{
  "best_metric": 0.0633806362748146,
  "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": 2.1983468532562256,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.0245,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.845805048942566,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.9453,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 2.114424467086792,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.8403,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.6073758602142334,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.6945,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 4.094237804412842,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.6607,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.0681352615356445,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.5807,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 1.7710206508636475,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.5197,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.2429312467575073,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.3937,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.4406018257141113,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.3528,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.027031660079956,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.3667,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 1.9240713119506836,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.3334,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.638930559158325,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.2454,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 6.830716609954834,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2819,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 0.21523836255073547,
      "eval_runtime": 0.9557,
      "eval_samples_per_second": 139.162,
      "eval_steps_per_second": 17.788,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.518559455871582,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.1985,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.1490204334259033,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.2231,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.4814776182174683,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2045,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.5196593999862671,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.2222,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.6076135635375977,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2101,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 3.965792417526245,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.1627,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.4577326774597168,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2518,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.5628718137741089,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.1809,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 2.0682942867279053,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 0.2118,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 1.7764798402786255,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.1658,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 5.369350433349609,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 0.2622,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.4012261927127838,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.1406,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3495618999004364,
      "learning_rate": 1.2e-05,
      "loss": 0.1301,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.13008521497249603,
      "eval_runtime": 0.9119,
      "eval_samples_per_second": 145.855,
      "eval_steps_per_second": 18.643,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.45774486660957336,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.0893,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.33671364188194275,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 0.1445,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 7.082273960113525,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.1251,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 8.469789505004883,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.1379,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 5.470594882965088,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.1717,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.343715101480484,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.1519,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 5.930575370788574,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.1887,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.3529945611953735,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.074,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.34375670552253723,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.1337,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 10.965347290039062,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.1184,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.35521456599235535,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.1505,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.293394684791565,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.1321,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.2926313877105713,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.138,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.0965404212474823,
      "eval_runtime": 0.8504,
      "eval_samples_per_second": 156.398,
      "eval_steps_per_second": 19.991,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.190274477005005,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.1305,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 6.677399158477783,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.155,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 9.334908485412598,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.1401,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.23280958831310272,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.0723,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 3.986518621444702,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.1132,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 12.469901084899902,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.0852,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 4.926708698272705,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.0976,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 8.433639526367188,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.0786,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.2203906923532486,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.0983,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.21135270595550537,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.0777,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.9004859328269958,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.1418,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 0.21626874804496765,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.1231,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3029797673225403,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.087,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.0633806362748146,
      "eval_runtime": 0.8952,
      "eval_samples_per_second": 148.564,
      "eval_steps_per_second": 18.989,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 5.996098041534424,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.0983,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 1.5587939023971558,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.1635,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 0.3577156960964203,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.153,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 1.2371996641159058,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.0502,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 0.37243759632110596,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.1318,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 0.2695910632610321,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.0827,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 8.743815422058105,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.0731,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 3.647848606109619,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.0905,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 4.872845649719238,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.0836,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.2711491584777832,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.0837,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 0.21131394803524017,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.0661,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 0.23692815005779266,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.0774,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6001172661781311,
      "learning_rate": 0.0,
      "loss": 0.1113,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.07878894358873367,
      "eval_runtime": 0.8551,
      "eval_samples_per_second": 155.538,
      "eval_steps_per_second": 19.881,
      "step": 650
    },
    {
      "epoch": 5.0,
      "step": 650,
      "total_flos": 4.006371770595533e+17,
      "train_loss": 0.21883984464865464,
      "train_runtime": 73.5636,
      "train_samples_per_second": 70.279,
      "train_steps_per_second": 8.836
    }
  ],
  "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
}