{
  "best_metric": 0.06277719885110855,
  "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": 2.219852924346924,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.0246,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.8659707307815552,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.9458,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 2.0018815994262695,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.8431,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.6520726680755615,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.698,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 4.319951057434082,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.6624,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.149073600769043,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.5853,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 1.812466025352478,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.5222,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.2116239070892334,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.3928,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.4122917652130127,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.3535,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.0293128490448,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.3675,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 1.9505114555358887,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.3324,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.59676194190979,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.2437,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 7.577399253845215,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2816,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 0.21850371360778809,
      "eval_runtime": 11.5027,
      "eval_samples_per_second": 11.562,
      "eval_steps_per_second": 1.478,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.6011801958084106,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.2002,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.0274529457092285,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.2245,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.394081950187683,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2019,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.5880785584449768,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.2201,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.6059957146644592,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2029,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 3.7114431858062744,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.1611,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.4368187189102173,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2473,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.7314859628677368,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.1804,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 4.0826029777526855,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 0.2188,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 1.8598849773406982,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.1661,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 4.044610977172852,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 0.2667,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.3763611912727356,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.1362,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.34921592473983765,
      "learning_rate": 1.2e-05,
      "loss": 0.1309,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.12996529042720795,
      "eval_runtime": 12.443,
      "eval_samples_per_second": 10.689,
      "eval_steps_per_second": 1.366,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.4331963360309601,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.0956,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.3378084897994995,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 0.1437,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 7.3877129554748535,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.13,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 8.507857322692871,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.1471,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 5.198700428009033,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.1737,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.33929577469825745,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.152,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 6.399758338928223,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.1785,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.2437870502471924,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.0759,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.3521318733692169,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.1332,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 8.943926811218262,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.1153,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.30642780661582947,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.1547,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.232720971107483,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.1254,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.29669004678726196,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1404,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.09639156609773636,
      "eval_runtime": 10.385,
      "eval_samples_per_second": 12.807,
      "eval_steps_per_second": 1.637,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.410691022872925,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.1347,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 9.097871780395508,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.1633,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 9.103521347045898,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.1377,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.23236368596553802,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.0739,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 17.802663803100586,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.1265,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 3.214564085006714,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.075,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 4.612748622894287,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.0965,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 8.368934631347656,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.0758,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.2200031727552414,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.0933,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.21209445595741272,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.0687,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.8884344100952148,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.1415,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 0.2086007297039032,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.136,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3007648289203644,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0866,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.06277719885110855,
      "eval_runtime": 10.39,
      "eval_samples_per_second": 12.801,
      "eval_steps_per_second": 1.636,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 2.7943220138549805,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.0956,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 1.370833396911621,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.1624,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 0.3506203293800354,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.15,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 1.352747917175293,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.0502,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 0.4083927869796753,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.1344,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 0.27641379833221436,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.0808,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 8.493270874023438,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.0654,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 3.093735933303833,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.1045,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 5.344343662261963,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.0935,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.2540881931781769,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.0771,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 0.22097405791282654,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.0624,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 0.22977662086486816,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.0751,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.59930819272995,
      "learning_rate": 0.0,
      "loss": 0.1156,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.08297071605920792,
      "eval_runtime": 11.1539,
      "eval_samples_per_second": 11.924,
      "eval_steps_per_second": 1.524,
      "step": 650
    },
    {
      "epoch": 5.0,
      "step": 650,
      "total_flos": 4.006371770595533e+17,
      "train_loss": 0.0,
      "train_runtime": 0.0789,
      "train_samples_per_second": 65533.623,
      "train_steps_per_second": 8239.237
    }
  ],
  "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
}