{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 59.922928709055874,
  "eval_steps": 1866,
  "global_step": 18660,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.992292870905588,
      "grad_norm": 2.7764830589294434,
      "learning_rate": 9.001607717041802e-06,
      "loss": 2.4298,
      "step": 1866
    },
    {
      "epoch": 5.992292870905588,
      "eval_accuracy": 0.043435941112057425,
      "eval_loss": 3.7282466888427734,
      "eval_runtime": 31.9019,
      "eval_samples_per_second": 257.634,
      "eval_steps_per_second": 6.457,
      "step": 1866
    },
    {
      "epoch": 11.984585741811175,
      "grad_norm": 4.700397491455078,
      "learning_rate": 8.001607717041802e-06,
      "loss": 2.246,
      "step": 3732
    },
    {
      "epoch": 11.984585741811175,
      "eval_accuracy": 0.037899987833069713,
      "eval_loss": 3.9666006565093994,
      "eval_runtime": 31.9111,
      "eval_samples_per_second": 257.559,
      "eval_steps_per_second": 6.455,
      "step": 3732
    },
    {
      "epoch": 17.976878612716764,
      "grad_norm": 11.175146102905273,
      "learning_rate": 7.002143622722402e-06,
      "loss": 2.1293,
      "step": 5598
    },
    {
      "epoch": 17.976878612716764,
      "eval_accuracy": 0.03601411363912885,
      "eval_loss": 4.101926326751709,
      "eval_runtime": 31.9673,
      "eval_samples_per_second": 257.106,
      "eval_steps_per_second": 6.444,
      "step": 5598
    },
    {
      "epoch": 23.96917148362235,
      "grad_norm": 4.427066802978516,
      "learning_rate": 6.002143622722402e-06,
      "loss": 2.0539,
      "step": 7464
    },
    {
      "epoch": 23.96917148362235,
      "eval_accuracy": 0.03516242851928458,
      "eval_loss": 4.282138824462891,
      "eval_runtime": 31.8094,
      "eval_samples_per_second": 258.383,
      "eval_steps_per_second": 6.476,
      "step": 7464
    },
    {
      "epoch": 29.961464354527937,
      "grad_norm": 9.087589263916016,
      "learning_rate": 5.002679528403002e-06,
      "loss": 2.0135,
      "step": 9330
    },
    {
      "epoch": 29.961464354527937,
      "eval_accuracy": 0.03452974814454313,
      "eval_loss": 4.331849575042725,
      "eval_runtime": 31.6904,
      "eval_samples_per_second": 259.353,
      "eval_steps_per_second": 6.5,
      "step": 9330
    },
    {
      "epoch": 35.95375722543353,
      "grad_norm": 5.957192420959473,
      "learning_rate": 4.003215434083601e-06,
      "loss": 1.9879,
      "step": 11196
    },
    {
      "epoch": 35.95375722543353,
      "eval_accuracy": 0.03414851766232713,
      "eval_loss": 4.370044708251953,
      "eval_runtime": 31.7505,
      "eval_samples_per_second": 258.862,
      "eval_steps_per_second": 6.488,
      "step": 11196
    },
    {
      "epoch": 41.946050096339114,
      "grad_norm": 3.3914530277252197,
      "learning_rate": 3.003215434083602e-06,
      "loss": 1.9648,
      "step": 13062
    },
    {
      "epoch": 41.946050096339114,
      "eval_accuracy": 0.03413693010967619,
      "eval_loss": 4.461861610412598,
      "eval_runtime": 31.7169,
      "eval_samples_per_second": 259.137,
      "eval_steps_per_second": 6.495,
      "step": 13062
    },
    {
      "epoch": 47.9383429672447,
      "grad_norm": 5.636336326599121,
      "learning_rate": 2.0037513397642016e-06,
      "loss": 1.9495,
      "step": 14928
    },
    {
      "epoch": 47.9383429672447,
      "eval_accuracy": 0.03411303078233362,
      "eval_loss": 4.5998759269714355,
      "eval_runtime": 31.8541,
      "eval_samples_per_second": 258.02,
      "eval_steps_per_second": 6.467,
      "step": 14928
    },
    {
      "epoch": 53.93063583815029,
      "grad_norm": 1.326487421989441,
      "learning_rate": 1.0042872454448018e-06,
      "loss": 1.9391,
      "step": 16794
    },
    {
      "epoch": 53.93063583815029,
      "eval_accuracy": 0.03390517905665734,
      "eval_loss": 4.680637359619141,
      "eval_runtime": 32.1441,
      "eval_samples_per_second": 255.692,
      "eval_steps_per_second": 6.409,
      "step": 16794
    },
    {
      "epoch": 59.922928709055874,
      "grad_norm": 2.050564765930176,
      "learning_rate": 4.82315112540193e-09,
      "loss": 1.9372,
      "step": 18660
    },
    {
      "epoch": 59.922928709055874,
      "eval_accuracy": 0.03361722837328142,
      "eval_loss": 4.748708724975586,
      "eval_runtime": 31.9078,
      "eval_samples_per_second": 257.586,
      "eval_steps_per_second": 6.456,
      "step": 18660
    },
    {
      "epoch": 59.922928709055874,
      "step": 18660,
      "total_flos": 1.4737892295759744e+18,
      "train_loss": 2.065115170003517,
      "train_runtime": 33403.3218,
      "train_samples_per_second": 111.824,
      "train_steps_per_second": 0.559
    }
  ],
  "logging_steps": 1866,
  "max_steps": 18660,
  "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": 1.4737892295759744e+18,
  "train_batch_size": 40,
  "trial_name": null,
  "trial_params": null
}