{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8514851485148514,
  "eval_steps": 3,
  "global_step": 36,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07920792079207921,
      "grad_norm": 0.6195542216300964,
      "learning_rate": 5e-05,
      "loss": 2.7446,
      "step": 1
    },
    {
      "epoch": 0.15841584158415842,
      "grad_norm": 0.6150830984115601,
      "learning_rate": 0.0001,
      "loss": 2.7192,
      "step": 2
    },
    {
      "epoch": 0.2376237623762376,
      "grad_norm": 0.6174027323722839,
      "learning_rate": 9.705882352941177e-05,
      "loss": 2.7788,
      "step": 3
    },
    {
      "epoch": 0.2376237623762376,
      "eval_loss": 2.845029830932617,
      "eval_runtime": 2.5957,
      "eval_samples_per_second": 0.771,
      "eval_steps_per_second": 0.771,
      "step": 3
    },
    {
      "epoch": 0.31683168316831684,
      "grad_norm": 0.5594128966331482,
      "learning_rate": 9.411764705882353e-05,
      "loss": 2.6762,
      "step": 4
    },
    {
      "epoch": 0.39603960396039606,
      "grad_norm": 0.4553978145122528,
      "learning_rate": 9.11764705882353e-05,
      "loss": 2.577,
      "step": 5
    },
    {
      "epoch": 0.4752475247524752,
      "grad_norm": 0.31477224826812744,
      "learning_rate": 8.823529411764706e-05,
      "loss": 2.7767,
      "step": 6
    },
    {
      "epoch": 0.4752475247524752,
      "eval_loss": 2.74169921875,
      "eval_runtime": 2.6084,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.767,
      "step": 6
    },
    {
      "epoch": 0.5544554455445545,
      "grad_norm": 0.6173616051673889,
      "learning_rate": 8.529411764705883e-05,
      "loss": 2.5605,
      "step": 7
    },
    {
      "epoch": 0.6336633663366337,
      "grad_norm": 0.7040618062019348,
      "learning_rate": 8.23529411764706e-05,
      "loss": 2.6044,
      "step": 8
    },
    {
      "epoch": 0.7128712871287128,
      "grad_norm": 0.5611338019371033,
      "learning_rate": 7.941176470588235e-05,
      "loss": 2.6618,
      "step": 9
    },
    {
      "epoch": 0.7128712871287128,
      "eval_loss": 2.7047266960144043,
      "eval_runtime": 2.596,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.77,
      "step": 9
    },
    {
      "epoch": 0.7920792079207921,
      "grad_norm": 0.4442787766456604,
      "learning_rate": 7.647058823529411e-05,
      "loss": 2.5313,
      "step": 10
    },
    {
      "epoch": 0.8712871287128713,
      "grad_norm": 0.292217880487442,
      "learning_rate": 7.352941176470589e-05,
      "loss": 2.6006,
      "step": 11
    },
    {
      "epoch": 0.9504950495049505,
      "grad_norm": 0.2892855107784271,
      "learning_rate": 7.058823529411765e-05,
      "loss": 2.4469,
      "step": 12
    },
    {
      "epoch": 0.9504950495049505,
      "eval_loss": 2.671910524368286,
      "eval_runtime": 2.5993,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.769,
      "step": 12
    },
    {
      "epoch": 1.0297029702970297,
      "grad_norm": 0.2804577350616455,
      "learning_rate": 6.764705882352942e-05,
      "loss": 2.5202,
      "step": 13
    },
    {
      "epoch": 1.108910891089109,
      "grad_norm": 0.29348304867744446,
      "learning_rate": 6.470588235294118e-05,
      "loss": 2.4257,
      "step": 14
    },
    {
      "epoch": 1.188118811881188,
      "grad_norm": 0.27386218309402466,
      "learning_rate": 6.176470588235295e-05,
      "loss": 2.4943,
      "step": 15
    },
    {
      "epoch": 1.188118811881188,
      "eval_loss": 2.6478095054626465,
      "eval_runtime": 2.6061,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.767,
      "step": 15
    },
    {
      "epoch": 1.2673267326732673,
      "grad_norm": 0.28914758563041687,
      "learning_rate": 5.882352941176471e-05,
      "loss": 2.5296,
      "step": 16
    },
    {
      "epoch": 1.3465346534653464,
      "grad_norm": 0.2648894786834717,
      "learning_rate": 5.588235294117647e-05,
      "loss": 2.5639,
      "step": 17
    },
    {
      "epoch": 1.4257425742574257,
      "grad_norm": 0.2344515174627304,
      "learning_rate": 5.294117647058824e-05,
      "loss": 2.4419,
      "step": 18
    },
    {
      "epoch": 1.4257425742574257,
      "eval_loss": 2.626154899597168,
      "eval_runtime": 2.6016,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.769,
      "step": 18
    },
    {
      "epoch": 1.504950495049505,
      "grad_norm": 0.2299778014421463,
      "learning_rate": 5e-05,
      "loss": 2.4946,
      "step": 19
    },
    {
      "epoch": 1.5841584158415842,
      "grad_norm": 0.2452971488237381,
      "learning_rate": 4.705882352941177e-05,
      "loss": 2.4017,
      "step": 20
    },
    {
      "epoch": 1.6633663366336635,
      "grad_norm": 0.2606826722621918,
      "learning_rate": 4.411764705882353e-05,
      "loss": 2.5081,
      "step": 21
    },
    {
      "epoch": 1.6633663366336635,
      "eval_loss": 2.6083157062530518,
      "eval_runtime": 2.6059,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.767,
      "step": 21
    },
    {
      "epoch": 1.7425742574257426,
      "grad_norm": 0.2621608376502991,
      "learning_rate": 4.11764705882353e-05,
      "loss": 2.5019,
      "step": 22
    },
    {
      "epoch": 1.8217821782178216,
      "grad_norm": 0.24822287261486053,
      "learning_rate": 3.8235294117647055e-05,
      "loss": 2.4289,
      "step": 23
    },
    {
      "epoch": 1.900990099009901,
      "grad_norm": 0.26251110434532166,
      "learning_rate": 3.529411764705883e-05,
      "loss": 2.4366,
      "step": 24
    },
    {
      "epoch": 1.900990099009901,
      "eval_loss": 2.592751979827881,
      "eval_runtime": 2.6021,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.769,
      "step": 24
    },
    {
      "epoch": 1.9801980198019802,
      "grad_norm": 0.29046395421028137,
      "learning_rate": 3.235294117647059e-05,
      "loss": 2.4076,
      "step": 25
    },
    {
      "epoch": 2.0594059405940595,
      "grad_norm": 0.3276838958263397,
      "learning_rate": 2.9411764705882354e-05,
      "loss": 2.3266,
      "step": 26
    },
    {
      "epoch": 2.1386138613861387,
      "grad_norm": 0.25077176094055176,
      "learning_rate": 2.647058823529412e-05,
      "loss": 2.3362,
      "step": 27
    },
    {
      "epoch": 2.1386138613861387,
      "eval_loss": 2.579805374145508,
      "eval_runtime": 2.6029,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.768,
      "step": 27
    },
    {
      "epoch": 2.217821782178218,
      "grad_norm": 0.2501392662525177,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 2.3981,
      "step": 28
    },
    {
      "epoch": 2.297029702970297,
      "grad_norm": 0.23780333995819092,
      "learning_rate": 2.058823529411765e-05,
      "loss": 2.4174,
      "step": 29
    },
    {
      "epoch": 2.376237623762376,
      "grad_norm": 0.3056803047657013,
      "learning_rate": 1.7647058823529414e-05,
      "loss": 2.3146,
      "step": 30
    },
    {
      "epoch": 2.376237623762376,
      "eval_loss": 2.5697760581970215,
      "eval_runtime": 2.6051,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.768,
      "step": 30
    },
    {
      "epoch": 2.4554455445544554,
      "grad_norm": 0.22868944704532623,
      "learning_rate": 1.4705882352941177e-05,
      "loss": 2.3982,
      "step": 31
    },
    {
      "epoch": 2.5346534653465347,
      "grad_norm": 0.23521411418914795,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 2.3992,
      "step": 32
    },
    {
      "epoch": 2.613861386138614,
      "grad_norm": 0.267884224653244,
      "learning_rate": 8.823529411764707e-06,
      "loss": 2.4405,
      "step": 33
    },
    {
      "epoch": 2.613861386138614,
      "eval_loss": 2.5635159015655518,
      "eval_runtime": 2.5955,
      "eval_samples_per_second": 0.771,
      "eval_steps_per_second": 0.771,
      "step": 33
    },
    {
      "epoch": 2.693069306930693,
      "grad_norm": 0.21910974383354187,
      "learning_rate": 5.882352941176471e-06,
      "loss": 2.4613,
      "step": 34
    },
    {
      "epoch": 2.772277227722772,
      "grad_norm": 0.262515127658844,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 2.3542,
      "step": 35
    },
    {
      "epoch": 2.8514851485148514,
      "grad_norm": 0.2211056649684906,
      "learning_rate": 0.0,
      "loss": 2.4264,
      "step": 36
    },
    {
      "epoch": 2.8514851485148514,
      "eval_loss": 2.5613255500793457,
      "eval_runtime": 2.603,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.768,
      "step": 36
    }
  ],
  "logging_steps": 1,
  "max_steps": 36,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4554329481230336e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}