{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.15630035449565,
  "eval_steps": 100,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12890750886239125,
      "grad_norm": 0.9527155866900715,
      "learning_rate": 0.001,
      "loss": 4.2951,
      "step": 50
    },
    {
      "epoch": 0.2578150177247825,
      "grad_norm": 0.8613266657338488,
      "learning_rate": 0.0009995777619868967,
      "loss": 4.1174,
      "step": 100
    },
    {
      "epoch": 0.2578150177247825,
      "eval_loss": 4.061758041381836,
      "eval_runtime": 13.6791,
      "eval_samples_per_second": 73.104,
      "eval_steps_per_second": 2.339,
      "step": 100
    },
    {
      "epoch": 0.3867225265871737,
      "grad_norm": 0.588963638591784,
      "learning_rate": 0.000998311761801199,
      "loss": 4.035,
      "step": 150
    },
    {
      "epoch": 0.515630035449565,
      "grad_norm": 0.3000507476828121,
      "learning_rate": 0.000996204139796873,
      "loss": 3.9794,
      "step": 200
    },
    {
      "epoch": 0.515630035449565,
      "eval_loss": 3.951843500137329,
      "eval_runtime": 13.2341,
      "eval_samples_per_second": 75.563,
      "eval_steps_per_second": 2.418,
      "step": 200
    },
    {
      "epoch": 0.6445375443119562,
      "grad_norm": 1.0882364902367292,
      "learning_rate": 0.0009932584592096643,
      "loss": 3.9453,
      "step": 250
    },
    {
      "epoch": 0.7734450531743474,
      "grad_norm": 0.6363211565344357,
      "learning_rate": 0.0009894797001329398,
      "loss": 3.917,
      "step": 300
    },
    {
      "epoch": 0.7734450531743474,
      "eval_loss": 3.8788869380950928,
      "eval_runtime": 13.3005,
      "eval_samples_per_second": 75.185,
      "eval_steps_per_second": 2.406,
      "step": 300
    },
    {
      "epoch": 0.9023525620367386,
      "grad_norm": 0.9910006063725697,
      "learning_rate": 0.0009848742510981292,
      "loss": 3.805,
      "step": 350
    },
    {
      "epoch": 1.03126007089913,
      "grad_norm": 5.070998393490313,
      "learning_rate": 0.0009794498982740008,
      "loss": 3.7973,
      "step": 400
    },
    {
      "epoch": 1.03126007089913,
      "eval_loss": 3.929041862487793,
      "eval_runtime": 13.2112,
      "eval_samples_per_second": 75.693,
      "eval_steps_per_second": 2.422,
      "step": 400
    },
    {
      "epoch": 1.1601675797615212,
      "grad_norm": 1.5111119419703132,
      "learning_rate": 0.0009732158123030324,
      "loss": 3.98,
      "step": 450
    },
    {
      "epoch": 1.2890750886239124,
      "grad_norm": 2.9266272457246685,
      "learning_rate": 0.0009661825327971316,
      "loss": 3.9676,
      "step": 500
    },
    {
      "epoch": 1.2890750886239124,
      "eval_loss": 3.992887258529663,
      "eval_runtime": 13.2769,
      "eval_samples_per_second": 75.319,
      "eval_steps_per_second": 2.41,
      "step": 500
    },
    {
      "epoch": 1.4179825974863036,
      "grad_norm": 1.9372695993602536,
      "learning_rate": 0.0009583619505189177,
      "loss": 4.0102,
      "step": 550
    },
    {
      "epoch": 1.5468901063486948,
      "grad_norm": 1.1325425311995052,
      "learning_rate": 0.0009497672872786908,
      "loss": 4.0347,
      "step": 600
    },
    {
      "epoch": 1.5468901063486948,
      "eval_loss": 4.0086283683776855,
      "eval_runtime": 13.2747,
      "eval_samples_per_second": 75.331,
      "eval_steps_per_second": 2.411,
      "step": 600
    },
    {
      "epoch": 1.675797615211086,
      "grad_norm": 1.0653229015184138,
      "learning_rate": 0.0009404130735810749,
      "loss": 4.015,
      "step": 650
    },
    {
      "epoch": 1.8047051240734773,
      "grad_norm": 0.9867803701996282,
      "learning_rate": 0.0009303151240591263,
      "loss": 4.0545,
      "step": 700
    },
    {
      "epoch": 1.8047051240734773,
      "eval_loss": 4.0230817794799805,
      "eval_runtime": 13.2485,
      "eval_samples_per_second": 75.48,
      "eval_steps_per_second": 2.415,
      "step": 700
    },
    {
      "epoch": 1.9336126329358685,
      "grad_norm": 1.3505264820000504,
      "learning_rate": 0.0009194905107374401,
      "loss": 4.0391,
      "step": 750
    },
    {
      "epoch": 2.06252014179826,
      "grad_norm": 1.4563064030505048,
      "learning_rate": 0.0009079575341694557,
      "loss": 4.0552,
      "step": 800
    },
    {
      "epoch": 2.06252014179826,
      "eval_loss": 4.040332317352295,
      "eval_runtime": 13.2387,
      "eval_samples_per_second": 75.536,
      "eval_steps_per_second": 2.417,
      "step": 800
    },
    {
      "epoch": 2.191427650660651,
      "grad_norm": 1.2688404260592696,
      "learning_rate": 0.0008957356924977609,
      "loss": 4.0285,
      "step": 850
    },
    {
      "epoch": 2.3203351595230424,
      "grad_norm": 1.922527341942018,
      "learning_rate": 0.0008828456484896984,
      "loss": 4.0291,
      "step": 900
    },
    {
      "epoch": 2.3203351595230424,
      "eval_loss": 4.0184197425842285,
      "eval_runtime": 13.2837,
      "eval_samples_per_second": 75.28,
      "eval_steps_per_second": 2.409,
      "step": 900
    },
    {
      "epoch": 2.4492426683854336,
      "grad_norm": 1.1751317607621916,
      "learning_rate": 0.0008693091946040104,
      "loss": 4.0198,
      "step": 950
    },
    {
      "epoch": 2.578150177247825,
      "grad_norm": 1.4781189301871374,
      "learning_rate": 0.000855149216147576,
      "loss": 4.0389,
      "step": 1000
    },
    {
      "epoch": 2.578150177247825,
      "eval_loss": 4.022523403167725,
      "eval_runtime": 13.2647,
      "eval_samples_per_second": 75.388,
      "eval_steps_per_second": 2.412,
      "step": 1000
    },
    {
      "epoch": 2.707057686110216,
      "grad_norm": 2.8481102078113363,
      "learning_rate": 0.000840389652584536,
      "loss": 4.0337,
      "step": 1050
    },
    {
      "epoch": 2.8359651949726072,
      "grad_norm": 1.2842129496698058,
      "learning_rate": 0.0008250554570632107,
      "loss": 4.0293,
      "step": 1100
    },
    {
      "epoch": 2.8359651949726072,
      "eval_loss": 4.018503189086914,
      "eval_runtime": 13.2539,
      "eval_samples_per_second": 75.45,
      "eval_steps_per_second": 2.414,
      "step": 1100
    },
    {
      "epoch": 2.9648727038349985,
      "grad_norm": 3.7613179214902197,
      "learning_rate": 0.0008091725542292438,
      "loss": 4.0283,
      "step": 1150
    },
    {
      "epoch": 3.0937802126973897,
      "grad_norm": 2.6521101503823346,
      "learning_rate": 0.000792767796396289,
      "loss": 4.0347,
      "step": 1200
    },
    {
      "epoch": 3.0937802126973897,
      "eval_loss": 4.021185398101807,
      "eval_runtime": 13.2939,
      "eval_samples_per_second": 75.222,
      "eval_steps_per_second": 2.407,
      "step": 1200
    },
    {
      "epoch": 3.222687721559781,
      "grad_norm": 1.7003701014793156,
      "learning_rate": 0.0007758689181483412,
      "loss": 4.0275,
      "step": 1250
    },
    {
      "epoch": 3.351595230422172,
      "grad_norm": 1.3295380943734334,
      "learning_rate": 0.000758504489450466,
      "loss": 4.0208,
      "step": 1300
    },
    {
      "epoch": 3.351595230422172,
      "eval_loss": 4.011188507080078,
      "eval_runtime": 13.2452,
      "eval_samples_per_second": 75.499,
      "eval_steps_per_second": 2.416,
      "step": 1300
    },
    {
      "epoch": 3.4805027392845633,
      "grad_norm": 1.3487659259714693,
      "learning_rate": 0.0007407038673471959,
      "loss": 4.0191,
      "step": 1350
    },
    {
      "epoch": 3.6094102481469545,
      "grad_norm": 0.9861657777387535,
      "learning_rate": 0.0007224971463302587,
      "loss": 4.0204,
      "step": 1400
    },
    {
      "epoch": 3.6094102481469545,
      "eval_loss": 4.0111846923828125,
      "eval_runtime": 13.2369,
      "eval_samples_per_second": 75.546,
      "eval_steps_per_second": 2.417,
      "step": 1400
    },
    {
      "epoch": 3.7383177570093458,
      "grad_norm": 1.0800491613545316,
      "learning_rate": 0.0007039151074595432,
      "loss": 4.024,
      "step": 1450
    },
    {
      "epoch": 3.867225265871737,
      "grad_norm": 0.9644339387320809,
      "learning_rate": 0.0006849891663233264,
      "loss": 4.023,
      "step": 1500
    },
    {
      "epoch": 3.867225265871737,
      "eval_loss": 4.018366813659668,
      "eval_runtime": 13.2404,
      "eval_samples_per_second": 75.526,
      "eval_steps_per_second": 2.417,
      "step": 1500
    },
    {
      "epoch": 3.996132774734128,
      "grad_norm": 0.8344627432775175,
      "learning_rate": 0.0006657513199257385,
      "loss": 4.0233,
      "step": 1550
    },
    {
      "epoch": 4.12504028359652,
      "grad_norm": 1.4143353248285266,
      "learning_rate": 0.0006462340925912611,
      "loss": 4.0253,
      "step": 1600
    },
    {
      "epoch": 4.12504028359652,
      "eval_loss": 4.0222063064575195,
      "eval_runtime": 13.2332,
      "eval_samples_per_second": 75.567,
      "eval_steps_per_second": 2.418,
      "step": 1600
    },
    {
      "epoch": 4.253947792458911,
      "grad_norm": 0.970349395404879,
      "learning_rate": 0.0006264704809777159,
      "loss": 4.02,
      "step": 1650
    },
    {
      "epoch": 4.382855301321302,
      "grad_norm": 0.7142593470613441,
      "learning_rate": 0.0006064938982907064,
      "loss": 4.0186,
      "step": 1700
    },
    {
      "epoch": 4.382855301321302,
      "eval_loss": 4.019730567932129,
      "eval_runtime": 13.2783,
      "eval_samples_per_second": 75.311,
      "eval_steps_per_second": 2.41,
      "step": 1700
    },
    {
      "epoch": 4.5117628101836935,
      "grad_norm": 1.088690826939789,
      "learning_rate": 0.0005863381177938257,
      "loss": 4.0203,
      "step": 1750
    },
    {
      "epoch": 4.640670319046085,
      "grad_norm": 0.81043587990677,
      "learning_rate": 0.0005660372157101351,
      "loss": 4.0124,
      "step": 1800
    },
    {
      "epoch": 4.640670319046085,
      "eval_loss": 4.00410795211792,
      "eval_runtime": 13.2665,
      "eval_samples_per_second": 75.378,
      "eval_steps_per_second": 2.412,
      "step": 1800
    },
    {
      "epoch": 4.769577827908476,
      "grad_norm": 1.218343880260603,
      "learning_rate": 0.0005456255136114464,
      "loss": 4.0095,
      "step": 1850
    },
    {
      "epoch": 4.898485336770867,
      "grad_norm": 1.2906531132627939,
      "learning_rate": 0.0005251375203928073,
      "loss": 4.013,
      "step": 1900
    },
    {
      "epoch": 4.898485336770867,
      "eval_loss": 4.00699520111084,
      "eval_runtime": 13.2744,
      "eval_samples_per_second": 75.333,
      "eval_steps_per_second": 2.411,
      "step": 1900
    },
    {
      "epoch": 5.027392845633258,
      "grad_norm": 1.0003413201127862,
      "learning_rate": 0.0005046078739302906,
      "loss": 4.0152,
      "step": 1950
    },
    {
      "epoch": 5.15630035449565,
      "grad_norm": 0.7713832151052984,
      "learning_rate": 0.00048407128252072126,
      "loss": 4.0107,
      "step": 2000
    },
    {
      "epoch": 5.15630035449565,
      "eval_loss": 4.002959728240967,
      "eval_runtime": 13.3166,
      "eval_samples_per_second": 75.094,
      "eval_steps_per_second": 2.403,
      "step": 2000
    }
  ],
  "logging_steps": 50,
  "max_steps": 3870,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1778471009255424.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}