{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 12.991322457480042,
  "eval_steps": 500,
  "global_step": 18720,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3471017007983339,
      "grad_norm": 9.175567626953125,
      "learning_rate": 2e-05,
      "loss": 10.9573,
      "step": 500
    },
    {
      "epoch": 0.6942034015966678,
      "grad_norm": 6.246857166290283,
      "learning_rate": 1.9451152579582877e-05,
      "loss": 2.2463,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.4781490564346313,
      "eval_runtime": 78.7792,
      "eval_samples_per_second": 146.282,
      "eval_steps_per_second": 4.582,
      "step": 1441
    },
    {
      "epoch": 1.0409580006942034,
      "grad_norm": 0.7370556592941284,
      "learning_rate": 1.890230515916575e-05,
      "loss": 1.8168,
      "step": 1500
    },
    {
      "epoch": 1.3880597014925373,
      "grad_norm": 600.9544677734375,
      "learning_rate": 1.835345773874863e-05,
      "loss": 2.8979,
      "step": 2000
    },
    {
      "epoch": 1.7351614022908712,
      "grad_norm": 0.528658390045166,
      "learning_rate": 1.7804610318331505e-05,
      "loss": 1.701,
      "step": 2500
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.3760780096054077,
      "eval_runtime": 78.8326,
      "eval_samples_per_second": 146.183,
      "eval_steps_per_second": 4.579,
      "step": 2882
    },
    {
      "epoch": 2.081916001388407,
      "grad_norm": 0.45837515592575073,
      "learning_rate": 1.725576289791438e-05,
      "loss": 1.5956,
      "step": 3000
    },
    {
      "epoch": 2.4290177021867407,
      "grad_norm": 0.41563984751701355,
      "learning_rate": 1.6706915477497255e-05,
      "loss": 1.5615,
      "step": 3500
    },
    {
      "epoch": 2.7761194029850746,
      "grad_norm": 0.4553076922893524,
      "learning_rate": 1.6158068057080134e-05,
      "loss": 1.5261,
      "step": 4000
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.3131805658340454,
      "eval_runtime": 78.7183,
      "eval_samples_per_second": 146.396,
      "eval_steps_per_second": 4.586,
      "step": 4323
    },
    {
      "epoch": 3.12287400208261,
      "grad_norm": 0.4429566562175751,
      "learning_rate": 1.560922063666301e-05,
      "loss": 1.4933,
      "step": 4500
    },
    {
      "epoch": 3.469975702880944,
      "grad_norm": 0.4545910656452179,
      "learning_rate": 1.5060373216245885e-05,
      "loss": 1.4766,
      "step": 5000
    },
    {
      "epoch": 3.817077403679278,
      "grad_norm": 0.4629240930080414,
      "learning_rate": 1.451152579582876e-05,
      "loss": 1.4626,
      "step": 5500
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.2775077819824219,
      "eval_runtime": 78.6454,
      "eval_samples_per_second": 146.531,
      "eval_steps_per_second": 4.59,
      "step": 5764
    },
    {
      "epoch": 4.163832002776814,
      "grad_norm": 0.3974866271018982,
      "learning_rate": 1.3962678375411635e-05,
      "loss": 1.448,
      "step": 6000
    },
    {
      "epoch": 4.510933703575148,
      "grad_norm": 0.4241460859775543,
      "learning_rate": 1.3413830954994514e-05,
      "loss": 1.4241,
      "step": 6500
    },
    {
      "epoch": 4.8580354043734815,
      "grad_norm": 0.4666702151298523,
      "learning_rate": 1.2864983534577389e-05,
      "loss": 1.4085,
      "step": 7000
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.254936695098877,
      "eval_runtime": 78.5765,
      "eval_samples_per_second": 146.66,
      "eval_steps_per_second": 4.594,
      "step": 7205
    },
    {
      "epoch": 5.204790003471017,
      "grad_norm": 0.4225795567035675,
      "learning_rate": 1.2316136114160264e-05,
      "loss": 1.395,
      "step": 7500
    },
    {
      "epoch": 5.551891704269351,
      "grad_norm": 0.42448151111602783,
      "learning_rate": 1.176728869374314e-05,
      "loss": 1.3938,
      "step": 8000
    },
    {
      "epoch": 5.898993405067685,
      "grad_norm": 0.47224199771881104,
      "learning_rate": 1.1218441273326017e-05,
      "loss": 1.3844,
      "step": 8500
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.2364274263381958,
      "eval_runtime": 78.7665,
      "eval_samples_per_second": 146.306,
      "eval_steps_per_second": 4.583,
      "step": 8646
    },
    {
      "epoch": 6.24574800416522,
      "grad_norm": 0.3676798939704895,
      "learning_rate": 1.0669593852908892e-05,
      "loss": 1.3668,
      "step": 9000
    },
    {
      "epoch": 6.592849704963554,
      "grad_norm": 0.37000757455825806,
      "learning_rate": 1.0120746432491769e-05,
      "loss": 1.3629,
      "step": 9500
    },
    {
      "epoch": 6.939951405761888,
      "grad_norm": 0.45095133781433105,
      "learning_rate": 9.571899012074644e-06,
      "loss": 1.3647,
      "step": 10000
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.222563624382019,
      "eval_runtime": 78.5655,
      "eval_samples_per_second": 146.68,
      "eval_steps_per_second": 4.595,
      "step": 10087
    },
    {
      "epoch": 7.286706004859424,
      "grad_norm": 0.4932011365890503,
      "learning_rate": 9.02305159165752e-06,
      "loss": 1.358,
      "step": 10500
    },
    {
      "epoch": 7.633807705657758,
      "grad_norm": 0.36681583523750305,
      "learning_rate": 8.474204171240396e-06,
      "loss": 1.3422,
      "step": 11000
    },
    {
      "epoch": 7.9809094064560915,
      "grad_norm": 0.39713332056999207,
      "learning_rate": 7.925356750823272e-06,
      "loss": 1.3442,
      "step": 11500
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.214061975479126,
      "eval_runtime": 78.596,
      "eval_samples_per_second": 146.623,
      "eval_steps_per_second": 4.593,
      "step": 11528
    },
    {
      "epoch": 8.327664005553627,
      "grad_norm": 0.4920157194137573,
      "learning_rate": 7.376509330406147e-06,
      "loss": 1.3358,
      "step": 12000
    },
    {
      "epoch": 8.67476570635196,
      "grad_norm": 0.4292038083076477,
      "learning_rate": 6.827661909989024e-06,
      "loss": 1.3373,
      "step": 12500
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.2043074369430542,
      "eval_runtime": 78.6255,
      "eval_samples_per_second": 146.568,
      "eval_steps_per_second": 4.591,
      "step": 12969
    },
    {
      "epoch": 9.021520305449497,
      "grad_norm": 0.390511691570282,
      "learning_rate": 6.278814489571899e-06,
      "loss": 1.3305,
      "step": 13000
    },
    {
      "epoch": 9.368622006247831,
      "grad_norm": 0.42470699548721313,
      "learning_rate": 5.729967069154776e-06,
      "loss": 1.3294,
      "step": 13500
    },
    {
      "epoch": 9.715723707046164,
      "grad_norm": 0.4358290135860443,
      "learning_rate": 5.181119648737652e-06,
      "loss": 1.3182,
      "step": 14000
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.1995124816894531,
      "eval_runtime": 78.7492,
      "eval_samples_per_second": 146.338,
      "eval_steps_per_second": 4.584,
      "step": 14410
    },
    {
      "epoch": 10.0624783061437,
      "grad_norm": 0.37242305278778076,
      "learning_rate": 4.632272228320528e-06,
      "loss": 1.3192,
      "step": 14500
    },
    {
      "epoch": 10.409580006942035,
      "grad_norm": 0.4692925810813904,
      "learning_rate": 4.0834248079034035e-06,
      "loss": 1.3177,
      "step": 15000
    },
    {
      "epoch": 10.756681707740368,
      "grad_norm": 0.42067646980285645,
      "learning_rate": 3.5345773874862793e-06,
      "loss": 1.3169,
      "step": 15500
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.1953924894332886,
      "eval_runtime": 78.6673,
      "eval_samples_per_second": 146.49,
      "eval_steps_per_second": 4.589,
      "step": 15851
    },
    {
      "epoch": 11.103436306837903,
      "grad_norm": 0.4010460674762726,
      "learning_rate": 2.9857299670691548e-06,
      "loss": 1.3145,
      "step": 16000
    },
    {
      "epoch": 11.450538007636238,
      "grad_norm": 0.4176557660102844,
      "learning_rate": 2.436882546652031e-06,
      "loss": 1.3063,
      "step": 16500
    },
    {
      "epoch": 11.797639708434572,
      "grad_norm": 0.3621806800365448,
      "learning_rate": 1.888035126234907e-06,
      "loss": 1.3195,
      "step": 17000
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.1941872835159302,
      "eval_runtime": 78.6747,
      "eval_samples_per_second": 146.477,
      "eval_steps_per_second": 4.589,
      "step": 17292
    },
    {
      "epoch": 12.144394307532107,
      "grad_norm": 0.41595444083213806,
      "learning_rate": 1.3391877058177828e-06,
      "loss": 1.3222,
      "step": 17500
    },
    {
      "epoch": 12.49149600833044,
      "grad_norm": 0.40309804677963257,
      "learning_rate": 7.903402854006587e-07,
      "loss": 1.3122,
      "step": 18000
    },
    {
      "epoch": 12.838597709128775,
      "grad_norm": 0.4489055871963501,
      "learning_rate": 2.414928649835346e-07,
      "loss": 1.3016,
      "step": 18500
    },
    {
      "epoch": 12.991322457480042,
      "eval_loss": 1.19288170337677,
      "eval_runtime": 78.6526,
      "eval_samples_per_second": 146.518,
      "eval_steps_per_second": 4.59,
      "step": 18720
    }
  ],
  "logging_steps": 500,
  "max_steps": 18720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 13,
  "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": 4.070590055763149e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}