "loss": 0.0134,
      "step": 190
    },
    {
      "epoch": 1.3513513513513513,
      "grad_norm": 0.10962755233049393,
      "learning_rate": 1.4594594594594596e-05,
      "loss": 0.0775,
      "step": 200
    },
    {
      "epoch": 1.4189189189189189,
      "grad_norm": 0.10070909559726715,
      "learning_rate": 1.4324324324324326e-05,
      "loss": 0.0524,
      "step": 210
    },
    {
      "epoch": 1.4864864864864864,
      "grad_norm": 0.09208140522241592,
      "learning_rate": 1.4054054054054055e-05,
      "loss": 0.0243,
      "step": 220
    },
    {
      "epoch": 1.554054054054054,
      "grad_norm": 5.711668968200684,
      "learning_rate": 1.3783783783783784e-05,
      "loss": 0.0358,
      "step": 230
    },
    {
      "epoch": 1.6216216216216215,
      "grad_norm": 0.09372463822364807,
      "learning_rate": 1.3513513513513515e-05,
      "loss": 0.0125,
      "step": 240
    },
    {
      "epoch": 1.689189189189189,
      "grad_norm": 0.08806200325489044,
      "learning_rate": 1.3243243243243244e-05,
      "loss": 0.0532,
      "step": 250
    },
    {
      "epoch": 1.7567567567567568,
      "grad_norm": 25.615135192871094,
      "learning_rate": 1.2972972972972975e-05,
      "loss": 0.0761,
      "step": 260
    },
    {
      "epoch": 1.8243243243243243,
      "grad_norm": 0.10457191616296768,
      "learning_rate": 1.2702702702702702e-05,
      "loss": 0.0417,
      "step": 270
    },
    {
      "epoch": 1.8918918918918919,
      "grad_norm": 0.09721718728542328,
      "learning_rate": 1.2432432432432433e-05,
      "loss": 0.0144,
      "step": 280
    },
    {
      "epoch": 1.9594594594594594,
      "grad_norm": 0.0960867628455162,
      "learning_rate": 1.2162162162162164e-05,
      "loss": 0.0121,
      "step": 290
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9904306220095693,
      "eval_loss": 0.04049909487366676,
      "eval_runtime": 27.4721,
      "eval_samples_per_second": 7.608,
      "eval_steps_per_second": 0.983,
      "step": 296
    },
    {
      "epoch": 2.027027027027027,
      "grad_norm": 0.08215823024511337,
      "learning_rate": 1.1891891891891894e-05,
      "loss": 0.0587,
      "step": 300
    },
    {
      "epoch": 2.0945945945945947,
      "grad_norm": 0.08944419026374817,
      "learning_rate": 1.1621621621621622e-05,
      "loss": 0.0616,
      "step": 310
    },
    {
      "epoch": 2.1621621621621623,
      "grad_norm": 0.07855982333421707,
      "learning_rate": 1.1351351351351352e-05,
      "loss": 0.0139,
      "step": 320
    },
    {
      "epoch": 2.22972972972973,
      "grad_norm": 0.07730930298566818,
      "learning_rate": 1.1081081081081081e-05,
      "loss": 0.0461,
      "step": 330
    },
    {
      "epoch": 2.2972972972972974,
      "grad_norm": 0.0793551355600357,
      "learning_rate": 1.0810810810810812e-05,
      "loss": 0.0101,
      "step": 340
    },
    {
      "epoch": 2.364864864864865,
      "grad_norm": 0.07596410065889359,
      "learning_rate": 1.0540540540540541e-05,
      "loss": 0.01,
      "step": 350
    },
    {
      "epoch": 2.4324324324324325,
      "grad_norm": 0.08209327608346939,
      "learning_rate": 1.027027027027027e-05,
      "loss": 0.0147,
      "step": 360
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.07307254523038864,
      "learning_rate": 1e-05,
      "loss": 0.01,
      "step": 370
    },
    {
      "epoch": 2.5675675675675675,
      "grad_norm": 0.07255389541387558,
      "learning_rate": 9.729729729729732e-06,
      "loss": 0.0418,
      "step": 380
    },
    {
      "epoch": 2.635135135135135,
      "grad_norm": 0.08818744868040085,
      "learning_rate": 9.45945945945946e-06,
      "loss": 0.0703,
      "step": 390
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 0.0737389624118805,
      "learning_rate": 9.189189189189191e-06,
      "loss": 0.0279,
      "step": 400
    },
    {
      "epoch": 2.77027027027027,
      "grad_norm": 0.06960991024971008,
      "learning_rate": 8.91891891891892e-06,
      "loss": 0.009,
      "step": 410
    },
    {
      "epoch": 2.8378378378378377,
      "grad_norm": 0.06750474125146866,
      "learning_rate": 8.64864864864865e-06,
      "loss": 0.009,
      "step": 420
    },
    {
      "epoch": 2.9054054054054053,
      "grad_norm": 0.07102343440055847,
      "learning_rate": 8.378378378378378e-06,
      "loss": 0.0081,
      "step": 430
    },
    {
      "epoch": 2.972972972972973,
      "grad_norm": 0.06397751718759537,
      "learning_rate": 8.108108108108109e-06,
      "loss": 0.0112,
      "step": 440
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9856459330143541,
      "eval_loss": 0.038280732929706573,
      "eval_runtime": 27.6482,
      "eval_samples_per_second": 7.559,
      "eval_steps_per_second": 0.977,
      "step": 444
    },
    {
      "epoch": 3.0405405405405403,
      "grad_norm": 0.07215096056461334,
      "learning_rate": 7.837837837837838e-06,
      "loss": 0.0153,
      "step": 450
    },
    {
      "epoch": 3.108108108108108,
      "grad_norm": 0.06951985508203506,
      "learning_rate": 7.567567567567569e-06,
      "loss": 0.0085,
      "step": 460
    },
    {
      "epoch": 3.175675675675676,
      "grad_norm": 0.060418710112571716,
      "learning_rate": 7.297297297297298e-06,
      "loss": 0.0095,
      "step": 470
    },
    {
      "epoch": 3.2432432432432434,
      "grad_norm": 2.964932441711426,
      "learning_rate": 7.027027027027028e-06,
      "loss": 0.0109,
      "step": 480
    },
    {
      "epoch": 3.310810810810811,
      "grad_norm": 0.06342046707868576,
      "learning_rate": 6.7567567567567575e-06,
      "loss": 0.0088,
      "step": 490
    },
    {
      "epoch": 3.3783783783783785,
      "grad_norm": 0.06240825727581978,
      "learning_rate": 6.486486486486487e-06,
      "loss": 0.008,
      "step": 500
    },
    {
      "epoch": 3.445945945945946,
      "grad_norm": 0.06396786123514175,
      "learning_rate": 6.2162162162162164e-06,
      "loss": 0.0453,
      "step": 510
    },
    {
      "epoch": 3.5135135135135136,
      "grad_norm": 0.06505484133958817,
      "learning_rate": 5.945945945945947e-06,
      "loss": 0.0075,
      "step": 520
    },
    {
      "epoch": 3.581081081081081,
      "grad_norm": 0.057356029748916626,
      "learning_rate": 5.675675675675676e-06,
      "loss": 0.0239,
      "step": 530
    },
    {
      "epoch": 3.6486486486486487,
      "grad_norm": 0.0630515068769455,
      "learning_rate": 5.405405405405406e-06,
      "loss": 0.0169,
      "step": 540
    },
    {
      "epoch": 3.7162162162162162,
      "grad_norm": 0.06203653663396835,
      "learning_rate": 5.135135135135135e-06,
      "loss": 0.0162,
      "step": 550
    },
    {
      "epoch": 3.7837837837837838,
      "grad_norm": 0.06313777714967728,
      "learning_rate": 4.864864864864866e-06,
      "loss": 0.0072,
      "step": 560
    },
    {
      "epoch": 3.8513513513513513,
      "grad_norm": 0.06151194870471954,
      "learning_rate": 4.594594594594596e-06,
      "loss": 0.0108,
      "step": 570
    },
    {
      "epoch": 3.918918918918919,
      "grad_norm": 0.06091887503862381,
      "learning_rate": 4.324324324324325e-06,
      "loss": 0.1161,
      "step": 580
    },
    {
      "epoch": 3.9864864864864864,
      "grad_norm": 0.14475536346435547,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 0.01,
      "step": 590
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9952153110047847,
      "eval_loss": 0.0269942469894886,
      "eval_runtime": 27.6031,
      "eval_samples_per_second": 7.572,
      "eval_steps_per_second": 0.978,
      "step": 592
    },
    {
      "epoch": 4.054054054054054,
      "grad_norm": 0.056809261441230774,
      "learning_rate": 3.7837837837837844e-06,
      "loss": 0.0187,
      "step": 600
    },
    {
      "epoch": 4.121621621621622,
      "grad_norm": 0.056312572211027145,
      "learning_rate": 3.513513513513514e-06,
      "loss": 0.0073,
      "step": 610
    },
    {
      "epoch": 4.1891891891891895,
      "grad_norm": 0.08059559762477875,
      "learning_rate": 3.2432432432432437e-06,
      "loss": 0.0557,
      "step": 620
    },
    {
      "epoch": 4.256756756756757,
      "grad_norm": 0.065690778195858,
      "learning_rate": 2.9729729729729736e-06,
      "loss": 0.0076,
      "step": 630
    },
    {
      "epoch": 4.324324324324325,
      "grad_norm": 0.05522730574011803,
      "learning_rate": 2.702702702702703e-06,
      "loss": 0.0069,
      "step": 640
    },
    {
      "epoch": 4.391891891891892,
      "grad_norm": 0.054099809378385544,
      "learning_rate": 2.432432432432433e-06,
      "loss": 0.0082,
      "step": 650
    },
    {
      "epoch": 4.45945945945946,
      "grad_norm": 0.06278300285339355,
      "learning_rate": 2.1621621621621623e-06,
      "loss": 0.0105,
      "step": 660
    },
    {
      "epoch": 4.527027027027027,
      "grad_norm": 0.05770929157733917,
      "learning_rate": 1.8918918918918922e-06,
      "loss": 0.0071,
      "step": 670
    },
    {
      "epoch": 4.594594594594595,
      "grad_norm": 0.054900798946619034,
      "learning_rate": 1.6216216216216219e-06,
      "loss": 0.0087,
      "step": 680
    },
    {
      "epoch": 4.662162162162162,
      "grad_norm": 0.07473506033420563,
      "learning_rate": 1.3513513513513515e-06,
      "loss": 0.0577,
      "step": 690
    },
    {
      "epoch": 4.72972972972973,
      "grad_norm": 0.053288765251636505,
      "learning_rate": 1.0810810810810812e-06,
      "loss": 0.0069,
      "step": 700
    },
    {
      "epoch": 4.797297297297297,
      "grad_norm": 0.05479385703802109,
      "learning_rate": 8.108108108108109e-07,
      "loss": 0.0069,
      "step": 710
    },
    {
      "epoch": 4.864864864864865,
      "grad_norm": 6.568968772888184,
      "learning_rate": 5.405405405405406e-07,
      "loss": 0.0143,
      "step": 720
    },
    {
      "epoch": 4.9324324324324325,
      "grad_norm": 0.06765338033437729,
      "learning_rate": 2.702702702702703e-07,
      "loss": 0.0074,
      "step": 730
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0712118148803711,
      "learning_rate": 0.0,
      "loss": 0.0098,
      "step": 740
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9952153110047847,
      "eval_loss": 0.02545962668955326,
      "eval_runtime": 28.1075,
      "eval_samples_per_second": 7.436,
      "eval_steps_per_second": 0.961,
      "step": 740
    },
    {
      "epoch": 5.0,
      "step": 740,
      "total_flos": 4.583651185703117e+17,
      "train_loss": 0.044068760142938514,
      "train_runtime": 1744.9647,
      "train_samples_per_second": 3.39,
      "train_steps_per_second": 0.424
    }
  ],
  "logging_steps": 10,
  "max_steps": 740,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.583651185703117e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}