{
  "best_metric": 0.06271149218082428,
  "best_model_checkpoint": "./beans_outputs/checkpoint-520",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 650,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 2.1943726539611816,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.0249,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.8665084838867188,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.9463,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 2.0366761684417725,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.8401,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.6267013549804688,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.6974,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 4.034597396850586,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.664,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.0909483432769775,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.5928,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 1.782705307006836,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.5262,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.3501217365264893,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.4015,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.443540573120117,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.3581,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.095242738723755,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.3728,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 2.682217836380005,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.3364,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.5428781509399414,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.2471,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 6.824251651763916,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.281,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 0.2193511426448822,
      "eval_runtime": 2.0821,
      "eval_samples_per_second": 63.878,
      "eval_steps_per_second": 8.165,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.4056776762008667,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.2017,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 2.9976863861083984,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.225,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.4828065633773804,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2011,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.5160149335861206,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.216,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.606515645980835,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2028,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 4.552361011505127,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.1627,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.44607630372047424,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2465,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.686824381351471,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.1797,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 2.306919574737549,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 0.2161,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 2.050942897796631,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.1433,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 7.248354434967041,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 0.2762,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.40929391980171204,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.1286,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3473702669143677,
      "learning_rate": 1.2e-05,
      "loss": 0.1268,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.12887412309646606,
      "eval_runtime": 2.0415,
      "eval_samples_per_second": 65.15,
      "eval_steps_per_second": 8.327,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.6053388118743896,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.0969,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.3397085666656494,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 0.1396,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 7.664570331573486,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.1308,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 9.331936836242676,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.1427,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 6.082279682159424,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.1687,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.3458414375782013,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.1621,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 5.815878391265869,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.1863,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.0016790628433228,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.0756,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.35424068570137024,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.1243,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 13.017029762268066,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.1215,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.30303019285202026,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.1552,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.2382951974868774,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.1296,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.29275208711624146,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1385,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.09677492827177048,
      "eval_runtime": 2.1789,
      "eval_samples_per_second": 61.041,
      "eval_steps_per_second": 7.802,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.064225673675537,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.1348,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 6.044634819030762,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.1496,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 9.788597106933594,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.142,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.23322972655296326,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.0788,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 2.807680368423462,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.1074,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 4.750285625457764,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.0719,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 5.354732990264893,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.0966,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 8.170781135559082,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.0753,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.22035281360149384,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.0902,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.2130032181739807,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.077,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.9436342716217041,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.1247,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 0.2058902233839035,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.1309,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.30152463912963867,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0749,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.06271149218082428,
      "eval_runtime": 1.9746,
      "eval_samples_per_second": 67.355,
      "eval_steps_per_second": 8.609,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 2.3252828121185303,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.0948,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 2.0056347846984863,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.1599,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 0.3621722161769867,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.1402,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 1.570186734199524,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.0493,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 0.31239092350006104,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.1283,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 0.39084771275520325,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.0804,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 9.627459526062012,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.064,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 3.2321269512176514,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.0828,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 5.409714221954346,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.0899,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.2573850750923157,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.0794,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 0.21551428735256195,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.0616,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 0.23159781098365784,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.0776,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.7027397155761719,
      "learning_rate": 0.0,
      "loss": 0.1089,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.07966959476470947,
      "eval_runtime": 1.9647,
      "eval_samples_per_second": 67.695,
      "eval_steps_per_second": 8.653,
      "step": 650
    },
    {
      "epoch": 5.0,
      "step": 650,
      "total_flos": 4.006371770595533e+17,
      "train_loss": 0.2178187003502479,
      "train_runtime": 261.0621,
      "train_samples_per_second": 19.804,
      "train_steps_per_second": 2.49
    }
  ],
  "logging_steps": 10,
  "max_steps": 650,
  "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": 4.006371770595533e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}