{
  "best_metric": 0.06357112526893616,
  "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": 2.219912052154541,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.0246,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.8681813478469849,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.9458,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 2.1414010524749756,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.843,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.6058433055877686,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.6974,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 4.402042388916016,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.6629,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.1224277019500732,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.5859,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 1.8387095928192139,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.5239,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.2264832258224487,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.3948,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.4116926193237305,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.3542,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.2245359420776367,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.3714,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 1.8246150016784668,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.3337,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.579549789428711,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.245,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 7.8159403800964355,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2848,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 0.21651236712932587,
      "eval_runtime": 4.5968,
      "eval_samples_per_second": 28.933,
      "eval_steps_per_second": 3.698,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.6502667665481567,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.2002,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.110250473022461,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.2245,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.432618498802185,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2014,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.5069489479064941,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.2194,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.616087794303894,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2059,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 3.305881977081299,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.1563,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.44331827759742737,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2556,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.6302493214607239,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.1847,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 3.098297119140625,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 0.2169,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 1.7016505002975464,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.1744,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 2.809041976928711,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 0.2587,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.39431506395339966,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.1388,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.34844547510147095,
      "learning_rate": 1.2e-05,
      "loss": 0.1354,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.1263737827539444,
      "eval_runtime": 4.4788,
      "eval_samples_per_second": 29.696,
      "eval_steps_per_second": 3.796,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.46993738412857056,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.0904,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.33621352910995483,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 0.1381,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 6.853353023529053,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.1324,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 9.30165958404541,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.1401,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 5.741597652435303,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.1749,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.3132116198539734,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.1525,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 7.111733436584473,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.1783,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.1952487230300903,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.075,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.3361908197402954,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.1362,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 7.930729389190674,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.1209,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.30117857456207275,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.1499,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.120028018951416,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.1284,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.28898346424102783,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1425,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.09617571532726288,
      "eval_runtime": 4.5523,
      "eval_samples_per_second": 29.216,
      "eval_steps_per_second": 3.734,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.4895524978637695,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.1354,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 8.028548240661621,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.1611,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 8.18461799621582,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.1389,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.23116138577461243,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.0745,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 4.462396144866943,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.1249,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 11.275718688964844,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.0792,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 5.329637050628662,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.0952,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 8.307212829589844,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.0717,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.2321697175502777,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.0842,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.21286359429359436,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.0762,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.9805808067321777,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.1455,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 0.22639550268650055,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.1349,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3098979890346527,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0847,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.06357112526893616,
      "eval_runtime": 4.6155,
      "eval_samples_per_second": 28.816,
      "eval_steps_per_second": 3.683,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 5.721865653991699,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.0969,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 1.4805147647857666,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.1625,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 0.37912416458129883,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.1505,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 1.2134712934494019,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.0487,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 0.5184732675552368,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.133,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 0.2708996832370758,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.082,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 9.838260650634766,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.0671,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 3.5887091159820557,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.0989,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 5.537531852722168,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.09,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.2328355312347412,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.076,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 0.21624204516410828,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.0627,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 0.22994200885295868,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.0735,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6011123657226562,
      "learning_rate": 0.0,
      "loss": 0.11,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.08139312267303467,
      "eval_runtime": 4.6501,
      "eval_samples_per_second": 28.602,
      "eval_steps_per_second": 3.656,
      "step": 650
    },
    {
      "epoch": 5.0,
      "step": 650,
      "total_flos": 4.006371770595533e+17,
      "train_loss": 0.21934667711074535,
      "train_runtime": 562.2583,
      "train_samples_per_second": 9.195,
      "train_steps_per_second": 1.156
    }
  ],
  "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
}