{
  "best_global_step": 3415,
  "best_metric": 0.6445827232796486,
  "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": 9.181918144226074,
      "learning_rate": 1.971303074670571e-05,
      "loss": 0.8158,
      "step": 50
    },
    {
      "epoch": 0.14641288433382138,
      "grad_norm": 10.360370635986328,
      "learning_rate": 1.942020497803807e-05,
      "loss": 0.7489,
      "step": 100
    },
    {
      "epoch": 0.21961932650073207,
      "grad_norm": 5.784631252288818,
      "learning_rate": 1.9127379209370426e-05,
      "loss": 0.7137,
      "step": 150
    },
    {
      "epoch": 0.29282576866764276,
      "grad_norm": 6.866419315338135,
      "learning_rate": 1.8834553440702785e-05,
      "loss": 0.7168,
      "step": 200
    },
    {
      "epoch": 0.36603221083455345,
      "grad_norm": 12.543447494506836,
      "learning_rate": 1.854172767203514e-05,
      "loss": 0.7341,
      "step": 250
    },
    {
      "epoch": 0.43923865300146414,
      "grad_norm": 5.011623382568359,
      "learning_rate": 1.8248901903367496e-05,
      "loss": 0.7068,
      "step": 300
    },
    {
      "epoch": 0.5124450951683748,
      "grad_norm": 8.569839477539062,
      "learning_rate": 1.7956076134699855e-05,
      "loss": 0.702,
      "step": 350
    },
    {
      "epoch": 0.5856515373352855,
      "grad_norm": 4.254383087158203,
      "learning_rate": 1.766325036603221e-05,
      "loss": 0.6995,
      "step": 400
    },
    {
      "epoch": 0.6588579795021962,
      "grad_norm": 5.654411792755127,
      "learning_rate": 1.737042459736457e-05,
      "loss": 0.6922,
      "step": 450
    },
    {
      "epoch": 0.7320644216691069,
      "grad_norm": 4.375262260437012,
      "learning_rate": 1.7077598828696925e-05,
      "loss": 0.6808,
      "step": 500
    },
    {
      "epoch": 0.8052708638360175,
      "grad_norm": 4.428417205810547,
      "learning_rate": 1.6784773060029284e-05,
      "loss": 0.7041,
      "step": 550
    },
    {
      "epoch": 0.8784773060029283,
      "grad_norm": 7.5497612953186035,
      "learning_rate": 1.649194729136164e-05,
      "loss": 0.6856,
      "step": 600
    },
    {
      "epoch": 0.9516837481698389,
      "grad_norm": 6.1082658767700195,
      "learning_rate": 1.6199121522694e-05,
      "loss": 0.6843,
      "step": 650
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5728404099560761,
      "eval_loss": 0.6616985201835632,
      "eval_runtime": 9.1365,
      "eval_samples_per_second": 299.02,
      "eval_steps_per_second": 4.706,
      "step": 683
    },
    {
      "epoch": 1.0248901903367496,
      "grad_norm": 7.093502521514893,
      "learning_rate": 1.5906295754026355e-05,
      "loss": 0.6763,
      "step": 700
    },
    {
      "epoch": 1.0980966325036603,
      "grad_norm": 7.498149394989014,
      "learning_rate": 1.5613469985358714e-05,
      "loss": 0.6657,
      "step": 750
    },
    {
      "epoch": 1.171303074670571,
      "grad_norm": 4.592875957489014,
      "learning_rate": 1.532064421669107e-05,
      "loss": 0.6725,
      "step": 800
    },
    {
      "epoch": 1.2445095168374818,
      "grad_norm": 3.1138663291931152,
      "learning_rate": 1.5027818448023428e-05,
      "loss": 0.6793,
      "step": 850
    },
    {
      "epoch": 1.3177159590043923,
      "grad_norm": 5.978513240814209,
      "learning_rate": 1.4734992679355784e-05,
      "loss": 0.6698,
      "step": 900
    },
    {
      "epoch": 1.390922401171303,
      "grad_norm": 7.4365339279174805,
      "learning_rate": 1.4442166910688143e-05,
      "loss": 0.6803,
      "step": 950
    },
    {
      "epoch": 1.4641288433382138,
      "grad_norm": 4.018332004547119,
      "learning_rate": 1.4149341142020499e-05,
      "loss": 0.6619,
      "step": 1000
    },
    {
      "epoch": 1.5373352855051245,
      "grad_norm": 3.3087430000305176,
      "learning_rate": 1.3856515373352856e-05,
      "loss": 0.6647,
      "step": 1050
    },
    {
      "epoch": 1.610541727672035,
      "grad_norm": 3.5859272480010986,
      "learning_rate": 1.3563689604685213e-05,
      "loss": 0.6641,
      "step": 1100
    },
    {
      "epoch": 1.6837481698389458,
      "grad_norm": 3.992220163345337,
      "learning_rate": 1.327086383601757e-05,
      "loss": 0.6625,
      "step": 1150
    },
    {
      "epoch": 1.7569546120058566,
      "grad_norm": 5.505192756652832,
      "learning_rate": 1.2978038067349928e-05,
      "loss": 0.653,
      "step": 1200
    },
    {
      "epoch": 1.830161054172767,
      "grad_norm": 3.4128470420837402,
      "learning_rate": 1.2685212298682286e-05,
      "loss": 0.6467,
      "step": 1250
    },
    {
      "epoch": 1.903367496339678,
      "grad_norm": 3.6012327671051025,
      "learning_rate": 1.2392386530014641e-05,
      "loss": 0.6585,
      "step": 1300
    },
    {
      "epoch": 1.9765739385065886,
      "grad_norm": 2.7060303688049316,
      "learning_rate": 1.2099560761347e-05,
      "loss": 0.6403,
      "step": 1350
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5904099560761347,
      "eval_loss": 0.6402733325958252,
      "eval_runtime": 9.1043,
      "eval_samples_per_second": 300.077,
      "eval_steps_per_second": 4.723,
      "step": 1366
    },
    {
      "epoch": 2.049780380673499,
      "grad_norm": 8.067757606506348,
      "learning_rate": 1.1806734992679356e-05,
      "loss": 0.641,
      "step": 1400
    },
    {
      "epoch": 2.12298682284041,
      "grad_norm": 6.0310540199279785,
      "learning_rate": 1.1513909224011715e-05,
      "loss": 0.639,
      "step": 1450
    },
    {
      "epoch": 2.1961932650073206,
      "grad_norm": 8.116523742675781,
      "learning_rate": 1.122108345534407e-05,
      "loss": 0.6338,
      "step": 1500
    },
    {
      "epoch": 2.269399707174231,
      "grad_norm": 2.589458703994751,
      "learning_rate": 1.092825768667643e-05,
      "loss": 0.6359,
      "step": 1550
    },
    {
      "epoch": 2.342606149341142,
      "grad_norm": 3.3060238361358643,
      "learning_rate": 1.0635431918008785e-05,
      "loss": 0.6387,
      "step": 1600
    },
    {
      "epoch": 2.4158125915080526,
      "grad_norm": 4.6258134841918945,
      "learning_rate": 1.0342606149341143e-05,
      "loss": 0.6279,
      "step": 1650
    },
    {
      "epoch": 2.4890190336749636,
      "grad_norm": 5.477193832397461,
      "learning_rate": 1.00497803806735e-05,
      "loss": 0.6455,
      "step": 1700
    },
    {
      "epoch": 2.562225475841874,
      "grad_norm": 4.201015949249268,
      "learning_rate": 9.756954612005857e-06,
      "loss": 0.6321,
      "step": 1750
    },
    {
      "epoch": 2.6354319180087846,
      "grad_norm": 4.213687896728516,
      "learning_rate": 9.464128843338215e-06,
      "loss": 0.6319,
      "step": 1800
    },
    {
      "epoch": 2.7086383601756956,
      "grad_norm": 15.101090431213379,
      "learning_rate": 9.171303074670572e-06,
      "loss": 0.6306,
      "step": 1850
    },
    {
      "epoch": 2.781844802342606,
      "grad_norm": 3.422281265258789,
      "learning_rate": 8.87847730600293e-06,
      "loss": 0.6436,
      "step": 1900
    },
    {
      "epoch": 2.855051244509517,
      "grad_norm": 8.450922012329102,
      "learning_rate": 8.585651537335287e-06,
      "loss": 0.6355,
      "step": 1950
    },
    {
      "epoch": 2.9282576866764276,
      "grad_norm": 7.880731105804443,
      "learning_rate": 8.292825768667644e-06,
      "loss": 0.6226,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6284773060029283,
      "eval_loss": 0.6120732426643372,
      "eval_runtime": 9.1117,
      "eval_samples_per_second": 299.834,
      "eval_steps_per_second": 4.719,
      "step": 2049
    },
    {
      "epoch": 3.001464128843338,
      "grad_norm": 3.401750326156616,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.6201,
      "step": 2050
    },
    {
      "epoch": 3.074670571010249,
      "grad_norm": 3.4573326110839844,
      "learning_rate": 7.707174231332359e-06,
      "loss": 0.6152,
      "step": 2100
    },
    {
      "epoch": 3.1478770131771596,
      "grad_norm": 3.4153525829315186,
      "learning_rate": 7.414348462664715e-06,
      "loss": 0.6124,
      "step": 2150
    },
    {
      "epoch": 3.22108345534407,
      "grad_norm": 3.386751651763916,
      "learning_rate": 7.1215226939970725e-06,
      "loss": 0.6098,
      "step": 2200
    },
    {
      "epoch": 3.294289897510981,
      "grad_norm": 8.25568962097168,
      "learning_rate": 6.82869692532943e-06,
      "loss": 0.6131,
      "step": 2250
    },
    {
      "epoch": 3.3674963396778916,
      "grad_norm": 3.146071195602417,
      "learning_rate": 6.535871156661787e-06,
      "loss": 0.6109,
      "step": 2300
    },
    {
      "epoch": 3.440702781844802,
      "grad_norm": 3.564626455307007,
      "learning_rate": 6.2430453879941446e-06,
      "loss": 0.6024,
      "step": 2350
    },
    {
      "epoch": 3.513909224011713,
      "grad_norm": 4.476875305175781,
      "learning_rate": 5.950219619326502e-06,
      "loss": 0.6121,
      "step": 2400
    },
    {
      "epoch": 3.5871156661786237,
      "grad_norm": 4.083089828491211,
      "learning_rate": 5.657393850658858e-06,
      "loss": 0.6155,
      "step": 2450
    },
    {
      "epoch": 3.660322108345534,
      "grad_norm": 4.565993785858154,
      "learning_rate": 5.364568081991216e-06,
      "loss": 0.6037,
      "step": 2500
    },
    {
      "epoch": 3.733528550512445,
      "grad_norm": 3.9050488471984863,
      "learning_rate": 5.071742313323573e-06,
      "loss": 0.6008,
      "step": 2550
    },
    {
      "epoch": 3.8067349926793557,
      "grad_norm": 3.331878185272217,
      "learning_rate": 4.77891654465593e-06,
      "loss": 0.5896,
      "step": 2600
    },
    {
      "epoch": 3.8799414348462666,
      "grad_norm": 7.776970863342285,
      "learning_rate": 4.486090775988287e-06,
      "loss": 0.608,
      "step": 2650
    },
    {
      "epoch": 3.953147877013177,
      "grad_norm": 4.320408821105957,
      "learning_rate": 4.193265007320644e-06,
      "loss": 0.5978,
      "step": 2700
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.637628111273792,
      "eval_loss": 0.604773759841919,
      "eval_runtime": 9.1056,
      "eval_samples_per_second": 300.036,
      "eval_steps_per_second": 4.722,
      "step": 2732
    },
    {
      "epoch": 4.026354319180088,
      "grad_norm": 4.932640075683594,
      "learning_rate": 3.900439238653002e-06,
      "loss": 0.592,
      "step": 2750
    },
    {
      "epoch": 4.099560761346998,
      "grad_norm": 4.4933929443359375,
      "learning_rate": 3.607613469985359e-06,
      "loss": 0.6146,
      "step": 2800
    },
    {
      "epoch": 4.172767203513909,
      "grad_norm": 5.861272811889648,
      "learning_rate": 3.314787701317716e-06,
      "loss": 0.5881,
      "step": 2850
    },
    {
      "epoch": 4.24597364568082,
      "grad_norm": 2.9945833683013916,
      "learning_rate": 3.0219619326500732e-06,
      "loss": 0.5937,
      "step": 2900
    },
    {
      "epoch": 4.31918008784773,
      "grad_norm": 5.6117658615112305,
      "learning_rate": 2.7291361639824306e-06,
      "loss": 0.5811,
      "step": 2950
    },
    {
      "epoch": 4.392386530014641,
      "grad_norm": 3.8142032623291016,
      "learning_rate": 2.436310395314788e-06,
      "loss": 0.5561,
      "step": 3000
    },
    {
      "epoch": 4.465592972181552,
      "grad_norm": 3.9350202083587646,
      "learning_rate": 2.1434846266471453e-06,
      "loss": 0.581,
      "step": 3050
    },
    {
      "epoch": 4.538799414348462,
      "grad_norm": 5.021022796630859,
      "learning_rate": 1.8506588579795024e-06,
      "loss": 0.581,
      "step": 3100
    },
    {
      "epoch": 4.612005856515373,
      "grad_norm": 4.0799760818481445,
      "learning_rate": 1.5578330893118595e-06,
      "loss": 0.5673,
      "step": 3150
    },
    {
      "epoch": 4.685212298682284,
      "grad_norm": 3.7053580284118652,
      "learning_rate": 1.2650073206442169e-06,
      "loss": 0.5759,
      "step": 3200
    },
    {
      "epoch": 4.758418740849194,
      "grad_norm": 6.692492961883545,
      "learning_rate": 9.72181551976574e-07,
      "loss": 0.572,
      "step": 3250
    },
    {
      "epoch": 4.831625183016105,
      "grad_norm": 3.743006467819214,
      "learning_rate": 6.793557833089313e-07,
      "loss": 0.5728,
      "step": 3300
    },
    {
      "epoch": 4.904831625183016,
      "grad_norm": 4.34169340133667,
      "learning_rate": 3.865300146412885e-07,
      "loss": 0.5952,
      "step": 3350
    },
    {
      "epoch": 4.978038067349927,
      "grad_norm": 6.44805383682251,
      "learning_rate": 9.370424597364569e-08,
      "loss": 0.5987,
      "step": 3400
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6445827232796486,
      "eval_loss": 0.5975594520568848,
      "eval_runtime": 9.1017,
      "eval_samples_per_second": 300.165,
      "eval_steps_per_second": 4.724,
      "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": 3.33697966920192e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}