{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 9590,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22007166537097106,
      "eval_loss": 4.9203200340271,
      "eval_runtime": 6.672,
      "eval_samples_per_second": 32.824,
      "eval_steps_per_second": 1.049,
      "step": 959
    },
    {
      "epoch": 1.0427528675703859,
      "grad_norm": 14641.076171875,
      "learning_rate": 0.0005993999999999999,
      "loss": 1.251,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.26578738081834347,
      "eval_loss": 4.47660493850708,
      "eval_runtime": 5.7576,
      "eval_samples_per_second": 38.037,
      "eval_steps_per_second": 1.216,
      "step": 1918
    },
    {
      "epoch": 2.0855057351407718,
      "grad_norm": 10230.9931640625,
      "learning_rate": 0.0005302211874272409,
      "loss": 1.0398,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.27962898426400024,
      "eval_loss": 4.3035736083984375,
      "eval_runtime": 5.7523,
      "eval_samples_per_second": 38.071,
      "eval_steps_per_second": 1.217,
      "step": 2877
    },
    {
      "epoch": 3.1282586027111576,
      "grad_norm": 13416.8994140625,
      "learning_rate": 0.00046037252619324795,
      "loss": 0.978,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2879125003350937,
      "eval_loss": 4.201538562774658,
      "eval_runtime": 5.7517,
      "eval_samples_per_second": 38.076,
      "eval_steps_per_second": 1.217,
      "step": 3836
    },
    {
      "epoch": 4.1710114702815435,
      "grad_norm": 12218.20703125,
      "learning_rate": 0.00039052386495925487,
      "loss": 0.9464,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.2940960959350901,
      "eval_loss": 4.131160259246826,
      "eval_runtime": 5.7355,
      "eval_samples_per_second": 38.183,
      "eval_steps_per_second": 1.22,
      "step": 4795
    },
    {
      "epoch": 5.213764337851929,
      "grad_norm": 11805.5654296875,
      "learning_rate": 0.0003206752037252619,
      "loss": 0.9242,
      "step": 5000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3038450884200556,
      "eval_loss": 4.05396842956543,
      "eval_runtime": 5.7181,
      "eval_samples_per_second": 38.3,
      "eval_steps_per_second": 1.224,
      "step": 5754
    },
    {
      "epoch": 6.256517205422315,
      "grad_norm": 13648.1845703125,
      "learning_rate": 0.0002508265424912689,
      "loss": 0.9027,
      "step": 6000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.30757133027727884,
      "eval_loss": 3.9987871646881104,
      "eval_runtime": 5.7301,
      "eval_samples_per_second": 38.219,
      "eval_steps_per_second": 1.222,
      "step": 6713
    },
    {
      "epoch": 7.299270072992701,
      "grad_norm": 12588.6240234375,
      "learning_rate": 0.0001809778812572759,
      "loss": 0.8825,
      "step": 7000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.31707905530386293,
      "eval_loss": 3.932004690170288,
      "eval_runtime": 5.7156,
      "eval_samples_per_second": 38.316,
      "eval_steps_per_second": 1.225,
      "step": 7672
    },
    {
      "epoch": 8.342022940563087,
      "grad_norm": 10292.294921875,
      "learning_rate": 0.00011112922002328288,
      "loss": 0.8632,
      "step": 8000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.32071593884316724,
      "eval_loss": 3.891136884689331,
      "eval_runtime": 5.8599,
      "eval_samples_per_second": 37.372,
      "eval_steps_per_second": 1.195,
      "step": 8631
    },
    {
      "epoch": 9.384775808133472,
      "grad_norm": 10414.537109375,
      "learning_rate": 4.128055878928987e-05,
      "loss": 0.8454,
      "step": 9000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.32392390245646013,
      "eval_loss": 3.869386911392212,
      "eval_runtime": 5.7187,
      "eval_samples_per_second": 38.295,
      "eval_steps_per_second": 1.224,
      "step": 9590
    },
    {
      "epoch": 10.0,
      "step": 9590,
      "total_flos": 8.010952409088e+16,
      "train_loss": 0.9517836235611232,
      "train_runtime": 10856.3074,
      "train_samples_per_second": 28.241,
      "train_steps_per_second": 0.883
    }
  ],
  "logging_steps": 1000,
  "max_steps": 9590,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 8.010952409088e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}