{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 825,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03636363636363636,
      "grad_norm": 3.2293971017711174,
      "learning_rate": 5e-06,
      "loss": 1.0337,
      "step": 10
    },
    {
      "epoch": 0.07272727272727272,
      "grad_norm": 1.2461654883314972,
      "learning_rate": 5e-06,
      "loss": 0.9092,
      "step": 20
    },
    {
      "epoch": 0.10909090909090909,
      "grad_norm": 1.0937923635217501,
      "learning_rate": 5e-06,
      "loss": 0.8658,
      "step": 30
    },
    {
      "epoch": 0.14545454545454545,
      "grad_norm": 1.3350225945199414,
      "learning_rate": 5e-06,
      "loss": 0.844,
      "step": 40
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 1.029425810987488,
      "learning_rate": 5e-06,
      "loss": 0.8249,
      "step": 50
    },
    {
      "epoch": 0.21818181818181817,
      "grad_norm": 1.5219290967515304,
      "learning_rate": 5e-06,
      "loss": 0.8068,
      "step": 60
    },
    {
      "epoch": 0.2545454545454545,
      "grad_norm": 1.6740495880819521,
      "learning_rate": 5e-06,
      "loss": 0.7989,
      "step": 70
    },
    {
      "epoch": 0.2909090909090909,
      "grad_norm": 1.2973735477904815,
      "learning_rate": 5e-06,
      "loss": 0.7921,
      "step": 80
    },
    {
      "epoch": 0.32727272727272727,
      "grad_norm": 0.8566363002967183,
      "learning_rate": 5e-06,
      "loss": 0.781,
      "step": 90
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.9961672641644985,
      "learning_rate": 5e-06,
      "loss": 0.7745,
      "step": 100
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.8927257959121373,
      "learning_rate": 5e-06,
      "loss": 0.7754,
      "step": 110
    },
    {
      "epoch": 0.43636363636363634,
      "grad_norm": 0.6135178704985191,
      "learning_rate": 5e-06,
      "loss": 0.772,
      "step": 120
    },
    {
      "epoch": 0.4727272727272727,
      "grad_norm": 0.7431505188106242,
      "learning_rate": 5e-06,
      "loss": 0.7686,
      "step": 130
    },
    {
      "epoch": 0.509090909090909,
      "grad_norm": 0.7150787812569424,
      "learning_rate": 5e-06,
      "loss": 0.7618,
      "step": 140
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 0.6352342662453642,
      "learning_rate": 5e-06,
      "loss": 0.7611,
      "step": 150
    },
    {
      "epoch": 0.5818181818181818,
      "grad_norm": 0.6257901300873526,
      "learning_rate": 5e-06,
      "loss": 0.7569,
      "step": 160
    },
    {
      "epoch": 0.6181818181818182,
      "grad_norm": 0.6387102446786417,
      "learning_rate": 5e-06,
      "loss": 0.7611,
      "step": 170
    },
    {
      "epoch": 0.6545454545454545,
      "grad_norm": 0.5983754152683597,
      "learning_rate": 5e-06,
      "loss": 0.7546,
      "step": 180
    },
    {
      "epoch": 0.6909090909090909,
      "grad_norm": 0.7480127979666656,
      "learning_rate": 5e-06,
      "loss": 0.7566,
      "step": 190
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.5804396007389026,
      "learning_rate": 5e-06,
      "loss": 0.75,
      "step": 200
    },
    {
      "epoch": 0.7636363636363637,
      "grad_norm": 0.682148918886327,
      "learning_rate": 5e-06,
      "loss": 0.7476,
      "step": 210
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.8039336411015884,
      "learning_rate": 5e-06,
      "loss": 0.7462,
      "step": 220
    },
    {
      "epoch": 0.8363636363636363,
      "grad_norm": 0.6876607052536684,
      "learning_rate": 5e-06,
      "loss": 0.7411,
      "step": 230
    },
    {
      "epoch": 0.8727272727272727,
      "grad_norm": 0.6588151842699974,
      "learning_rate": 5e-06,
      "loss": 0.7469,
      "step": 240
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.6715213794720472,
      "learning_rate": 5e-06,
      "loss": 0.7378,
      "step": 250
    },
    {
      "epoch": 0.9454545454545454,
      "grad_norm": 0.5870957383826958,
      "learning_rate": 5e-06,
      "loss": 0.7457,
      "step": 260
    },
    {
      "epoch": 0.9818181818181818,
      "grad_norm": 0.6643986810785624,
      "learning_rate": 5e-06,
      "loss": 0.7466,
      "step": 270
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7417545914649963,
      "eval_runtime": 26.6218,
      "eval_samples_per_second": 278.268,
      "eval_steps_per_second": 1.089,
      "step": 275
    },
    {
      "epoch": 1.018181818181818,
      "grad_norm": 0.9781239153342394,
      "learning_rate": 5e-06,
      "loss": 0.7136,
      "step": 280
    },
    {
      "epoch": 1.0545454545454545,
      "grad_norm": 0.7152925984087143,
      "learning_rate": 5e-06,
      "loss": 0.6871,
      "step": 290
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.6929492576277494,
      "learning_rate": 5e-06,
      "loss": 0.6894,
      "step": 300
    },
    {
      "epoch": 1.1272727272727272,
      "grad_norm": 0.728764264622129,
      "learning_rate": 5e-06,
      "loss": 0.6935,
      "step": 310
    },
    {
      "epoch": 1.1636363636363636,
      "grad_norm": 0.7252517543389313,
      "learning_rate": 5e-06,
      "loss": 0.6945,
      "step": 320
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.6665160391388197,
      "learning_rate": 5e-06,
      "loss": 0.6886,
      "step": 330
    },
    {
      "epoch": 1.2363636363636363,
      "grad_norm": 0.7161659905517039,
      "learning_rate": 5e-06,
      "loss": 0.6898,
      "step": 340
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 0.5719039452566653,
      "learning_rate": 5e-06,
      "loss": 0.6934,
      "step": 350
    },
    {
      "epoch": 1.309090909090909,
      "grad_norm": 0.6060853746189843,
      "learning_rate": 5e-06,
      "loss": 0.6922,
      "step": 360
    },
    {
      "epoch": 1.3454545454545455,
      "grad_norm": 0.6563719933283224,
      "learning_rate": 5e-06,
      "loss": 0.6912,
      "step": 370
    },
    {
      "epoch": 1.3818181818181818,
      "grad_norm": 0.5958006047997326,
      "learning_rate": 5e-06,
      "loss": 0.6904,
      "step": 380
    },
    {
      "epoch": 1.4181818181818182,
      "grad_norm": 0.7430218105320606,
      "learning_rate": 5e-06,
      "loss": 0.688,
      "step": 390
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.6322073230662588,
      "learning_rate": 5e-06,
      "loss": 0.6883,
      "step": 400
    },
    {
      "epoch": 1.490909090909091,
      "grad_norm": 0.7151221978666452,
      "learning_rate": 5e-06,
      "loss": 0.6934,
      "step": 410
    },
    {
      "epoch": 1.5272727272727273,
      "grad_norm": 0.6184168187218901,
      "learning_rate": 5e-06,
      "loss": 0.6916,
      "step": 420
    },
    {
      "epoch": 1.5636363636363635,
      "grad_norm": 0.6280848540221795,
      "learning_rate": 5e-06,
      "loss": 0.6916,
      "step": 430
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.6568705155050817,
      "learning_rate": 5e-06,
      "loss": 0.6856,
      "step": 440
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 0.6359258851827682,
      "learning_rate": 5e-06,
      "loss": 0.6851,
      "step": 450
    },
    {
      "epoch": 1.6727272727272728,
      "grad_norm": 0.710888538426671,
      "learning_rate": 5e-06,
      "loss": 0.6872,
      "step": 460
    },
    {
      "epoch": 1.709090909090909,
      "grad_norm": 0.7584066029266229,
      "learning_rate": 5e-06,
      "loss": 0.6849,
      "step": 470
    },
    {
      "epoch": 1.7454545454545456,
      "grad_norm": 0.5960492892442344,
      "learning_rate": 5e-06,
      "loss": 0.6891,
      "step": 480
    },
    {
      "epoch": 1.7818181818181817,
      "grad_norm": 0.5629377755020811,
      "learning_rate": 5e-06,
      "loss": 0.6847,
      "step": 490
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.589716689792314,
      "learning_rate": 5e-06,
      "loss": 0.6871,
      "step": 500
    },
    {
      "epoch": 1.8545454545454545,
      "grad_norm": 0.5740509121739076,
      "learning_rate": 5e-06,
      "loss": 0.6888,
      "step": 510
    },
    {
      "epoch": 1.8909090909090909,
      "grad_norm": 0.5891046247600111,
      "learning_rate": 5e-06,
      "loss": 0.6884,
      "step": 520
    },
    {
      "epoch": 1.9272727272727272,
      "grad_norm": 0.6447276827053491,
      "learning_rate": 5e-06,
      "loss": 0.6893,
      "step": 530
    },
    {
      "epoch": 1.9636363636363636,
      "grad_norm": 0.6935516132206995,
      "learning_rate": 5e-06,
      "loss": 0.6868,
      "step": 540
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5781509823001448,
      "learning_rate": 5e-06,
      "loss": 0.6841,
      "step": 550
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7281343340873718,
      "eval_runtime": 26.4698,
      "eval_samples_per_second": 279.867,
      "eval_steps_per_second": 1.096,
      "step": 550
    },
    {
      "epoch": 2.036363636363636,
      "grad_norm": 0.7551729949207574,
      "learning_rate": 5e-06,
      "loss": 0.6351,
      "step": 560
    },
    {
      "epoch": 2.0727272727272728,
      "grad_norm": 0.6070448901420726,
      "learning_rate": 5e-06,
      "loss": 0.6307,
      "step": 570
    },
    {
      "epoch": 2.109090909090909,
      "grad_norm": 0.7225948313371118,
      "learning_rate": 5e-06,
      "loss": 0.6357,
      "step": 580
    },
    {
      "epoch": 2.1454545454545455,
      "grad_norm": 1.3944109200671733,
      "learning_rate": 5e-06,
      "loss": 0.6375,
      "step": 590
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 1.1390572133302885,
      "learning_rate": 5e-06,
      "loss": 0.635,
      "step": 600
    },
    {
      "epoch": 2.2181818181818183,
      "grad_norm": 0.7900509422330505,
      "learning_rate": 5e-06,
      "loss": 0.6383,
      "step": 610
    },
    {
      "epoch": 2.2545454545454544,
      "grad_norm": 0.594871030626621,
      "learning_rate": 5e-06,
      "loss": 0.6321,
      "step": 620
    },
    {
      "epoch": 2.290909090909091,
      "grad_norm": 0.665898906007086,
      "learning_rate": 5e-06,
      "loss": 0.6341,
      "step": 630
    },
    {
      "epoch": 2.327272727272727,
      "grad_norm": 0.6509722897169726,
      "learning_rate": 5e-06,
      "loss": 0.6326,
      "step": 640
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.6231670817005929,
      "learning_rate": 5e-06,
      "loss": 0.6385,
      "step": 650
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.6425410588561774,
      "learning_rate": 5e-06,
      "loss": 0.6373,
      "step": 660
    },
    {
      "epoch": 2.4363636363636365,
      "grad_norm": 0.621241338432262,
      "learning_rate": 5e-06,
      "loss": 0.6399,
      "step": 670
    },
    {
      "epoch": 2.4727272727272727,
      "grad_norm": 0.6924233110335524,
      "learning_rate": 5e-06,
      "loss": 0.6393,
      "step": 680
    },
    {
      "epoch": 2.509090909090909,
      "grad_norm": 0.6419114963815122,
      "learning_rate": 5e-06,
      "loss": 0.6405,
      "step": 690
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.7336852368102121,
      "learning_rate": 5e-06,
      "loss": 0.6385,
      "step": 700
    },
    {
      "epoch": 2.581818181818182,
      "grad_norm": 0.7922288944252411,
      "learning_rate": 5e-06,
      "loss": 0.6377,
      "step": 710
    },
    {
      "epoch": 2.618181818181818,
      "grad_norm": 0.6500377491351792,
      "learning_rate": 5e-06,
      "loss": 0.6427,
      "step": 720
    },
    {
      "epoch": 2.6545454545454543,
      "grad_norm": 0.6853834065254241,
      "learning_rate": 5e-06,
      "loss": 0.6346,
      "step": 730
    },
    {
      "epoch": 2.690909090909091,
      "grad_norm": 0.8156333668312422,
      "learning_rate": 5e-06,
      "loss": 0.6414,
      "step": 740
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.6294215183471213,
      "learning_rate": 5e-06,
      "loss": 0.6363,
      "step": 750
    },
    {
      "epoch": 2.7636363636363637,
      "grad_norm": 0.8237171162592375,
      "learning_rate": 5e-06,
      "loss": 0.6421,
      "step": 760
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.6772752476166749,
      "learning_rate": 5e-06,
      "loss": 0.6356,
      "step": 770
    },
    {
      "epoch": 2.8363636363636364,
      "grad_norm": 0.7780500988065099,
      "learning_rate": 5e-06,
      "loss": 0.6425,
      "step": 780
    },
    {
      "epoch": 2.8727272727272726,
      "grad_norm": 0.6862874007983163,
      "learning_rate": 5e-06,
      "loss": 0.6368,
      "step": 790
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.5748210856771035,
      "learning_rate": 5e-06,
      "loss": 0.6405,
      "step": 800
    },
    {
      "epoch": 2.9454545454545453,
      "grad_norm": 0.6351457621560951,
      "learning_rate": 5e-06,
      "loss": 0.6357,
      "step": 810
    },
    {
      "epoch": 2.981818181818182,
      "grad_norm": 0.586627253325874,
      "learning_rate": 5e-06,
      "loss": 0.6412,
      "step": 820
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7299705147743225,
      "eval_runtime": 25.9228,
      "eval_samples_per_second": 285.772,
      "eval_steps_per_second": 1.119,
      "step": 825
    },
    {
      "epoch": 3.0,
      "step": 825,
      "total_flos": 1381905727488000.0,
      "train_loss": 0.7044297796307188,
      "train_runtime": 5353.5806,
      "train_samples_per_second": 78.866,
      "train_steps_per_second": 0.154
    }
  ],
  "logging_steps": 10,
  "max_steps": 825,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1381905727488000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}