{
  "best_global_step": 162,
  "best_metric": 0.24394482374191284,
  "best_model_checkpoint": "trainer_output/checkpoint-162",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 405,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12345679012345678,
      "grad_norm": 10.78885269165039,
      "learning_rate": 1.9555555555555557e-05,
      "loss": 1.2609,
      "num_tokens": 2954.0,
      "step": 10
    },
    {
      "epoch": 0.24691358024691357,
      "grad_norm": 15.32591438293457,
      "learning_rate": 1.906172839506173e-05,
      "loss": 0.2937,
      "num_tokens": 5841.0,
      "step": 20
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 9.532804489135742,
      "learning_rate": 1.85679012345679e-05,
      "loss": 0.3133,
      "num_tokens": 8816.0,
      "step": 30
    },
    {
      "epoch": 0.49382716049382713,
      "grad_norm": 7.899057865142822,
      "learning_rate": 1.8074074074074074e-05,
      "loss": 0.2841,
      "num_tokens": 11751.0,
      "step": 40
    },
    {
      "epoch": 0.6172839506172839,
      "grad_norm": 7.562534332275391,
      "learning_rate": 1.7580246913580247e-05,
      "loss": 0.2418,
      "num_tokens": 14693.0,
      "step": 50
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 5.633542537689209,
      "learning_rate": 1.708641975308642e-05,
      "loss": 0.2967,
      "num_tokens": 17674.0,
      "step": 60
    },
    {
      "epoch": 0.8641975308641975,
      "grad_norm": 8.396917343139648,
      "learning_rate": 1.6592592592592594e-05,
      "loss": 0.2145,
      "num_tokens": 20612.0,
      "step": 70
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 4.925112247467041,
      "learning_rate": 1.6098765432098767e-05,
      "loss": 0.2634,
      "num_tokens": 23597.0,
      "step": 80
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.2501472234725952,
      "eval_num_tokens": 23906.0,
      "eval_runtime": 0.396,
      "eval_samples_per_second": 181.836,
      "eval_steps_per_second": 22.73,
      "step": 81
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 8.67037296295166,
      "learning_rate": 1.560493827160494e-05,
      "loss": 0.1568,
      "num_tokens": 26580.0,
      "step": 90
    },
    {
      "epoch": 1.2345679012345678,
      "grad_norm": 5.7295098304748535,
      "learning_rate": 1.5111111111111112e-05,
      "loss": 0.1577,
      "num_tokens": 29566.0,
      "step": 100
    },
    {
      "epoch": 1.3580246913580247,
      "grad_norm": 5.680588245391846,
      "learning_rate": 1.4617283950617284e-05,
      "loss": 0.1314,
      "num_tokens": 32458.0,
      "step": 110
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 4.126427173614502,
      "learning_rate": 1.4123456790123457e-05,
      "loss": 0.1553,
      "num_tokens": 35447.0,
      "step": 120
    },
    {
      "epoch": 1.6049382716049383,
      "grad_norm": 5.138214588165283,
      "learning_rate": 1.362962962962963e-05,
      "loss": 0.1172,
      "num_tokens": 38392.0,
      "step": 130
    },
    {
      "epoch": 1.7283950617283952,
      "grad_norm": 6.607668399810791,
      "learning_rate": 1.3135802469135804e-05,
      "loss": 0.136,
      "num_tokens": 41302.0,
      "step": 140
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 3.6715426445007324,
      "learning_rate": 1.2641975308641976e-05,
      "loss": 0.1304,
      "num_tokens": 44248.0,
      "step": 150
    },
    {
      "epoch": 1.9753086419753085,
      "grad_norm": 3.3442978858947754,
      "learning_rate": 1.2148148148148149e-05,
      "loss": 0.1309,
      "num_tokens": 47224.0,
      "step": 160
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.24394482374191284,
      "eval_num_tokens": 47812.0,
      "eval_runtime": 0.3659,
      "eval_samples_per_second": 196.773,
      "eval_steps_per_second": 24.597,
      "step": 162
    },
    {
      "epoch": 2.0987654320987654,
      "grad_norm": 3.3987321853637695,
      "learning_rate": 1.1654320987654322e-05,
      "loss": 0.0948,
      "num_tokens": 50182.0,
      "step": 170
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 2.5394458770751953,
      "learning_rate": 1.1160493827160494e-05,
      "loss": 0.0914,
      "num_tokens": 53168.0,
      "step": 180
    },
    {
      "epoch": 2.3456790123456788,
      "grad_norm": 4.553636074066162,
      "learning_rate": 1.0666666666666667e-05,
      "loss": 0.076,
      "num_tokens": 56109.0,
      "step": 190
    },
    {
      "epoch": 2.4691358024691357,
      "grad_norm": 11.147275924682617,
      "learning_rate": 1.017283950617284e-05,
      "loss": 0.0807,
      "num_tokens": 59056.0,
      "step": 200
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 4.174232482910156,
      "learning_rate": 9.679012345679012e-06,
      "loss": 0.0669,
      "num_tokens": 62001.0,
      "step": 210
    },
    {
      "epoch": 2.7160493827160495,
      "grad_norm": 8.098738670349121,
      "learning_rate": 9.185185185185186e-06,
      "loss": 0.0772,
      "num_tokens": 64924.0,
      "step": 220
    },
    {
      "epoch": 2.8395061728395063,
      "grad_norm": 5.179749011993408,
      "learning_rate": 8.691358024691359e-06,
      "loss": 0.0795,
      "num_tokens": 67919.0,
      "step": 230
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 4.310792446136475,
      "learning_rate": 8.197530864197532e-06,
      "loss": 0.0795,
      "num_tokens": 70830.0,
      "step": 240
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.2799757719039917,
      "eval_num_tokens": 71718.0,
      "eval_runtime": 0.3677,
      "eval_samples_per_second": 195.795,
      "eval_steps_per_second": 24.474,
      "step": 243
    },
    {
      "epoch": 3.0864197530864197,
      "grad_norm": 4.194922924041748,
      "learning_rate": 7.703703703703704e-06,
      "loss": 0.0585,
      "num_tokens": 73826.0,
      "step": 250
    },
    {
      "epoch": 3.2098765432098766,
      "grad_norm": 2.2344257831573486,
      "learning_rate": 7.209876543209877e-06,
      "loss": 0.0453,
      "num_tokens": 76798.0,
      "step": 260
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 6.8052449226379395,
      "learning_rate": 6.71604938271605e-06,
      "loss": 0.0321,
      "num_tokens": 79712.0,
      "step": 270
    },
    {
      "epoch": 3.45679012345679,
      "grad_norm": 1.2683576345443726,
      "learning_rate": 6.222222222222223e-06,
      "loss": 0.0385,
      "num_tokens": 82643.0,
      "step": 280
    },
    {
      "epoch": 3.580246913580247,
      "grad_norm": 4.096933841705322,
      "learning_rate": 5.728395061728396e-06,
      "loss": 0.0314,
      "num_tokens": 85536.0,
      "step": 290
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 4.044299125671387,
      "learning_rate": 5.234567901234568e-06,
      "loss": 0.037,
      "num_tokens": 88507.0,
      "step": 300
    },
    {
      "epoch": 3.8271604938271606,
      "grad_norm": 4.812356472015381,
      "learning_rate": 4.7407407407407415e-06,
      "loss": 0.0488,
      "num_tokens": 91478.0,
      "step": 310
    },
    {
      "epoch": 3.950617283950617,
      "grad_norm": 3.086695909500122,
      "learning_rate": 4.246913580246914e-06,
      "loss": 0.0448,
      "num_tokens": 94424.0,
      "step": 320
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.344700425863266,
      "eval_num_tokens": 95624.0,
      "eval_runtime": 0.3661,
      "eval_samples_per_second": 196.678,
      "eval_steps_per_second": 24.585,
      "step": 324
    },
    {
      "epoch": 4.074074074074074,
      "grad_norm": 2.5405914783477783,
      "learning_rate": 3.753086419753087e-06,
      "loss": 0.023,
      "num_tokens": 97401.0,
      "step": 330
    },
    {
      "epoch": 4.197530864197531,
      "grad_norm": 0.751119077205658,
      "learning_rate": 3.25925925925926e-06,
      "loss": 0.0104,
      "num_tokens": 100338.0,
      "step": 340
    },
    {
      "epoch": 4.320987654320987,
      "grad_norm": 0.3113096356391907,
      "learning_rate": 2.765432098765432e-06,
      "loss": 0.0088,
      "num_tokens": 103320.0,
      "step": 350
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 0.25087428092956543,
      "learning_rate": 2.2716049382716052e-06,
      "loss": 0.009,
      "num_tokens": 106228.0,
      "step": 360
    },
    {
      "epoch": 4.567901234567901,
      "grad_norm": 1.1306395530700684,
      "learning_rate": 1.777777777777778e-06,
      "loss": 0.0126,
      "num_tokens": 109170.0,
      "step": 370
    },
    {
      "epoch": 4.6913580246913575,
      "grad_norm": 5.73167085647583,
      "learning_rate": 1.2839506172839509e-06,
      "loss": 0.0098,
      "num_tokens": 112126.0,
      "step": 380
    },
    {
      "epoch": 4.814814814814815,
      "grad_norm": 0.4754239022731781,
      "learning_rate": 7.901234567901235e-07,
      "loss": 0.0119,
      "num_tokens": 115124.0,
      "step": 390
    },
    {
      "epoch": 4.938271604938271,
      "grad_norm": 1.1381328105926514,
      "learning_rate": 2.9629629629629634e-07,
      "loss": 0.0097,
      "num_tokens": 118062.0,
      "step": 400
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.36827170848846436,
      "eval_num_tokens": 119530.0,
      "eval_runtime": 0.3639,
      "eval_samples_per_second": 197.856,
      "eval_steps_per_second": 24.732,
      "step": 405
    }
  ],
  "logging_steps": 10,
  "max_steps": 405,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 366504160788480.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}