{
  "best_global_step": 3415,
  "best_metric": 0.6295754026354319,
  "best_model_checkpoint": "./saved_models/checkpoint-3415",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 3415,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07320644216691069,
      "grad_norm": 1.5678623914718628,
      "learning_rate": 1.971303074670571e-05,
      "loss": 0.6936,
      "step": 50
    },
    {
      "epoch": 0.14641288433382138,
      "grad_norm": 1.88246488571167,
      "learning_rate": 1.942020497803807e-05,
      "loss": 0.6949,
      "step": 100
    },
    {
      "epoch": 0.21961932650073207,
      "grad_norm": 1.9367057085037231,
      "learning_rate": 1.9127379209370426e-05,
      "loss": 0.6881,
      "step": 150
    },
    {
      "epoch": 0.29282576866764276,
      "grad_norm": 1.304018497467041,
      "learning_rate": 1.8834553440702785e-05,
      "loss": 0.6898,
      "step": 200
    },
    {
      "epoch": 0.36603221083455345,
      "grad_norm": 3.572559356689453,
      "learning_rate": 1.854172767203514e-05,
      "loss": 0.6835,
      "step": 250
    },
    {
      "epoch": 0.43923865300146414,
      "grad_norm": 2.7547812461853027,
      "learning_rate": 1.8248901903367496e-05,
      "loss": 0.6758,
      "step": 300
    },
    {
      "epoch": 0.5124450951683748,
      "grad_norm": 5.36733341217041,
      "learning_rate": 1.7956076134699855e-05,
      "loss": 0.6763,
      "step": 350
    },
    {
      "epoch": 0.5856515373352855,
      "grad_norm": 2.518483877182007,
      "learning_rate": 1.766325036603221e-05,
      "loss": 0.6794,
      "step": 400
    },
    {
      "epoch": 0.6588579795021962,
      "grad_norm": 2.1962575912475586,
      "learning_rate": 1.737042459736457e-05,
      "loss": 0.6845,
      "step": 450
    },
    {
      "epoch": 0.7320644216691069,
      "grad_norm": 3.3848378658294678,
      "learning_rate": 1.7077598828696925e-05,
      "loss": 0.6792,
      "step": 500
    },
    {
      "epoch": 0.8052708638360175,
      "grad_norm": 0.9908949136734009,
      "learning_rate": 1.6784773060029284e-05,
      "loss": 0.6905,
      "step": 550
    },
    {
      "epoch": 0.8784773060029283,
      "grad_norm": 1.581642508506775,
      "learning_rate": 1.649194729136164e-05,
      "loss": 0.6801,
      "step": 600
    },
    {
      "epoch": 0.9516837481698389,
      "grad_norm": 3.641871213912964,
      "learning_rate": 1.6199121522694e-05,
      "loss": 0.6825,
      "step": 650
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6010248901903368,
      "eval_loss": 0.6722473502159119,
      "eval_runtime": 3.5824,
      "eval_samples_per_second": 762.619,
      "eval_steps_per_second": 12.003,
      "step": 683
    },
    {
      "epoch": 1.0248901903367496,
      "grad_norm": 1.8619815111160278,
      "learning_rate": 1.5906295754026355e-05,
      "loss": 0.6704,
      "step": 700
    },
    {
      "epoch": 1.0980966325036603,
      "grad_norm": 0.7537197470664978,
      "learning_rate": 1.5613469985358714e-05,
      "loss": 0.6758,
      "step": 750
    },
    {
      "epoch": 1.171303074670571,
      "grad_norm": 0.7405994534492493,
      "learning_rate": 1.532064421669107e-05,
      "loss": 0.6682,
      "step": 800
    },
    {
      "epoch": 1.2445095168374818,
      "grad_norm": 1.4860440492630005,
      "learning_rate": 1.5027818448023428e-05,
      "loss": 0.6664,
      "step": 850
    },
    {
      "epoch": 1.3177159590043923,
      "grad_norm": 1.5564947128295898,
      "learning_rate": 1.4734992679355784e-05,
      "loss": 0.6696,
      "step": 900
    },
    {
      "epoch": 1.390922401171303,
      "grad_norm": 2.0371503829956055,
      "learning_rate": 1.4442166910688143e-05,
      "loss": 0.6679,
      "step": 950
    },
    {
      "epoch": 1.4641288433382138,
      "grad_norm": 1.321709394454956,
      "learning_rate": 1.4149341142020499e-05,
      "loss": 0.6589,
      "step": 1000
    },
    {
      "epoch": 1.5373352855051245,
      "grad_norm": 2.6580564975738525,
      "learning_rate": 1.3856515373352856e-05,
      "loss": 0.6577,
      "step": 1050
    },
    {
      "epoch": 1.610541727672035,
      "grad_norm": 1.3632615804672241,
      "learning_rate": 1.3563689604685213e-05,
      "loss": 0.6773,
      "step": 1100
    },
    {
      "epoch": 1.6837481698389458,
      "grad_norm": 1.320080280303955,
      "learning_rate": 1.327086383601757e-05,
      "loss": 0.6573,
      "step": 1150
    },
    {
      "epoch": 1.7569546120058566,
      "grad_norm": 2.9068989753723145,
      "learning_rate": 1.2978038067349928e-05,
      "loss": 0.6548,
      "step": 1200
    },
    {
      "epoch": 1.830161054172767,
      "grad_norm": 0.9685165286064148,
      "learning_rate": 1.2685212298682286e-05,
      "loss": 0.6589,
      "step": 1250
    },
    {
      "epoch": 1.903367496339678,
      "grad_norm": 0.8692395687103271,
      "learning_rate": 1.2392386530014641e-05,
      "loss": 0.6604,
      "step": 1300
    },
    {
      "epoch": 1.9765739385065886,
      "grad_norm": 0.9666163325309753,
      "learning_rate": 1.2099560761347e-05,
      "loss": 0.6634,
      "step": 1350
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5647877013177159,
      "eval_loss": 0.6470807790756226,
      "eval_runtime": 3.57,
      "eval_samples_per_second": 765.268,
      "eval_steps_per_second": 12.045,
      "step": 1366
    },
    {
      "epoch": 2.049780380673499,
      "grad_norm": 6.759352207183838,
      "learning_rate": 1.1806734992679356e-05,
      "loss": 0.6592,
      "step": 1400
    },
    {
      "epoch": 2.12298682284041,
      "grad_norm": 2.8715744018554688,
      "learning_rate": 1.1513909224011715e-05,
      "loss": 0.6445,
      "step": 1450
    },
    {
      "epoch": 2.1961932650073206,
      "grad_norm": 4.17264461517334,
      "learning_rate": 1.122108345534407e-05,
      "loss": 0.6499,
      "step": 1500
    },
    {
      "epoch": 2.269399707174231,
      "grad_norm": 1.707520842552185,
      "learning_rate": 1.092825768667643e-05,
      "loss": 0.649,
      "step": 1550
    },
    {
      "epoch": 2.342606149341142,
      "grad_norm": 1.4020594358444214,
      "learning_rate": 1.0635431918008785e-05,
      "loss": 0.6468,
      "step": 1600
    },
    {
      "epoch": 2.4158125915080526,
      "grad_norm": 1.690818190574646,
      "learning_rate": 1.0342606149341143e-05,
      "loss": 0.6524,
      "step": 1650
    },
    {
      "epoch": 2.4890190336749636,
      "grad_norm": 2.177947521209717,
      "learning_rate": 1.00497803806735e-05,
      "loss": 0.6443,
      "step": 1700
    },
    {
      "epoch": 2.562225475841874,
      "grad_norm": 2.009894847869873,
      "learning_rate": 9.756954612005857e-06,
      "loss": 0.6491,
      "step": 1750
    },
    {
      "epoch": 2.6354319180087846,
      "grad_norm": 1.707280158996582,
      "learning_rate": 9.464128843338215e-06,
      "loss": 0.6364,
      "step": 1800
    },
    {
      "epoch": 2.7086383601756956,
      "grad_norm": 6.970835208892822,
      "learning_rate": 9.171303074670572e-06,
      "loss": 0.6442,
      "step": 1850
    },
    {
      "epoch": 2.781844802342606,
      "grad_norm": 1.682925820350647,
      "learning_rate": 8.87847730600293e-06,
      "loss": 0.6399,
      "step": 1900
    },
    {
      "epoch": 2.855051244509517,
      "grad_norm": 3.518610954284668,
      "learning_rate": 8.585651537335287e-06,
      "loss": 0.6423,
      "step": 1950
    },
    {
      "epoch": 2.9282576866764276,
      "grad_norm": 3.799734115600586,
      "learning_rate": 8.292825768667644e-06,
      "loss": 0.6385,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.623718887262079,
      "eval_loss": 0.6276178359985352,
      "eval_runtime": 3.5711,
      "eval_samples_per_second": 765.036,
      "eval_steps_per_second": 12.041,
      "step": 2049
    },
    {
      "epoch": 3.001464128843338,
      "grad_norm": 2.7766103744506836,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.651,
      "step": 2050
    },
    {
      "epoch": 3.074670571010249,
      "grad_norm": 4.394078731536865,
      "learning_rate": 7.707174231332359e-06,
      "loss": 0.6292,
      "step": 2100
    },
    {
      "epoch": 3.1478770131771596,
      "grad_norm": 1.7268424034118652,
      "learning_rate": 7.414348462664715e-06,
      "loss": 0.6345,
      "step": 2150
    },
    {
      "epoch": 3.22108345534407,
      "grad_norm": 1.5889354944229126,
      "learning_rate": 7.1215226939970725e-06,
      "loss": 0.6218,
      "step": 2200
    },
    {
      "epoch": 3.294289897510981,
      "grad_norm": 6.7957658767700195,
      "learning_rate": 6.82869692532943e-06,
      "loss": 0.6317,
      "step": 2250
    },
    {
      "epoch": 3.3674963396778916,
      "grad_norm": 2.915957450866699,
      "learning_rate": 6.535871156661787e-06,
      "loss": 0.6147,
      "step": 2300
    },
    {
      "epoch": 3.440702781844802,
      "grad_norm": 2.6085119247436523,
      "learning_rate": 6.2430453879941446e-06,
      "loss": 0.6226,
      "step": 2350
    },
    {
      "epoch": 3.513909224011713,
      "grad_norm": 2.2643730640411377,
      "learning_rate": 5.950219619326502e-06,
      "loss": 0.616,
      "step": 2400
    },
    {
      "epoch": 3.5871156661786237,
      "grad_norm": 2.4427967071533203,
      "learning_rate": 5.657393850658858e-06,
      "loss": 0.6235,
      "step": 2450
    },
    {
      "epoch": 3.660322108345534,
      "grad_norm": 3.067891836166382,
      "learning_rate": 5.364568081991216e-06,
      "loss": 0.6098,
      "step": 2500
    },
    {
      "epoch": 3.733528550512445,
      "grad_norm": 5.014751434326172,
      "learning_rate": 5.071742313323573e-06,
      "loss": 0.6162,
      "step": 2550
    },
    {
      "epoch": 3.8067349926793557,
      "grad_norm": 6.996451377868652,
      "learning_rate": 4.77891654465593e-06,
      "loss": 0.6057,
      "step": 2600
    },
    {
      "epoch": 3.8799414348462666,
      "grad_norm": 8.236844062805176,
      "learning_rate": 4.486090775988287e-06,
      "loss": 0.6065,
      "step": 2650
    },
    {
      "epoch": 3.953147877013177,
      "grad_norm": 2.6412973403930664,
      "learning_rate": 4.193265007320644e-06,
      "loss": 0.6071,
      "step": 2700
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5867496339677891,
      "eval_loss": 0.647341251373291,
      "eval_runtime": 3.5811,
      "eval_samples_per_second": 762.894,
      "eval_steps_per_second": 12.007,
      "step": 2732
    },
    {
      "epoch": 4.026354319180088,
      "grad_norm": 7.569336414337158,
      "learning_rate": 3.900439238653002e-06,
      "loss": 0.6059,
      "step": 2750
    },
    {
      "epoch": 4.099560761346998,
      "grad_norm": 4.824645519256592,
      "learning_rate": 3.607613469985359e-06,
      "loss": 0.6003,
      "step": 2800
    },
    {
      "epoch": 4.172767203513909,
      "grad_norm": 5.947188854217529,
      "learning_rate": 3.314787701317716e-06,
      "loss": 0.6082,
      "step": 2850
    },
    {
      "epoch": 4.24597364568082,
      "grad_norm": 3.865933895111084,
      "learning_rate": 3.0219619326500732e-06,
      "loss": 0.5993,
      "step": 2900
    },
    {
      "epoch": 4.31918008784773,
      "grad_norm": 9.159831047058105,
      "learning_rate": 2.7291361639824306e-06,
      "loss": 0.5923,
      "step": 2950
    },
    {
      "epoch": 4.392386530014641,
      "grad_norm": 6.4656782150268555,
      "learning_rate": 2.436310395314788e-06,
      "loss": 0.5715,
      "step": 3000
    },
    {
      "epoch": 4.465592972181552,
      "grad_norm": 3.221506118774414,
      "learning_rate": 2.1434846266471453e-06,
      "loss": 0.5921,
      "step": 3050
    },
    {
      "epoch": 4.538799414348462,
      "grad_norm": 3.973538398742676,
      "learning_rate": 1.8506588579795024e-06,
      "loss": 0.5866,
      "step": 3100
    },
    {
      "epoch": 4.612005856515373,
      "grad_norm": 3.1635329723358154,
      "learning_rate": 1.5578330893118595e-06,
      "loss": 0.5837,
      "step": 3150
    },
    {
      "epoch": 4.685212298682284,
      "grad_norm": 3.28462815284729,
      "learning_rate": 1.2650073206442169e-06,
      "loss": 0.5954,
      "step": 3200
    },
    {
      "epoch": 4.758418740849194,
      "grad_norm": 6.968575954437256,
      "learning_rate": 9.72181551976574e-07,
      "loss": 0.5951,
      "step": 3250
    },
    {
      "epoch": 4.831625183016105,
      "grad_norm": 3.1862399578094482,
      "learning_rate": 6.793557833089313e-07,
      "loss": 0.5928,
      "step": 3300
    },
    {
      "epoch": 4.904831625183016,
      "grad_norm": 9.21243953704834,
      "learning_rate": 3.865300146412885e-07,
      "loss": 0.5966,
      "step": 3350
    },
    {
      "epoch": 4.978038067349927,
      "grad_norm": 8.415995597839355,
      "learning_rate": 9.370424597364569e-08,
      "loss": 0.5912,
      "step": 3400
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6295754026354319,
      "eval_loss": 0.6141241192817688,
      "eval_runtime": 3.5919,
      "eval_samples_per_second": 760.595,
      "eval_steps_per_second": 11.971,
      "step": 3415
    }
  ],
  "logging_steps": 50,
  "max_steps": 3415,
  "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": 1.43750725095936e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}