{
  "best_global_step": 1000,
  "best_metric": 0.2766864597797394,
  "best_model_checkpoint": "/data/model_checkpoints/checkpoint-1000",
  "epoch": 0.8942544153811759,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0447127207690588,
      "grad_norm": 203.0,
      "learning_rate": 2.1894548704200177e-06,
      "loss": 6.45140380859375,
      "step": 50
    },
    {
      "epoch": 0.0894254415381176,
      "grad_norm": 62.5,
      "learning_rate": 4.423592493297587e-06,
      "loss": 2.8122402954101564,
      "step": 100
    },
    {
      "epoch": 0.0894254415381176,
      "eval_loss": 1.1310627460479736,
      "eval_runtime": 30.6359,
      "eval_samples_per_second": 32.478,
      "eval_steps_per_second": 16.255,
      "step": 100
    },
    {
      "epoch": 0.1341381623071764,
      "grad_norm": 67.0,
      "learning_rate": 6.6577301161751565e-06,
      "loss": 0.8708096313476562,
      "step": 150
    },
    {
      "epoch": 0.1788508830762352,
      "grad_norm": 56.75,
      "learning_rate": 8.891867739052725e-06,
      "loss": 0.5381386947631835,
      "step": 200
    },
    {
      "epoch": 0.1788508830762352,
      "eval_loss": 0.4545099437236786,
      "eval_runtime": 30.5557,
      "eval_samples_per_second": 32.564,
      "eval_steps_per_second": 16.298,
      "step": 200
    },
    {
      "epoch": 0.22356360384529397,
      "grad_norm": 50.75,
      "learning_rate": 1.1126005361930296e-05,
      "loss": 0.41018287658691405,
      "step": 250
    },
    {
      "epoch": 0.2682763246143528,
      "grad_norm": 42.0,
      "learning_rate": 1.3360142984807863e-05,
      "loss": 0.3585561370849609,
      "step": 300
    },
    {
      "epoch": 0.2682763246143528,
      "eval_loss": 0.3562505543231964,
      "eval_runtime": 30.6312,
      "eval_samples_per_second": 32.483,
      "eval_steps_per_second": 16.258,
      "step": 300
    },
    {
      "epoch": 0.3129890453834116,
      "grad_norm": 37.75,
      "learning_rate": 1.5594280607685434e-05,
      "loss": 0.3467108154296875,
      "step": 350
    },
    {
      "epoch": 0.3577017661524704,
      "grad_norm": 25.25,
      "learning_rate": 1.7828418230563005e-05,
      "loss": 0.33910282135009767,
      "step": 400
    },
    {
      "epoch": 0.3577017661524704,
      "eval_loss": 0.3491951823234558,
      "eval_runtime": 30.5449,
      "eval_samples_per_second": 32.575,
      "eval_steps_per_second": 16.304,
      "step": 400
    },
    {
      "epoch": 0.4024144869215292,
      "grad_norm": 30.25,
      "learning_rate": 2.0062555853440572e-05,
      "loss": 0.35506061553955076,
      "step": 450
    },
    {
      "epoch": 0.44712720769058795,
      "grad_norm": 41.0,
      "learning_rate": 2.2296693476318143e-05,
      "loss": 0.3203820037841797,
      "step": 500
    },
    {
      "epoch": 0.44712720769058795,
      "eval_loss": 0.33218100666999817,
      "eval_runtime": 30.4848,
      "eval_samples_per_second": 32.639,
      "eval_steps_per_second": 16.336,
      "step": 500
    },
    {
      "epoch": 0.49183992845964675,
      "grad_norm": 23.625,
      "learning_rate": 2.453083109919571e-05,
      "loss": 0.30755191802978515,
      "step": 550
    },
    {
      "epoch": 0.5365526492287056,
      "grad_norm": 20.5,
      "learning_rate": 2.676496872207328e-05,
      "loss": 0.30894683837890624,
      "step": 600
    },
    {
      "epoch": 0.5365526492287056,
      "eval_loss": 0.30529576539993286,
      "eval_runtime": 30.4934,
      "eval_samples_per_second": 32.63,
      "eval_steps_per_second": 16.331,
      "step": 600
    },
    {
      "epoch": 0.5812653699977643,
      "grad_norm": 31.125,
      "learning_rate": 2.899910634495085e-05,
      "loss": 0.30126373291015623,
      "step": 650
    },
    {
      "epoch": 0.6259780907668232,
      "grad_norm": 20.0,
      "learning_rate": 3.123324396782842e-05,
      "loss": 0.34950965881347656,
      "step": 700
    },
    {
      "epoch": 0.6259780907668232,
      "eval_loss": 0.3235093951225281,
      "eval_runtime": 30.5199,
      "eval_samples_per_second": 32.602,
      "eval_steps_per_second": 16.317,
      "step": 700
    },
    {
      "epoch": 0.670690811535882,
      "grad_norm": 18.125,
      "learning_rate": 3.346738159070599e-05,
      "loss": 0.31941865921020507,
      "step": 750
    },
    {
      "epoch": 0.7154035323049408,
      "grad_norm": 21.75,
      "learning_rate": 3.5701519213583554e-05,
      "loss": 0.2860032844543457,
      "step": 800
    },
    {
      "epoch": 0.7154035323049408,
      "eval_loss": 0.27899640798568726,
      "eval_runtime": 30.3648,
      "eval_samples_per_second": 32.768,
      "eval_steps_per_second": 16.401,
      "step": 800
    },
    {
      "epoch": 0.7601162530739995,
      "grad_norm": 11.625,
      "learning_rate": 3.7935656836461125e-05,
      "loss": 0.2760417366027832,
      "step": 850
    },
    {
      "epoch": 0.8048289738430584,
      "grad_norm": 17.25,
      "learning_rate": 4.01697944593387e-05,
      "loss": 0.24729475021362304,
      "step": 900
    },
    {
      "epoch": 0.8048289738430584,
      "eval_loss": 0.29079964756965637,
      "eval_runtime": 30.3913,
      "eval_samples_per_second": 32.74,
      "eval_steps_per_second": 16.386,
      "step": 900
    },
    {
      "epoch": 0.8495416946121171,
      "grad_norm": 15.6875,
      "learning_rate": 4.240393208221627e-05,
      "loss": 0.2624860382080078,
      "step": 950
    },
    {
      "epoch": 0.8942544153811759,
      "grad_norm": 42.75,
      "learning_rate": 4.463806970509384e-05,
      "loss": 0.27374263763427736,
      "step": 1000
    },
    {
      "epoch": 0.8942544153811759,
      "eval_loss": 0.2766864597797394,
      "eval_runtime": 30.3793,
      "eval_samples_per_second": 32.753,
      "eval_steps_per_second": 16.393,
      "step": 1000
    }
  ],
  "logging_steps": 50,
  "max_steps": 11190,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.980791074558669e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}