{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 96980,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 4.406856536865234,
      "learning_rate": 4.750206228088266e-05,
      "loss": 3.599,
      "step": 4849
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.830796480178833,
      "eval_runtime": 63.7615,
      "eval_samples_per_second": 608.329,
      "eval_steps_per_second": 19.024,
      "step": 4849
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.6979801654815674,
      "learning_rate": 4.5002577851103326e-05,
      "loss": 2.6769,
      "step": 9698
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.430718183517456,
      "eval_runtime": 62.9007,
      "eval_samples_per_second": 616.654,
      "eval_steps_per_second": 19.284,
      "step": 9698
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.8321845531463623,
      "learning_rate": 4.2503608991544655e-05,
      "loss": 2.3727,
      "step": 14547
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.206211566925049,
      "eval_runtime": 62.7764,
      "eval_samples_per_second": 617.876,
      "eval_steps_per_second": 19.323,
      "step": 14547
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.3764877319335938,
      "learning_rate": 4.000464013198598e-05,
      "loss": 2.1798,
      "step": 19396
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.0566577911376953,
      "eval_runtime": 62.7,
      "eval_samples_per_second": 618.629,
      "eval_steps_per_second": 19.346,
      "step": 19396
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.017418146133423,
      "learning_rate": 3.750567127242731e-05,
      "loss": 2.041,
      "step": 24245
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.943306803703308,
      "eval_runtime": 62.7554,
      "eval_samples_per_second": 618.082,
      "eval_steps_per_second": 19.329,
      "step": 24245
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.028724431991577,
      "learning_rate": 3.500567127242731e-05,
      "loss": 1.9417,
      "step": 29094
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.8640257120132446,
      "eval_runtime": 63.2843,
      "eval_samples_per_second": 612.916,
      "eval_steps_per_second": 19.167,
      "step": 29094
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.9556593894958496,
      "learning_rate": 3.2506702412868636e-05,
      "loss": 1.8625,
      "step": 33943
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.7964398860931396,
      "eval_runtime": 63.0672,
      "eval_samples_per_second": 615.027,
      "eval_steps_per_second": 19.233,
      "step": 33943
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.954166889190674,
      "learning_rate": 3.000670241286863e-05,
      "loss": 1.797,
      "step": 38792
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.7490918636322021,
      "eval_runtime": 63.1115,
      "eval_samples_per_second": 614.595,
      "eval_steps_per_second": 19.22,
      "step": 38792
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.954655885696411,
      "learning_rate": 2.7508249123530627e-05,
      "loss": 1.7491,
      "step": 43641
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.7073553800582886,
      "eval_runtime": 63.2567,
      "eval_samples_per_second": 613.184,
      "eval_steps_per_second": 19.176,
      "step": 43641
    },
    {
      "epoch": 10.0,
      "grad_norm": 3.0339479446411133,
      "learning_rate": 2.5009280263971953e-05,
      "loss": 1.7049,
      "step": 48490
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.6738481521606445,
      "eval_runtime": 63.2477,
      "eval_samples_per_second": 613.271,
      "eval_steps_per_second": 19.179,
      "step": 48490
    },
    {
      "epoch": 11.0,
      "grad_norm": 3.4147965908050537,
      "learning_rate": 2.2509795834192618e-05,
      "loss": 1.6655,
      "step": 53339
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.6354576349258423,
      "eval_runtime": 63.6046,
      "eval_samples_per_second": 609.83,
      "eval_steps_per_second": 19.071,
      "step": 53339
    },
    {
      "epoch": 12.0,
      "grad_norm": 3.4822511672973633,
      "learning_rate": 2.0010826974633944e-05,
      "loss": 1.6295,
      "step": 58188
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.6142407655715942,
      "eval_runtime": 63.1801,
      "eval_samples_per_second": 613.927,
      "eval_steps_per_second": 19.199,
      "step": 58188
    },
    {
      "epoch": 13.0,
      "grad_norm": 3.052323341369629,
      "learning_rate": 1.751134254485461e-05,
      "loss": 1.6022,
      "step": 63037
    },
    {
      "epoch": 13.0,
      "eval_loss": 1.595221757888794,
      "eval_runtime": 63.3841,
      "eval_samples_per_second": 611.951,
      "eval_steps_per_second": 19.137,
      "step": 63037
    },
    {
      "epoch": 14.0,
      "grad_norm": 3.0843849182128906,
      "learning_rate": 1.5012373685295938e-05,
      "loss": 1.5769,
      "step": 67886
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.5756992101669312,
      "eval_runtime": 61.4761,
      "eval_samples_per_second": 630.944,
      "eval_steps_per_second": 19.731,
      "step": 67886
    },
    {
      "epoch": 15.0,
      "grad_norm": 3.4379537105560303,
      "learning_rate": 1.2513404825737265e-05,
      "loss": 1.5571,
      "step": 72735
    },
    {
      "epoch": 15.0,
      "eval_loss": 1.554135799407959,
      "eval_runtime": 62.2027,
      "eval_samples_per_second": 623.574,
      "eval_steps_per_second": 19.501,
      "step": 72735
    },
    {
      "epoch": 16.0,
      "grad_norm": 3.0627856254577637,
      "learning_rate": 1.0014435966178593e-05,
      "loss": 1.5343,
      "step": 77584
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.541438102722168,
      "eval_runtime": 61.9741,
      "eval_samples_per_second": 625.874,
      "eval_steps_per_second": 19.573,
      "step": 77584
    },
    {
      "epoch": 17.0,
      "grad_norm": 3.1744439601898193,
      "learning_rate": 7.5154671066199224e-06,
      "loss": 1.5188,
      "step": 82433
    },
    {
      "epoch": 17.0,
      "eval_loss": 1.5268478393554688,
      "eval_runtime": 61.9172,
      "eval_samples_per_second": 626.449,
      "eval_steps_per_second": 19.591,
      "step": 82433
    },
    {
      "epoch": 18.0,
      "grad_norm": 3.5251266956329346,
      "learning_rate": 5.015982676840586e-06,
      "loss": 1.5037,
      "step": 87282
    },
    {
      "epoch": 18.0,
      "eval_loss": 1.5217667818069458,
      "eval_runtime": 61.3504,
      "eval_samples_per_second": 632.237,
      "eval_steps_per_second": 19.772,
      "step": 87282
    },
    {
      "epoch": 19.0,
      "grad_norm": 3.248511552810669,
      "learning_rate": 2.517013817281914e-06,
      "loss": 1.4927,
      "step": 92131
    },
    {
      "epoch": 19.0,
      "eval_loss": 1.511409044265747,
      "eval_runtime": 62.5116,
      "eval_samples_per_second": 620.493,
      "eval_steps_per_second": 19.404,
      "step": 92131
    },
    {
      "epoch": 20.0,
      "grad_norm": 2.856531858444214,
      "learning_rate": 1.7529387502577853e-08,
      "loss": 1.4862,
      "step": 96980
    },
    {
      "epoch": 20.0,
      "eval_loss": 1.5111480951309204,
      "eval_runtime": 61.8388,
      "eval_samples_per_second": 627.244,
      "eval_steps_per_second": 19.616,
      "step": 96980
    }
  ],
  "logging_steps": 500,
  "max_steps": 96980,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 2.0420834325692416e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}