{
  "best_metric": 0.06797418743371964,
  "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": 2.198302984237671,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.0245,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.845097541809082,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.9453,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 2.150439739227295,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.8406,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.6366426944732666,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.6947,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 4.09306526184082,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.661,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.1485249996185303,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.58,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 2.1387364864349365,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.5187,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.241268277168274,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.3923,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.5380561351776123,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.3482,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.8183135986328125,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.3693,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 2.0173871517181396,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.3363,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.347116231918335,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.2488,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 7.064352035522461,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2869,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 0.21880216896533966,
      "eval_runtime": 10.9454,
      "eval_samples_per_second": 12.151,
      "eval_steps_per_second": 1.553,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.7081693410873413,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.2015,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.1946160793304443,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.2237,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.4541012048721313,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2086,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.5262917280197144,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.2227,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.5964558124542236,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2137,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 3.347987413406372,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.1612,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.45679983496665955,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2579,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.6202031970024109,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.182,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 2.1473233699798584,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 0.2099,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 1.916715145111084,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.1568,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 4.764385223388672,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 0.2773,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.4081193804740906,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.1441,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3510228395462036,
      "learning_rate": 1.2e-05,
      "loss": 0.131,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.13096114993095398,
      "eval_runtime": 10.0017,
      "eval_samples_per_second": 13.298,
      "eval_steps_per_second": 1.7,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.4616471230983734,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.0926,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.338736355304718,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 0.1428,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 6.58615255355835,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.1229,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 6.827399730682373,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.135,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 4.440911293029785,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.1728,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.308022141456604,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.1501,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 7.213817596435547,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.1811,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.0687382221221924,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.0751,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.3557477593421936,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.1345,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 13.175363540649414,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.1435,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.29770728945732117,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.145,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.1821846961975098,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.1319,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.29596850275993347,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1467,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.09739137440919876,
      "eval_runtime": 9.9904,
      "eval_samples_per_second": 13.313,
      "eval_steps_per_second": 1.702,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.110072374343872,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.138,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 5.901464462280273,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.1542,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 8.865696907043457,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.1449,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.23266926407814026,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.0757,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 6.493763446807861,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.116,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 10.815217018127441,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.0784,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 4.633830547332764,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.0971,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 7.9144182205200195,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.0842,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.22700750827789307,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.0956,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.2125973105430603,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.0806,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.8930318355560303,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.144,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 0.20944423973560333,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.1241,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3025684654712677,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0797,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.06797418743371964,
      "eval_runtime": 10.7931,
      "eval_samples_per_second": 12.323,
      "eval_steps_per_second": 1.575,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 10.744412422180176,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.1032,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 1.7659765481948853,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.1622,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 0.34627851843833923,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.1526,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 1.1866897344589233,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.0505,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 0.27596765756607056,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.1352,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 0.27586278319358826,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.0826,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 9.091331481933594,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.0725,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 4.240416526794434,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.0941,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 5.184774398803711,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.0898,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.2620495557785034,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.0804,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 0.2159629911184311,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.067,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 0.23236247897148132,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.0754,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6177453398704529,
      "learning_rate": 0.0,
      "loss": 0.1236,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.08286623656749725,
      "eval_runtime": 10.4228,
      "eval_samples_per_second": 12.76,
      "eval_steps_per_second": 1.631,
      "step": 650
    },
    {
      "epoch": 5.0,
      "step": 650,
      "total_flos": 4.006371770595533e+17,
      "train_loss": 0.22018987765678993,
      "train_runtime": 1372.1584,
      "train_samples_per_second": 3.768,
      "train_steps_per_second": 0.474
    }
  ],
  "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
}