{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 59.67567567567568,
  "eval_steps": 276,
  "global_step": 2760,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.967567567567568,
      "grad_norm": 7.085669040679932,
      "learning_rate": 9.007246376811595e-06,
      "loss": 1.1802,
      "step": 276
    },
    {
      "epoch": 5.967567567567568,
      "eval_accuracy": 0.09781729991915926,
      "eval_loss": 2.998732089996338,
      "eval_runtime": 4.6867,
      "eval_samples_per_second": 263.941,
      "eval_steps_per_second": 13.229,
      "step": 276
    },
    {
      "epoch": 11.935135135135136,
      "grad_norm": 12.396671295166016,
      "learning_rate": 8.007246376811595e-06,
      "loss": 0.8645,
      "step": 552
    },
    {
      "epoch": 11.935135135135136,
      "eval_accuracy": 0.09822150363783347,
      "eval_loss": 3.489548683166504,
      "eval_runtime": 4.7384,
      "eval_samples_per_second": 261.059,
      "eval_steps_per_second": 13.085,
      "step": 552
    },
    {
      "epoch": 17.9027027027027,
      "grad_norm": 13.397128105163574,
      "learning_rate": 7.007246376811595e-06,
      "loss": 0.6579,
      "step": 828
    },
    {
      "epoch": 17.9027027027027,
      "eval_accuracy": 0.10158986796011857,
      "eval_loss": 3.6031882762908936,
      "eval_runtime": 4.7432,
      "eval_samples_per_second": 260.795,
      "eval_steps_per_second": 13.071,
      "step": 828
    },
    {
      "epoch": 23.87027027027027,
      "grad_norm": 8.087945938110352,
      "learning_rate": 6.007246376811595e-06,
      "loss": 0.595,
      "step": 1104
    },
    {
      "epoch": 23.87027027027027,
      "eval_accuracy": 0.09822150363783347,
      "eval_loss": 3.6242969036102295,
      "eval_runtime": 4.9069,
      "eval_samples_per_second": 252.096,
      "eval_steps_per_second": 12.635,
      "step": 1104
    },
    {
      "epoch": 29.83783783783784,
      "grad_norm": 4.8504486083984375,
      "learning_rate": 5.0072463768115946e-06,
      "loss": 0.5518,
      "step": 1380
    },
    {
      "epoch": 29.83783783783784,
      "eval_accuracy": 0.09749393694421989,
      "eval_loss": 3.6076433658599854,
      "eval_runtime": 4.7024,
      "eval_samples_per_second": 263.059,
      "eval_steps_per_second": 13.185,
      "step": 1380
    },
    {
      "epoch": 35.8054054054054,
      "grad_norm": 7.603194713592529,
      "learning_rate": 4.007246376811594e-06,
      "loss": 0.5298,
      "step": 1656
    },
    {
      "epoch": 35.8054054054054,
      "eval_accuracy": 0.09714362705470224,
      "eval_loss": 3.6512610912323,
      "eval_runtime": 4.7172,
      "eval_samples_per_second": 262.229,
      "eval_steps_per_second": 13.143,
      "step": 1656
    },
    {
      "epoch": 41.77297297297297,
      "grad_norm": 5.10509729385376,
      "learning_rate": 3.0072463768115946e-06,
      "loss": 0.5155,
      "step": 1932
    },
    {
      "epoch": 41.77297297297297,
      "eval_accuracy": 0.09700889248181083,
      "eval_loss": 3.5870466232299805,
      "eval_runtime": 4.6922,
      "eval_samples_per_second": 263.627,
      "eval_steps_per_second": 13.213,
      "step": 1932
    },
    {
      "epoch": 47.74054054054054,
      "grad_norm": 2.8337128162384033,
      "learning_rate": 2.0072463768115945e-06,
      "loss": 0.5058,
      "step": 2208
    },
    {
      "epoch": 47.74054054054054,
      "eval_accuracy": 0.09660468876313662,
      "eval_loss": 3.6554582118988037,
      "eval_runtime": 4.7001,
      "eval_samples_per_second": 263.184,
      "eval_steps_per_second": 13.191,
      "step": 2208
    },
    {
      "epoch": 53.70810810810811,
      "grad_norm": 4.16039514541626,
      "learning_rate": 1.0108695652173914e-06,
      "loss": 0.4935,
      "step": 2484
    },
    {
      "epoch": 53.70810810810811,
      "eval_accuracy": 0.09602083894727387,
      "eval_loss": 3.673586130142212,
      "eval_runtime": 4.7276,
      "eval_samples_per_second": 261.655,
      "eval_steps_per_second": 13.114,
      "step": 2484
    },
    {
      "epoch": 59.67567567567568,
      "grad_norm": 2.723360061645508,
      "learning_rate": 1.0869565217391305e-08,
      "loss": 0.4928,
      "step": 2760
    },
    {
      "epoch": 59.67567567567568,
      "eval_accuracy": 0.09571544058205335,
      "eval_loss": 3.731440544128418,
      "eval_runtime": 4.6934,
      "eval_samples_per_second": 263.563,
      "eval_steps_per_second": 13.21,
      "step": 2760
    },
    {
      "epoch": 59.67567567567568,
      "step": 2760,
      "total_flos": 2.1787658809800192e+17,
      "train_loss": 0.6386864206065302,
      "train_runtime": 4685.4185,
      "train_samples_per_second": 118.222,
      "train_steps_per_second": 0.589
    }
  ],
  "logging_steps": 276,
  "max_steps": 2760,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 60,
  "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": 2.1787658809800192e+17,
  "train_batch_size": 25,
  "trial_name": null,
  "trial_params": null
}