{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 591,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.050761421319796954,
      "grad_norm": 2.884176452669992,
      "learning_rate": 5e-06,
      "loss": 0.7058,
      "step": 10
    },
    {
      "epoch": 0.10152284263959391,
      "grad_norm": 1.1031748057601416,
      "learning_rate": 5e-06,
      "loss": 0.6396,
      "step": 20
    },
    {
      "epoch": 0.15228426395939088,
      "grad_norm": 0.8763770464242356,
      "learning_rate": 5e-06,
      "loss": 0.6263,
      "step": 30
    },
    {
      "epoch": 0.20304568527918782,
      "grad_norm": 0.8637694735287882,
      "learning_rate": 5e-06,
      "loss": 0.6163,
      "step": 40
    },
    {
      "epoch": 0.25380710659898476,
      "grad_norm": 0.6443562464039193,
      "learning_rate": 5e-06,
      "loss": 0.5983,
      "step": 50
    },
    {
      "epoch": 0.30456852791878175,
      "grad_norm": 0.7181436435893084,
      "learning_rate": 5e-06,
      "loss": 0.5841,
      "step": 60
    },
    {
      "epoch": 0.3553299492385787,
      "grad_norm": 0.5736245789251513,
      "learning_rate": 5e-06,
      "loss": 0.5851,
      "step": 70
    },
    {
      "epoch": 0.40609137055837563,
      "grad_norm": 0.5370831215793392,
      "learning_rate": 5e-06,
      "loss": 0.5743,
      "step": 80
    },
    {
      "epoch": 0.45685279187817257,
      "grad_norm": 0.7868112154827747,
      "learning_rate": 5e-06,
      "loss": 0.5631,
      "step": 90
    },
    {
      "epoch": 0.5076142131979695,
      "grad_norm": 0.6384967672565564,
      "learning_rate": 5e-06,
      "loss": 0.5621,
      "step": 100
    },
    {
      "epoch": 0.5583756345177665,
      "grad_norm": 0.6105632833081082,
      "learning_rate": 5e-06,
      "loss": 0.5626,
      "step": 110
    },
    {
      "epoch": 0.6091370558375635,
      "grad_norm": 0.4980885433595067,
      "learning_rate": 5e-06,
      "loss": 0.5576,
      "step": 120
    },
    {
      "epoch": 0.6598984771573604,
      "grad_norm": 0.5567494710307067,
      "learning_rate": 5e-06,
      "loss": 0.5634,
      "step": 130
    },
    {
      "epoch": 0.7106598984771574,
      "grad_norm": 0.5747955173184647,
      "learning_rate": 5e-06,
      "loss": 0.5556,
      "step": 140
    },
    {
      "epoch": 0.7614213197969543,
      "grad_norm": 0.7422760458895346,
      "learning_rate": 5e-06,
      "loss": 0.5485,
      "step": 150
    },
    {
      "epoch": 0.8121827411167513,
      "grad_norm": 0.5519407925537647,
      "learning_rate": 5e-06,
      "loss": 0.5555,
      "step": 160
    },
    {
      "epoch": 0.8629441624365483,
      "grad_norm": 0.3938384364399605,
      "learning_rate": 5e-06,
      "loss": 0.5534,
      "step": 170
    },
    {
      "epoch": 0.9137055837563451,
      "grad_norm": 0.4274758585586587,
      "learning_rate": 5e-06,
      "loss": 0.5504,
      "step": 180
    },
    {
      "epoch": 0.9644670050761421,
      "grad_norm": 0.4623946226615239,
      "learning_rate": 5e-06,
      "loss": 0.5428,
      "step": 190
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.545667827129364,
      "eval_runtime": 69.9223,
      "eval_samples_per_second": 75.899,
      "eval_steps_per_second": 0.601,
      "step": 197
    },
    {
      "epoch": 1.015228426395939,
      "grad_norm": 0.4892463181375632,
      "learning_rate": 5e-06,
      "loss": 0.538,
      "step": 200
    },
    {
      "epoch": 1.0659898477157361,
      "grad_norm": 0.5258103265854674,
      "learning_rate": 5e-06,
      "loss": 0.5005,
      "step": 210
    },
    {
      "epoch": 1.116751269035533,
      "grad_norm": 0.43953346933881265,
      "learning_rate": 5e-06,
      "loss": 0.5116,
      "step": 220
    },
    {
      "epoch": 1.16751269035533,
      "grad_norm": 0.555350254923011,
      "learning_rate": 5e-06,
      "loss": 0.5039,
      "step": 230
    },
    {
      "epoch": 1.218274111675127,
      "grad_norm": 0.4240258545882722,
      "learning_rate": 5e-06,
      "loss": 0.4979,
      "step": 240
    },
    {
      "epoch": 1.2690355329949239,
      "grad_norm": 0.6055626365057643,
      "learning_rate": 5e-06,
      "loss": 0.4977,
      "step": 250
    },
    {
      "epoch": 1.3197969543147208,
      "grad_norm": 0.5183814019968731,
      "learning_rate": 5e-06,
      "loss": 0.5015,
      "step": 260
    },
    {
      "epoch": 1.3705583756345177,
      "grad_norm": 0.44198994392173363,
      "learning_rate": 5e-06,
      "loss": 0.5052,
      "step": 270
    },
    {
      "epoch": 1.4213197969543148,
      "grad_norm": 0.4597615345223239,
      "learning_rate": 5e-06,
      "loss": 0.4974,
      "step": 280
    },
    {
      "epoch": 1.4720812182741116,
      "grad_norm": 0.47036560762175594,
      "learning_rate": 5e-06,
      "loss": 0.5044,
      "step": 290
    },
    {
      "epoch": 1.5228426395939088,
      "grad_norm": 0.4356873877920564,
      "learning_rate": 5e-06,
      "loss": 0.5134,
      "step": 300
    },
    {
      "epoch": 1.5736040609137056,
      "grad_norm": 0.4547653150275499,
      "learning_rate": 5e-06,
      "loss": 0.5022,
      "step": 310
    },
    {
      "epoch": 1.6243654822335025,
      "grad_norm": 0.5243750306692014,
      "learning_rate": 5e-06,
      "loss": 0.5001,
      "step": 320
    },
    {
      "epoch": 1.6751269035532994,
      "grad_norm": 0.5072947429791255,
      "learning_rate": 5e-06,
      "loss": 0.504,
      "step": 330
    },
    {
      "epoch": 1.7258883248730963,
      "grad_norm": 0.4591167012346128,
      "learning_rate": 5e-06,
      "loss": 0.5008,
      "step": 340
    },
    {
      "epoch": 1.7766497461928934,
      "grad_norm": 0.559953933092174,
      "learning_rate": 5e-06,
      "loss": 0.5058,
      "step": 350
    },
    {
      "epoch": 1.8274111675126905,
      "grad_norm": 0.4615691289798925,
      "learning_rate": 5e-06,
      "loss": 0.5013,
      "step": 360
    },
    {
      "epoch": 1.8781725888324874,
      "grad_norm": 0.4673102096867509,
      "learning_rate": 5e-06,
      "loss": 0.505,
      "step": 370
    },
    {
      "epoch": 1.9289340101522843,
      "grad_norm": 0.41382969409293313,
      "learning_rate": 5e-06,
      "loss": 0.5029,
      "step": 380
    },
    {
      "epoch": 1.9796954314720812,
      "grad_norm": 0.4786556881881671,
      "learning_rate": 5e-06,
      "loss": 0.4991,
      "step": 390
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5367357730865479,
      "eval_runtime": 69.7466,
      "eval_samples_per_second": 76.09,
      "eval_steps_per_second": 0.602,
      "step": 394
    },
    {
      "epoch": 2.030456852791878,
      "grad_norm": 0.7058322019162009,
      "learning_rate": 5e-06,
      "loss": 0.4785,
      "step": 400
    },
    {
      "epoch": 2.081218274111675,
      "grad_norm": 0.45765334889032916,
      "learning_rate": 5e-06,
      "loss": 0.4563,
      "step": 410
    },
    {
      "epoch": 2.1319796954314723,
      "grad_norm": 0.4288566534513188,
      "learning_rate": 5e-06,
      "loss": 0.4564,
      "step": 420
    },
    {
      "epoch": 2.182741116751269,
      "grad_norm": 0.4543531173101596,
      "learning_rate": 5e-06,
      "loss": 0.4559,
      "step": 430
    },
    {
      "epoch": 2.233502538071066,
      "grad_norm": 0.5677467898296061,
      "learning_rate": 5e-06,
      "loss": 0.454,
      "step": 440
    },
    {
      "epoch": 2.284263959390863,
      "grad_norm": 0.5213901371480834,
      "learning_rate": 5e-06,
      "loss": 0.4584,
      "step": 450
    },
    {
      "epoch": 2.33502538071066,
      "grad_norm": 0.5106205098926287,
      "learning_rate": 5e-06,
      "loss": 0.4603,
      "step": 460
    },
    {
      "epoch": 2.3857868020304567,
      "grad_norm": 0.48058806334750254,
      "learning_rate": 5e-06,
      "loss": 0.4586,
      "step": 470
    },
    {
      "epoch": 2.436548223350254,
      "grad_norm": 0.4462504100857437,
      "learning_rate": 5e-06,
      "loss": 0.4547,
      "step": 480
    },
    {
      "epoch": 2.487309644670051,
      "grad_norm": 0.47373720391622604,
      "learning_rate": 5e-06,
      "loss": 0.4553,
      "step": 490
    },
    {
      "epoch": 2.5380710659898478,
      "grad_norm": 0.44268127361884035,
      "learning_rate": 5e-06,
      "loss": 0.4552,
      "step": 500
    },
    {
      "epoch": 2.5888324873096447,
      "grad_norm": 0.5168789659092264,
      "learning_rate": 5e-06,
      "loss": 0.4495,
      "step": 510
    },
    {
      "epoch": 2.6395939086294415,
      "grad_norm": 0.4759213683260235,
      "learning_rate": 5e-06,
      "loss": 0.4601,
      "step": 520
    },
    {
      "epoch": 2.6903553299492384,
      "grad_norm": 0.5405150525609788,
      "learning_rate": 5e-06,
      "loss": 0.4639,
      "step": 530
    },
    {
      "epoch": 2.7411167512690353,
      "grad_norm": 0.48554448131156325,
      "learning_rate": 5e-06,
      "loss": 0.4561,
      "step": 540
    },
    {
      "epoch": 2.7918781725888326,
      "grad_norm": 0.5715450026884347,
      "learning_rate": 5e-06,
      "loss": 0.458,
      "step": 550
    },
    {
      "epoch": 2.8426395939086295,
      "grad_norm": 0.45709891570844524,
      "learning_rate": 5e-06,
      "loss": 0.4582,
      "step": 560
    },
    {
      "epoch": 2.8934010152284264,
      "grad_norm": 0.4329230281361173,
      "learning_rate": 5e-06,
      "loss": 0.4535,
      "step": 570
    },
    {
      "epoch": 2.9441624365482233,
      "grad_norm": 0.4707895543051601,
      "learning_rate": 5e-06,
      "loss": 0.4609,
      "step": 580
    },
    {
      "epoch": 2.99492385786802,
      "grad_norm": 0.4717664390758382,
      "learning_rate": 5e-06,
      "loss": 0.4636,
      "step": 590
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5386124849319458,
      "eval_runtime": 67.8364,
      "eval_samples_per_second": 78.232,
      "eval_steps_per_second": 0.619,
      "step": 591
    },
    {
      "epoch": 3.0,
      "step": 591,
      "total_flos": 989528252743680.0,
      "train_loss": 0.5134775242224563,
      "train_runtime": 10543.3353,
      "train_samples_per_second": 28.688,
      "train_steps_per_second": 0.056
    }
  ],
  "logging_steps": 10,
  "max_steps": 591,
  "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": 989528252743680.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}