{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05937654626422563,
  "eval_steps": 3,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0019792182088075212,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 8.076898574829102,
      "num_input_tokens_seen": 2942,
      "step": 1,
      "train_runtime": 697.1739,
      "train_tokens_per_second": 4.22
    },
    {
      "epoch": 0.0039584364176150424,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 6.006780624389648,
      "num_input_tokens_seen": 8009,
      "step": 2,
      "train_runtime": 724.0489,
      "train_tokens_per_second": 11.061
    },
    {
      "epoch": 0.005937654626422563,
      "grad_norm": 0.33343738317489624,
      "learning_rate": 0.0,
      "loss": 9.303730964660645,
      "num_input_tokens_seen": 9957,
      "step": 3,
      "train_runtime": 729.6177,
      "train_tokens_per_second": 13.647
    },
    {
      "epoch": 0.005937654626422563,
      "eval_loss": 1.6509373188018799,
      "eval_runtime": 39.0128,
      "eval_samples_per_second": 3.281,
      "eval_steps_per_second": 0.82,
      "num_input_tokens_seen": 9957,
      "step": 3
    },
    {
      "epoch": 0.007916872835230085,
      "grad_norm": 0.35222020745277405,
      "learning_rate": 1e-05,
      "loss": 8.339298248291016,
      "num_input_tokens_seen": 12579,
      "step": 4,
      "train_runtime": 775.7485,
      "train_tokens_per_second": 16.215
    },
    {
      "epoch": 0.009896091044037604,
      "grad_norm": 0.4923153519630432,
      "learning_rate": 2e-05,
      "loss": 5.858340263366699,
      "num_input_tokens_seen": 18548,
      "step": 5,
      "train_runtime": 791.4629,
      "train_tokens_per_second": 23.435
    },
    {
      "epoch": 0.011875309252845126,
      "grad_norm": 0.4596147835254669,
      "learning_rate": 3e-05,
      "loss": 7.152117729187012,
      "num_input_tokens_seen": 22715,
      "step": 6,
      "train_runtime": 802.7905,
      "train_tokens_per_second": 28.295
    },
    {
      "epoch": 0.011875309252845126,
      "eval_loss": 1.6447160243988037,
      "eval_runtime": 38.6774,
      "eval_samples_per_second": 3.309,
      "eval_steps_per_second": 0.827,
      "num_input_tokens_seen": 22715,
      "step": 6
    },
    {
      "epoch": 0.013854527461652647,
      "grad_norm": 0.4033356308937073,
      "learning_rate": 4e-05,
      "loss": 8.132979393005371,
      "num_input_tokens_seen": 26012,
      "step": 7,
      "train_runtime": 850.5164,
      "train_tokens_per_second": 30.584
    },
    {
      "epoch": 0.01583374567046017,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 5.526519775390625,
      "num_input_tokens_seen": 32601,
      "step": 8,
      "train_runtime": 867.4081,
      "train_tokens_per_second": 37.584
    },
    {
      "epoch": 0.017812963879267688,
      "grad_norm": 0.4410555064678192,
      "learning_rate": 5e-05,
      "loss": 6.735212326049805,
      "num_input_tokens_seen": 36772,
      "step": 9,
      "train_runtime": 878.5352,
      "train_tokens_per_second": 41.856
    },
    {
      "epoch": 0.017812963879267688,
      "eval_loss": 1.6224464178085327,
      "eval_runtime": 38.4019,
      "eval_samples_per_second": 3.333,
      "eval_steps_per_second": 0.833,
      "num_input_tokens_seen": 36772,
      "step": 9
    },
    {
      "epoch": 0.01979218208807521,
      "grad_norm": 0.41144317388534546,
      "learning_rate": 4.8e-05,
      "loss": 7.8381853103637695,
      "num_input_tokens_seen": 41799,
      "step": 10,
      "train_runtime": 929.9841,
      "train_tokens_per_second": 44.946
    },
    {
      "epoch": 0.02177140029688273,
      "grad_norm": 0.49192050099372864,
      "learning_rate": 4.600000000000001e-05,
      "loss": 8.786320686340332,
      "num_input_tokens_seen": 44038,
      "step": 11,
      "train_runtime": 936.1529,
      "train_tokens_per_second": 47.041
    },
    {
      "epoch": 0.02375061850569025,
      "grad_norm": 0.3306628465652466,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 8.384254455566406,
      "num_input_tokens_seen": 46086,
      "step": 12,
      "train_runtime": 941.817,
      "train_tokens_per_second": 48.933
    },
    {
      "epoch": 0.02375061850569025,
      "eval_loss": 1.5842628479003906,
      "eval_runtime": 38.3991,
      "eval_samples_per_second": 3.333,
      "eval_steps_per_second": 0.833,
      "num_input_tokens_seen": 46086,
      "step": 12
    },
    {
      "epoch": 0.025729836714497772,
      "grad_norm": 0.4944745898246765,
      "learning_rate": 4.2e-05,
      "loss": 8.164417266845703,
      "num_input_tokens_seen": 49507,
      "step": 13,
      "train_runtime": 989.3818,
      "train_tokens_per_second": 50.038
    },
    {
      "epoch": 0.027709054923305294,
      "grad_norm": 0.3035735785961151,
      "learning_rate": 4e-05,
      "loss": 8.058252334594727,
      "num_input_tokens_seen": 52233,
      "step": 14,
      "train_runtime": 996.6464,
      "train_tokens_per_second": 52.409
    },
    {
      "epoch": 0.029688273132112815,
      "grad_norm": 0.33408379554748535,
      "learning_rate": 3.8e-05,
      "loss": 7.010341167449951,
      "num_input_tokens_seen": 55234,
      "step": 15,
      "train_runtime": 1004.6341,
      "train_tokens_per_second": 54.979
    },
    {
      "epoch": 0.029688273132112815,
      "eval_loss": 1.5526598691940308,
      "eval_runtime": 38.2948,
      "eval_samples_per_second": 3.342,
      "eval_steps_per_second": 0.836,
      "num_input_tokens_seen": 55234,
      "step": 15
    },
    {
      "epoch": 0.03166749134092034,
      "grad_norm": 0.28500697016716003,
      "learning_rate": 3.6e-05,
      "loss": 7.41168737411499,
      "num_input_tokens_seen": 58400,
      "step": 16,
      "train_runtime": 1051.3309,
      "train_tokens_per_second": 55.549
    },
    {
      "epoch": 0.03364670954972786,
      "grad_norm": 0.5121440887451172,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 7.725010395050049,
      "num_input_tokens_seen": 62169,
      "step": 17,
      "train_runtime": 1061.4008,
      "train_tokens_per_second": 58.573
    },
    {
      "epoch": 0.035625927758535375,
      "grad_norm": 0.6245442628860474,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 6.330207347869873,
      "num_input_tokens_seen": 66801,
      "step": 18,
      "train_runtime": 1073.4796,
      "train_tokens_per_second": 62.228
    },
    {
      "epoch": 0.035625927758535375,
      "eval_loss": 1.5273845195770264,
      "eval_runtime": 38.4036,
      "eval_samples_per_second": 3.333,
      "eval_steps_per_second": 0.833,
      "num_input_tokens_seen": 66801,
      "step": 18
    },
    {
      "epoch": 0.0376051459673429,
      "grad_norm": 0.3218696117401123,
      "learning_rate": 3e-05,
      "loss": 7.721973419189453,
      "num_input_tokens_seen": 68693,
      "step": 19,
      "train_runtime": 1117.2758,
      "train_tokens_per_second": 61.483
    },
    {
      "epoch": 0.03958436417615042,
      "grad_norm": 0.4316211938858032,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 5.789700031280518,
      "num_input_tokens_seen": 74505,
      "step": 20,
      "train_runtime": 1132.4346,
      "train_tokens_per_second": 65.792
    },
    {
      "epoch": 0.04156358238495794,
      "grad_norm": 0.36192160844802856,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 5.359489440917969,
      "num_input_tokens_seen": 82017,
      "step": 21,
      "train_runtime": 1151.4895,
      "train_tokens_per_second": 71.227
    },
    {
      "epoch": 0.04156358238495794,
      "eval_loss": 1.5085684061050415,
      "eval_runtime": 38.2169,
      "eval_samples_per_second": 3.349,
      "eval_steps_per_second": 0.837,
      "num_input_tokens_seen": 82017,
      "step": 21
    },
    {
      "epoch": 0.04354280059376546,
      "grad_norm": 0.35479193925857544,
      "learning_rate": 2.4e-05,
      "loss": 7.415735244750977,
      "num_input_tokens_seen": 84672,
      "step": 22,
      "train_runtime": 1197.1732,
      "train_tokens_per_second": 70.727
    },
    {
      "epoch": 0.045522018802572985,
      "grad_norm": 0.3102089464664459,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 5.640255928039551,
      "num_input_tokens_seen": 88890,
      "step": 23,
      "train_runtime": 1208.5904,
      "train_tokens_per_second": 73.548
    },
    {
      "epoch": 0.0475012370113805,
      "grad_norm": 0.31496089696884155,
      "learning_rate": 2e-05,
      "loss": 7.0548906326293945,
      "num_input_tokens_seen": 92773,
      "step": 24,
      "train_runtime": 1218.7108,
      "train_tokens_per_second": 76.124
    },
    {
      "epoch": 0.0475012370113805,
      "eval_loss": 1.495510220527649,
      "eval_runtime": 38.2529,
      "eval_samples_per_second": 3.346,
      "eval_steps_per_second": 0.837,
      "num_input_tokens_seen": 92773,
      "step": 24
    },
    {
      "epoch": 0.04948045522018803,
      "grad_norm": 0.3281867206096649,
      "learning_rate": 1.8e-05,
      "loss": 7.233792304992676,
      "num_input_tokens_seen": 95385,
      "step": 25,
      "train_runtime": 1263.8477,
      "train_tokens_per_second": 75.472
    },
    {
      "epoch": 0.051459673428995545,
      "grad_norm": 0.3076775372028351,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 6.263100624084473,
      "num_input_tokens_seen": 99968,
      "step": 26,
      "train_runtime": 1275.9673,
      "train_tokens_per_second": 78.347
    },
    {
      "epoch": 0.05343889163780307,
      "grad_norm": 0.36238089203834534,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 7.531805515289307,
      "num_input_tokens_seen": 103394,
      "step": 27,
      "train_runtime": 1285.2238,
      "train_tokens_per_second": 80.448
    },
    {
      "epoch": 0.05343889163780307,
      "eval_loss": 1.4868113994598389,
      "eval_runtime": 38.2858,
      "eval_samples_per_second": 3.343,
      "eval_steps_per_second": 0.836,
      "num_input_tokens_seen": 103394,
      "step": 27
    },
    {
      "epoch": 0.05541810984661059,
      "grad_norm": 0.36287209391593933,
      "learning_rate": 1.2e-05,
      "loss": 7.41123104095459,
      "num_input_tokens_seen": 107552,
      "step": 28,
      "train_runtime": 1334.3841,
      "train_tokens_per_second": 80.6
    },
    {
      "epoch": 0.05739732805541811,
      "grad_norm": 0.33840489387512207,
      "learning_rate": 1e-05,
      "loss": 8.186224937438965,
      "num_input_tokens_seen": 109305,
      "step": 29,
      "train_runtime": 1339.2503,
      "train_tokens_per_second": 81.617
    },
    {
      "epoch": 0.05937654626422563,
      "grad_norm": 0.29359596967697144,
      "learning_rate": 8.000000000000001e-06,
      "loss": 5.34782075881958,
      "num_input_tokens_seen": 113042,
      "step": 30,
      "train_runtime": 1349.1236,
      "train_tokens_per_second": 83.789
    },
    {
      "epoch": 0.05937654626422563,
      "eval_loss": 1.4815630912780762,
      "eval_runtime": 38.3292,
      "eval_samples_per_second": 3.339,
      "eval_steps_per_second": 0.835,
      "num_input_tokens_seen": 113042,
      "step": 30
    }
  ],
  "logging_steps": 1,
  "max_steps": 30,
  "num_input_tokens_seen": 113042,
  "num_train_epochs": 1,
  "save_steps": 30,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2423146296844416.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}