{
  "best_metric": 0.06631391495466232,
  "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": 2.1983516216278076,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.0245,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.845354437828064,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.9453,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 2.025181293487549,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.8403,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.6090736389160156,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.695,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 4.343094825744629,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.6612,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.0946686267852783,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.5814,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 1.7924715280532837,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.5193,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.244881510734558,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.3922,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.5017757415771484,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.3546,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.0490570068359375,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.372,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 1.9284653663635254,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.3317,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.431607246398926,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.2461,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 7.428433895111084,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2854,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 0.2194061428308487,
      "eval_runtime": 12.4335,
      "eval_samples_per_second": 10.697,
      "eval_steps_per_second": 1.367,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.7276313304901123,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.2007,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.1764917373657227,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.225,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.43402099609375,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2057,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.519515335559845,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.2213,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.6076896786689758,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2112,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 4.2212629318237305,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.1627,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.459544837474823,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2557,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.6737858653068542,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.1805,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 2.4183340072631836,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 0.2106,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 1.8529425859451294,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.1542,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 6.124072074890137,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 0.265,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.39290884137153625,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.1373,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.34861478209495544,
      "learning_rate": 1.2e-05,
      "loss": 0.1302,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.1308879852294922,
      "eval_runtime": 17.5451,
      "eval_samples_per_second": 7.58,
      "eval_steps_per_second": 0.969,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.45311787724494934,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.0912,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.3352016806602478,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 0.1463,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 7.322403907775879,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.1304,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 8.625993728637695,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.1438,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 4.359902381896973,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.174,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.3188760280609131,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.1547,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 6.000261306762695,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.1902,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.7377233505249023,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.0731,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.35091903805732727,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.1295,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 9.31395149230957,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.1161,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.31594040989875793,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.1503,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.2401353120803833,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.1275,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.29739508032798767,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.144,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.09733586758375168,
      "eval_runtime": 22.1644,
      "eval_samples_per_second": 6.001,
      "eval_steps_per_second": 0.767,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.3267691135406494,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.1336,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 6.462228775024414,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.1527,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 9.17526912689209,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.1366,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.23261962831020355,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.0757,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 11.900029182434082,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.1201,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 11.986618041992188,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.08,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 5.186712741851807,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.0967,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 8.64492130279541,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.0747,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.221771702170372,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.102,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.21222464740276337,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.0747,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.8941594958305359,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.1343,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 0.21074175834655762,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.1305,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3024842143058777,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0808,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.06631391495466232,
      "eval_runtime": 13.287,
      "eval_samples_per_second": 10.01,
      "eval_steps_per_second": 1.279,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 8.620423316955566,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.0999,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 1.5799357891082764,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.1632,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 0.3434726595878601,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.1532,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 1.2967414855957031,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.0504,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 0.31858542561531067,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.1351,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 0.2646137475967407,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.0842,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 9.598633766174316,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.0728,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 4.918817043304443,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.0944,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 5.175420761108398,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.0857,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.2623072862625122,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.0828,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 0.21628715097904205,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.0681,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 0.2312709391117096,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.0772,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5570812225341797,
      "learning_rate": 0.0,
      "loss": 0.1109,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.08231782168149948,
      "eval_runtime": 13.812,
      "eval_samples_per_second": 9.629,
      "eval_steps_per_second": 1.231,
      "step": 650
    },
    {
      "epoch": 5.0,
      "step": 650,
      "total_flos": 4.006371770595533e+17,
      "train_loss": 0.2192414202139928,
      "train_runtime": 873.7716,
      "train_samples_per_second": 5.917,
      "train_steps_per_second": 0.744
    }
  ],
  "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
}