{
  "best_global_step": 2049,
  "best_metric": 0.6537335285505125,
  "best_model_checkpoint": "./saved_models/checkpoint-2049",
  "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": 1.8367172479629517,
      "learning_rate": 1.971303074670571e-05,
      "loss": 0.6893,
      "step": 50
    },
    {
      "epoch": 0.14641288433382138,
      "grad_norm": 2.717869758605957,
      "learning_rate": 1.942020497803807e-05,
      "loss": 0.6873,
      "step": 100
    },
    {
      "epoch": 0.21961932650073207,
      "grad_norm": 2.075798273086548,
      "learning_rate": 1.9127379209370426e-05,
      "loss": 0.6654,
      "step": 150
    },
    {
      "epoch": 0.29282576866764276,
      "grad_norm": 2.1151416301727295,
      "learning_rate": 1.8834553440702785e-05,
      "loss": 0.6675,
      "step": 200
    },
    {
      "epoch": 0.36603221083455345,
      "grad_norm": 5.988334655761719,
      "learning_rate": 1.854172767203514e-05,
      "loss": 0.6523,
      "step": 250
    },
    {
      "epoch": 0.43923865300146414,
      "grad_norm": 3.652282238006592,
      "learning_rate": 1.8248901903367496e-05,
      "loss": 0.6479,
      "step": 300
    },
    {
      "epoch": 0.5124450951683748,
      "grad_norm": 10.128896713256836,
      "learning_rate": 1.7956076134699855e-05,
      "loss": 0.6526,
      "step": 350
    },
    {
      "epoch": 0.5856515373352855,
      "grad_norm": 1.4224460124969482,
      "learning_rate": 1.766325036603221e-05,
      "loss": 0.6529,
      "step": 400
    },
    {
      "epoch": 0.6588579795021962,
      "grad_norm": 3.7824015617370605,
      "learning_rate": 1.737042459736457e-05,
      "loss": 0.65,
      "step": 450
    },
    {
      "epoch": 0.7320644216691069,
      "grad_norm": 2.6415672302246094,
      "learning_rate": 1.7077598828696925e-05,
      "loss": 0.6407,
      "step": 500
    },
    {
      "epoch": 0.8052708638360175,
      "grad_norm": 1.3568845987319946,
      "learning_rate": 1.6784773060029284e-05,
      "loss": 0.661,
      "step": 550
    },
    {
      "epoch": 0.8784773060029283,
      "grad_norm": 2.8408727645874023,
      "learning_rate": 1.649194729136164e-05,
      "loss": 0.6468,
      "step": 600
    },
    {
      "epoch": 0.9516837481698389,
      "grad_norm": 1.2629034519195557,
      "learning_rate": 1.6199121522694e-05,
      "loss": 0.634,
      "step": 650
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6255490483162518,
      "eval_loss": 0.6187466979026794,
      "eval_runtime": 13.089,
      "eval_samples_per_second": 208.726,
      "eval_steps_per_second": 3.285,
      "step": 683
    },
    {
      "epoch": 1.0248901903367496,
      "grad_norm": 2.3205177783966064,
      "learning_rate": 1.5906295754026355e-05,
      "loss": 0.6254,
      "step": 700
    },
    {
      "epoch": 1.0980966325036603,
      "grad_norm": 2.2244133949279785,
      "learning_rate": 1.5613469985358714e-05,
      "loss": 0.6299,
      "step": 750
    },
    {
      "epoch": 1.171303074670571,
      "grad_norm": 2.3779757022857666,
      "learning_rate": 1.532064421669107e-05,
      "loss": 0.6285,
      "step": 800
    },
    {
      "epoch": 1.2445095168374818,
      "grad_norm": 1.8427557945251465,
      "learning_rate": 1.5027818448023428e-05,
      "loss": 0.61,
      "step": 850
    },
    {
      "epoch": 1.3177159590043923,
      "grad_norm": 10.884477615356445,
      "learning_rate": 1.4734992679355784e-05,
      "loss": 0.6175,
      "step": 900
    },
    {
      "epoch": 1.390922401171303,
      "grad_norm": 5.921830177307129,
      "learning_rate": 1.4442166910688143e-05,
      "loss": 0.61,
      "step": 950
    },
    {
      "epoch": 1.4641288433382138,
      "grad_norm": 5.03695821762085,
      "learning_rate": 1.4149341142020499e-05,
      "loss": 0.6033,
      "step": 1000
    },
    {
      "epoch": 1.5373352855051245,
      "grad_norm": 2.7754528522491455,
      "learning_rate": 1.3856515373352856e-05,
      "loss": 0.6001,
      "step": 1050
    },
    {
      "epoch": 1.610541727672035,
      "grad_norm": 3.685950517654419,
      "learning_rate": 1.3563689604685213e-05,
      "loss": 0.6234,
      "step": 1100
    },
    {
      "epoch": 1.6837481698389458,
      "grad_norm": 3.515249252319336,
      "learning_rate": 1.327086383601757e-05,
      "loss": 0.6111,
      "step": 1150
    },
    {
      "epoch": 1.7569546120058566,
      "grad_norm": 3.3894221782684326,
      "learning_rate": 1.2978038067349928e-05,
      "loss": 0.5906,
      "step": 1200
    },
    {
      "epoch": 1.830161054172767,
      "grad_norm": 4.961050510406494,
      "learning_rate": 1.2685212298682286e-05,
      "loss": 0.614,
      "step": 1250
    },
    {
      "epoch": 1.903367496339678,
      "grad_norm": 2.3843491077423096,
      "learning_rate": 1.2392386530014641e-05,
      "loss": 0.6117,
      "step": 1300
    },
    {
      "epoch": 1.9765739385065886,
      "grad_norm": 2.979031801223755,
      "learning_rate": 1.2099560761347e-05,
      "loss": 0.5859,
      "step": 1350
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6361639824304539,
      "eval_loss": 0.5908383131027222,
      "eval_runtime": 12.8649,
      "eval_samples_per_second": 212.361,
      "eval_steps_per_second": 3.342,
      "step": 1366
    },
    {
      "epoch": 2.049780380673499,
      "grad_norm": 5.897040367126465,
      "learning_rate": 1.1806734992679356e-05,
      "loss": 0.573,
      "step": 1400
    },
    {
      "epoch": 2.12298682284041,
      "grad_norm": 5.25750207901001,
      "learning_rate": 1.1513909224011715e-05,
      "loss": 0.5455,
      "step": 1450
    },
    {
      "epoch": 2.1961932650073206,
      "grad_norm": 14.578705787658691,
      "learning_rate": 1.122108345534407e-05,
      "loss": 0.5673,
      "step": 1500
    },
    {
      "epoch": 2.269399707174231,
      "grad_norm": 3.079087018966675,
      "learning_rate": 1.092825768667643e-05,
      "loss": 0.5545,
      "step": 1550
    },
    {
      "epoch": 2.342606149341142,
      "grad_norm": 3.9325766563415527,
      "learning_rate": 1.0635431918008785e-05,
      "loss": 0.5688,
      "step": 1600
    },
    {
      "epoch": 2.4158125915080526,
      "grad_norm": 5.2294087409973145,
      "learning_rate": 1.0342606149341143e-05,
      "loss": 0.5614,
      "step": 1650
    },
    {
      "epoch": 2.4890190336749636,
      "grad_norm": 5.683165073394775,
      "learning_rate": 1.00497803806735e-05,
      "loss": 0.5549,
      "step": 1700
    },
    {
      "epoch": 2.562225475841874,
      "grad_norm": 4.106119155883789,
      "learning_rate": 9.756954612005857e-06,
      "loss": 0.5565,
      "step": 1750
    },
    {
      "epoch": 2.6354319180087846,
      "grad_norm": 3.8661065101623535,
      "learning_rate": 9.464128843338215e-06,
      "loss": 0.5489,
      "step": 1800
    },
    {
      "epoch": 2.7086383601756956,
      "grad_norm": 12.783368110656738,
      "learning_rate": 9.171303074670572e-06,
      "loss": 0.5472,
      "step": 1850
    },
    {
      "epoch": 2.781844802342606,
      "grad_norm": 17.30908966064453,
      "learning_rate": 8.87847730600293e-06,
      "loss": 0.5696,
      "step": 1900
    },
    {
      "epoch": 2.855051244509517,
      "grad_norm": 13.78253173828125,
      "learning_rate": 8.585651537335287e-06,
      "loss": 0.5289,
      "step": 1950
    },
    {
      "epoch": 2.9282576866764276,
      "grad_norm": 11.748068809509277,
      "learning_rate": 8.292825768667644e-06,
      "loss": 0.5487,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6537335285505125,
      "eval_loss": 0.6090347170829773,
      "eval_runtime": 13.1904,
      "eval_samples_per_second": 207.12,
      "eval_steps_per_second": 3.26,
      "step": 2049
    },
    {
      "epoch": 3.001464128843338,
      "grad_norm": 4.5922698974609375,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.5447,
      "step": 2050
    },
    {
      "epoch": 3.074670571010249,
      "grad_norm": 5.039747714996338,
      "learning_rate": 7.707174231332359e-06,
      "loss": 0.5025,
      "step": 2100
    },
    {
      "epoch": 3.1478770131771596,
      "grad_norm": 7.7539825439453125,
      "learning_rate": 7.414348462664715e-06,
      "loss": 0.4964,
      "step": 2150
    },
    {
      "epoch": 3.22108345534407,
      "grad_norm": 7.477388381958008,
      "learning_rate": 7.1215226939970725e-06,
      "loss": 0.5053,
      "step": 2200
    },
    {
      "epoch": 3.294289897510981,
      "grad_norm": 11.205092430114746,
      "learning_rate": 6.82869692532943e-06,
      "loss": 0.487,
      "step": 2250
    },
    {
      "epoch": 3.3674963396778916,
      "grad_norm": 5.962555885314941,
      "learning_rate": 6.535871156661787e-06,
      "loss": 0.4947,
      "step": 2300
    },
    {
      "epoch": 3.440702781844802,
      "grad_norm": 9.716086387634277,
      "learning_rate": 6.2430453879941446e-06,
      "loss": 0.5043,
      "step": 2350
    },
    {
      "epoch": 3.513909224011713,
      "grad_norm": 7.83236026763916,
      "learning_rate": 5.950219619326502e-06,
      "loss": 0.5083,
      "step": 2400
    },
    {
      "epoch": 3.5871156661786237,
      "grad_norm": 7.700263500213623,
      "learning_rate": 5.657393850658858e-06,
      "loss": 0.5099,
      "step": 2450
    },
    {
      "epoch": 3.660322108345534,
      "grad_norm": 6.1209540367126465,
      "learning_rate": 5.364568081991216e-06,
      "loss": 0.5031,
      "step": 2500
    },
    {
      "epoch": 3.733528550512445,
      "grad_norm": 6.6275858879089355,
      "learning_rate": 5.071742313323573e-06,
      "loss": 0.5111,
      "step": 2550
    },
    {
      "epoch": 3.8067349926793557,
      "grad_norm": 11.429731369018555,
      "learning_rate": 4.77891654465593e-06,
      "loss": 0.4936,
      "step": 2600
    },
    {
      "epoch": 3.8799414348462666,
      "grad_norm": 13.412334442138672,
      "learning_rate": 4.486090775988287e-06,
      "loss": 0.5202,
      "step": 2650
    },
    {
      "epoch": 3.953147877013177,
      "grad_norm": 6.625152111053467,
      "learning_rate": 4.193265007320644e-06,
      "loss": 0.4826,
      "step": 2700
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6379941434846267,
      "eval_loss": 0.6155059933662415,
      "eval_runtime": 13.2833,
      "eval_samples_per_second": 205.672,
      "eval_steps_per_second": 3.237,
      "step": 2732
    },
    {
      "epoch": 4.026354319180088,
      "grad_norm": 7.482672214508057,
      "learning_rate": 3.900439238653002e-06,
      "loss": 0.4992,
      "step": 2750
    },
    {
      "epoch": 4.099560761346998,
      "grad_norm": 9.46927261352539,
      "learning_rate": 3.607613469985359e-06,
      "loss": 0.4473,
      "step": 2800
    },
    {
      "epoch": 4.172767203513909,
      "grad_norm": 8.915600776672363,
      "learning_rate": 3.314787701317716e-06,
      "loss": 0.4468,
      "step": 2850
    },
    {
      "epoch": 4.24597364568082,
      "grad_norm": 10.110650062561035,
      "learning_rate": 3.0219619326500732e-06,
      "loss": 0.4692,
      "step": 2900
    },
    {
      "epoch": 4.31918008784773,
      "grad_norm": 17.987337112426758,
      "learning_rate": 2.7291361639824306e-06,
      "loss": 0.4452,
      "step": 2950
    },
    {
      "epoch": 4.392386530014641,
      "grad_norm": 7.12765645980835,
      "learning_rate": 2.436310395314788e-06,
      "loss": 0.4401,
      "step": 3000
    },
    {
      "epoch": 4.465592972181552,
      "grad_norm": 6.777918815612793,
      "learning_rate": 2.1434846266471453e-06,
      "loss": 0.4511,
      "step": 3050
    },
    {
      "epoch": 4.538799414348462,
      "grad_norm": 15.6901273727417,
      "learning_rate": 1.8506588579795024e-06,
      "loss": 0.4427,
      "step": 3100
    },
    {
      "epoch": 4.612005856515373,
      "grad_norm": 9.44357967376709,
      "learning_rate": 1.5578330893118595e-06,
      "loss": 0.4564,
      "step": 3150
    },
    {
      "epoch": 4.685212298682284,
      "grad_norm": 7.72824239730835,
      "learning_rate": 1.2650073206442169e-06,
      "loss": 0.4319,
      "step": 3200
    },
    {
      "epoch": 4.758418740849194,
      "grad_norm": 14.447629928588867,
      "learning_rate": 9.72181551976574e-07,
      "loss": 0.4558,
      "step": 3250
    },
    {
      "epoch": 4.831625183016105,
      "grad_norm": 7.083517551422119,
      "learning_rate": 6.793557833089313e-07,
      "loss": 0.4537,
      "step": 3300
    },
    {
      "epoch": 4.904831625183016,
      "grad_norm": 15.930305480957031,
      "learning_rate": 3.865300146412885e-07,
      "loss": 0.4552,
      "step": 3350
    },
    {
      "epoch": 4.978038067349927,
      "grad_norm": 15.451827049255371,
      "learning_rate": 9.370424597364569e-08,
      "loss": 0.456,
      "step": 3400
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6471449487554904,
      "eval_loss": 0.6813626289367676,
      "eval_runtime": 12.97,
      "eval_samples_per_second": 210.639,
      "eval_steps_per_second": 3.315,
      "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
}