{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.018703824932198635,
  "eval_steps": 4,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00037407649864397267,
      "grad_norm": 9.773622512817383,
      "learning_rate": 2e-05,
      "loss": 4.8339,
      "step": 1
    },
    {
      "epoch": 0.00037407649864397267,
      "eval_loss": NaN,
      "eval_runtime": 22.0999,
      "eval_samples_per_second": 25.475,
      "eval_steps_per_second": 25.475,
      "step": 1
    },
    {
      "epoch": 0.0007481529972879453,
      "grad_norm": 8.580421447753906,
      "learning_rate": 4e-05,
      "loss": 4.5196,
      "step": 2
    },
    {
      "epoch": 0.0011222294959319181,
      "grad_norm": 8.180211067199707,
      "learning_rate": 6e-05,
      "loss": 4.344,
      "step": 3
    },
    {
      "epoch": 0.0014963059945758907,
      "grad_norm": 10.246196746826172,
      "learning_rate": 8e-05,
      "loss": 4.743,
      "step": 4
    },
    {
      "epoch": 0.0014963059945758907,
      "eval_loss": NaN,
      "eval_runtime": 21.684,
      "eval_samples_per_second": 25.964,
      "eval_steps_per_second": 25.964,
      "step": 4
    },
    {
      "epoch": 0.0018703824932198635,
      "grad_norm": 9.82059383392334,
      "learning_rate": 0.0001,
      "loss": 4.1244,
      "step": 5
    },
    {
      "epoch": 0.0022444589918638362,
      "grad_norm": 10.244694709777832,
      "learning_rate": 0.00012,
      "loss": 3.5537,
      "step": 6
    },
    {
      "epoch": 0.002618535490507809,
      "grad_norm": 9.037696838378906,
      "learning_rate": 0.00014,
      "loss": 2.7798,
      "step": 7
    },
    {
      "epoch": 0.0029926119891517814,
      "grad_norm": 10.064749717712402,
      "learning_rate": 0.00016,
      "loss": 2.0157,
      "step": 8
    },
    {
      "epoch": 0.0029926119891517814,
      "eval_loss": NaN,
      "eval_runtime": 21.7463,
      "eval_samples_per_second": 25.889,
      "eval_steps_per_second": 25.889,
      "step": 8
    },
    {
      "epoch": 0.0033666884877957544,
      "grad_norm": 10.570599555969238,
      "learning_rate": 0.00018,
      "loss": 1.5668,
      "step": 9
    },
    {
      "epoch": 0.003740764986439727,
      "grad_norm": 14.205665588378906,
      "learning_rate": 0.0002,
      "loss": 1.1739,
      "step": 10
    },
    {
      "epoch": 0.0041148414850837,
      "grad_norm": 11.37263011932373,
      "learning_rate": 0.0001996917333733128,
      "loss": 1.1603,
      "step": 11
    },
    {
      "epoch": 0.0044889179837276725,
      "grad_norm": 5.627699851989746,
      "learning_rate": 0.00019876883405951377,
      "loss": 0.8112,
      "step": 12
    },
    {
      "epoch": 0.0044889179837276725,
      "eval_loss": NaN,
      "eval_runtime": 21.8678,
      "eval_samples_per_second": 25.746,
      "eval_steps_per_second": 25.746,
      "step": 12
    },
    {
      "epoch": 0.004862994482371645,
      "grad_norm": 12.806142807006836,
      "learning_rate": 0.00019723699203976766,
      "loss": 0.7784,
      "step": 13
    },
    {
      "epoch": 0.005237070981015618,
      "grad_norm": 8.642436981201172,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.8493,
      "step": 14
    },
    {
      "epoch": 0.00561114747965959,
      "grad_norm": 6.148828506469727,
      "learning_rate": 0.0001923879532511287,
      "loss": 0.656,
      "step": 15
    },
    {
      "epoch": 0.005985223978303563,
      "grad_norm": 8.592019081115723,
      "learning_rate": 0.0001891006524188368,
      "loss": 1.0952,
      "step": 16
    },
    {
      "epoch": 0.005985223978303563,
      "eval_loss": NaN,
      "eval_runtime": 21.7489,
      "eval_samples_per_second": 25.886,
      "eval_steps_per_second": 25.886,
      "step": 16
    },
    {
      "epoch": 0.006359300476947536,
      "grad_norm": 2.953862190246582,
      "learning_rate": 0.00018526401643540922,
      "loss": 0.7376,
      "step": 17
    },
    {
      "epoch": 0.006733376975591509,
      "grad_norm": 10.415821075439453,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.9074,
      "step": 18
    },
    {
      "epoch": 0.007107453474235481,
      "grad_norm": 8.800667762756348,
      "learning_rate": 0.0001760405965600031,
      "loss": 0.9933,
      "step": 19
    },
    {
      "epoch": 0.007481529972879454,
      "grad_norm": 5.8638386726379395,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.8282,
      "step": 20
    },
    {
      "epoch": 0.007481529972879454,
      "eval_loss": NaN,
      "eval_runtime": 21.7567,
      "eval_samples_per_second": 25.877,
      "eval_steps_per_second": 25.877,
      "step": 20
    },
    {
      "epoch": 0.007855606471523427,
      "grad_norm": 6.939398765563965,
      "learning_rate": 0.00016494480483301836,
      "loss": 0.9589,
      "step": 21
    },
    {
      "epoch": 0.0082296829701674,
      "grad_norm": 5.484954357147217,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.7741,
      "step": 22
    },
    {
      "epoch": 0.008603759468811372,
      "grad_norm": 2.173372983932495,
      "learning_rate": 0.0001522498564715949,
      "loss": 0.7695,
      "step": 23
    },
    {
      "epoch": 0.008977835967455345,
      "grad_norm": 2.9671242237091064,
      "learning_rate": 0.00014539904997395468,
      "loss": 0.7306,
      "step": 24
    },
    {
      "epoch": 0.008977835967455345,
      "eval_loss": NaN,
      "eval_runtime": 21.7415,
      "eval_samples_per_second": 25.895,
      "eval_steps_per_second": 25.895,
      "step": 24
    },
    {
      "epoch": 0.009351912466099318,
      "grad_norm": 4.0381903648376465,
      "learning_rate": 0.000138268343236509,
      "loss": 0.7789,
      "step": 25
    },
    {
      "epoch": 0.00972598896474329,
      "grad_norm": 4.545593738555908,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.769,
      "step": 26
    },
    {
      "epoch": 0.010100065463387263,
      "grad_norm": 2.968888521194458,
      "learning_rate": 0.00012334453638559057,
      "loss": 0.8867,
      "step": 27
    },
    {
      "epoch": 0.010474141962031235,
      "grad_norm": 2.718158006668091,
      "learning_rate": 0.0001156434465040231,
      "loss": 0.8066,
      "step": 28
    },
    {
      "epoch": 0.010474141962031235,
      "eval_loss": NaN,
      "eval_runtime": 21.8902,
      "eval_samples_per_second": 25.719,
      "eval_steps_per_second": 25.719,
      "step": 28
    },
    {
      "epoch": 0.010848218460675208,
      "grad_norm": 4.827681541442871,
      "learning_rate": 0.0001078459095727845,
      "loss": 0.713,
      "step": 29
    },
    {
      "epoch": 0.01122229495931918,
      "grad_norm": 2.4585800170898438,
      "learning_rate": 0.0001,
      "loss": 0.82,
      "step": 30
    },
    {
      "epoch": 0.011596371457963153,
      "grad_norm": 3.3518271446228027,
      "learning_rate": 9.215409042721552e-05,
      "loss": 0.6706,
      "step": 31
    },
    {
      "epoch": 0.011970447956607125,
      "grad_norm": 1.3497822284698486,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.7892,
      "step": 32
    },
    {
      "epoch": 0.011970447956607125,
      "eval_loss": NaN,
      "eval_runtime": 22.3464,
      "eval_samples_per_second": 25.194,
      "eval_steps_per_second": 25.194,
      "step": 32
    },
    {
      "epoch": 0.012344524455251098,
      "grad_norm": 2.0146560668945312,
      "learning_rate": 7.66554636144095e-05,
      "loss": 0.8602,
      "step": 33
    },
    {
      "epoch": 0.012718600953895072,
      "grad_norm": 2.6399552822113037,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.6844,
      "step": 34
    },
    {
      "epoch": 0.013092677452539045,
      "grad_norm": 2.6146562099456787,
      "learning_rate": 6.173165676349103e-05,
      "loss": 0.6531,
      "step": 35
    },
    {
      "epoch": 0.013466753951183017,
      "grad_norm": 3.0831737518310547,
      "learning_rate": 5.4600950026045326e-05,
      "loss": 0.8027,
      "step": 36
    },
    {
      "epoch": 0.013466753951183017,
      "eval_loss": NaN,
      "eval_runtime": 21.7733,
      "eval_samples_per_second": 25.857,
      "eval_steps_per_second": 25.857,
      "step": 36
    },
    {
      "epoch": 0.01384083044982699,
      "grad_norm": 2.011878252029419,
      "learning_rate": 4.7750143528405126e-05,
      "loss": 0.6398,
      "step": 37
    },
    {
      "epoch": 0.014214906948470963,
      "grad_norm": 2.133044958114624,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.7345,
      "step": 38
    },
    {
      "epoch": 0.014588983447114935,
      "grad_norm": 3.275799036026001,
      "learning_rate": 3.5055195166981645e-05,
      "loss": 0.5697,
      "step": 39
    },
    {
      "epoch": 0.014963059945758908,
      "grad_norm": 1.622467279434204,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 0.5799,
      "step": 40
    },
    {
      "epoch": 0.014963059945758908,
      "eval_loss": NaN,
      "eval_runtime": 21.7692,
      "eval_samples_per_second": 25.862,
      "eval_steps_per_second": 25.862,
      "step": 40
    },
    {
      "epoch": 0.01533713644440288,
      "grad_norm": 3.586545944213867,
      "learning_rate": 2.3959403439996907e-05,
      "loss": 0.7168,
      "step": 41
    },
    {
      "epoch": 0.015711212943046855,
      "grad_norm": 3.455106496810913,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.6883,
      "step": 42
    },
    {
      "epoch": 0.016085289441690825,
      "grad_norm": 1.5402703285217285,
      "learning_rate": 1.4735983564590783e-05,
      "loss": 0.6855,
      "step": 43
    },
    {
      "epoch": 0.0164593659403348,
      "grad_norm": 3.069215774536133,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.4914,
      "step": 44
    },
    {
      "epoch": 0.0164593659403348,
      "eval_loss": NaN,
      "eval_runtime": 21.7546,
      "eval_samples_per_second": 25.88,
      "eval_steps_per_second": 25.88,
      "step": 44
    },
    {
      "epoch": 0.01683344243897877,
      "grad_norm": 3.156932830810547,
      "learning_rate": 7.612046748871327e-06,
      "loss": 0.7092,
      "step": 45
    },
    {
      "epoch": 0.017207518937622745,
      "grad_norm": 2.049065589904785,
      "learning_rate": 4.8943483704846475e-06,
      "loss": 0.572,
      "step": 46
    },
    {
      "epoch": 0.017581595436266716,
      "grad_norm": 2.168691396713257,
      "learning_rate": 2.7630079602323442e-06,
      "loss": 0.6909,
      "step": 47
    },
    {
      "epoch": 0.01795567193491069,
      "grad_norm": 3.115194320678711,
      "learning_rate": 1.231165940486234e-06,
      "loss": 0.6867,
      "step": 48
    },
    {
      "epoch": 0.01795567193491069,
      "eval_loss": NaN,
      "eval_runtime": 21.7292,
      "eval_samples_per_second": 25.91,
      "eval_steps_per_second": 25.91,
      "step": 48
    },
    {
      "epoch": 0.01832974843355466,
      "grad_norm": 1.8398240804672241,
      "learning_rate": 3.0826662668720364e-07,
      "loss": 0.7688,
      "step": 49
    },
    {
      "epoch": 0.018703824932198635,
      "grad_norm": 2.1569676399230957,
      "learning_rate": 0.0,
      "loss": 0.636,
      "step": 50
    }
  ],
  "logging_steps": 1,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 4,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2419306423910400.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}