{
  "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": 1.919080138206482,
      "learning_rate": 9.007246376811595e-06,
      "loss": 1.2383,
      "step": 276
    },
    {
      "epoch": 5.967567567567568,
      "eval_accuracy": 0.24171382376717865,
      "eval_loss": 2.6388285160064697,
      "eval_runtime": 3.7642,
      "eval_samples_per_second": 328.62,
      "eval_steps_per_second": 16.471,
      "step": 276
    },
    {
      "epoch": 11.935135135135136,
      "grad_norm": 4.005405426025391,
      "learning_rate": 8.007246376811595e-06,
      "loss": 0.9769,
      "step": 552
    },
    {
      "epoch": 11.935135135135136,
      "eval_accuracy": 0.22473726758286175,
      "eval_loss": 2.817387580871582,
      "eval_runtime": 3.6842,
      "eval_samples_per_second": 335.761,
      "eval_steps_per_second": 16.829,
      "step": 552
    },
    {
      "epoch": 17.9027027027027,
      "grad_norm": 5.196224689483643,
      "learning_rate": 7.007246376811595e-06,
      "loss": 0.8157,
      "step": 828
    },
    {
      "epoch": 17.9027027027027,
      "eval_accuracy": 0.21476690918889788,
      "eval_loss": 3.148566961288452,
      "eval_runtime": 3.7399,
      "eval_samples_per_second": 330.758,
      "eval_steps_per_second": 16.578,
      "step": 828
    },
    {
      "epoch": 23.87027027027027,
      "grad_norm": 5.394821643829346,
      "learning_rate": 6.007246376811595e-06,
      "loss": 0.7322,
      "step": 1104
    },
    {
      "epoch": 23.87027027027027,
      "eval_accuracy": 0.20796281325788196,
      "eval_loss": 3.3019931316375732,
      "eval_runtime": 3.7581,
      "eval_samples_per_second": 329.155,
      "eval_steps_per_second": 16.498,
      "step": 1104
    },
    {
      "epoch": 29.83783783783784,
      "grad_norm": 4.061520576477051,
      "learning_rate": 5.0072463768115946e-06,
      "loss": 0.6777,
      "step": 1380
    },
    {
      "epoch": 29.83783783783784,
      "eval_accuracy": 0.20258690379951497,
      "eval_loss": 3.393256187438965,
      "eval_runtime": 3.7268,
      "eval_samples_per_second": 331.922,
      "eval_steps_per_second": 16.636,
      "step": 1380
    },
    {
      "epoch": 35.8054054054054,
      "grad_norm": 4.988523483276367,
      "learning_rate": 4.007246376811594e-06,
      "loss": 0.6466,
      "step": 1656
    },
    {
      "epoch": 35.8054054054054,
      "eval_accuracy": 0.19967663702506064,
      "eval_loss": 3.418116807937622,
      "eval_runtime": 3.7118,
      "eval_samples_per_second": 333.261,
      "eval_steps_per_second": 16.703,
      "step": 1656
    },
    {
      "epoch": 41.77297297297297,
      "grad_norm": 4.484840393066406,
      "learning_rate": 3.0072463768115946e-06,
      "loss": 0.6272,
      "step": 1932
    },
    {
      "epoch": 41.77297297297297,
      "eval_accuracy": 0.1974823882665435,
      "eval_loss": 3.436495304107666,
      "eval_runtime": 3.7964,
      "eval_samples_per_second": 325.831,
      "eval_steps_per_second": 16.331,
      "step": 1932
    },
    {
      "epoch": 47.74054054054054,
      "grad_norm": 3.4457993507385254,
      "learning_rate": 2.0108695652173916e-06,
      "loss": 0.6188,
      "step": 2208
    },
    {
      "epoch": 47.74054054054054,
      "eval_accuracy": 0.19563459983831852,
      "eval_loss": 3.466963529586792,
      "eval_runtime": 3.7849,
      "eval_samples_per_second": 326.822,
      "eval_steps_per_second": 16.381,
      "step": 2208
    },
    {
      "epoch": 53.70810810810811,
      "grad_norm": 5.685315132141113,
      "learning_rate": 1.0108695652173914e-06,
      "loss": 0.6052,
      "step": 2484
    },
    {
      "epoch": 53.70810810810811,
      "eval_accuracy": 0.19455672325518727,
      "eval_loss": 3.4625163078308105,
      "eval_runtime": 3.7562,
      "eval_samples_per_second": 329.32,
      "eval_steps_per_second": 16.506,
      "step": 2484
    },
    {
      "epoch": 59.67567567567568,
      "grad_norm": 3.5206406116485596,
      "learning_rate": 1.0869565217391305e-08,
      "loss": 0.6049,
      "step": 2760
    },
    {
      "epoch": 59.67567567567568,
      "eval_accuracy": 0.1936944219886823,
      "eval_loss": 3.4762697219848633,
      "eval_runtime": 3.721,
      "eval_samples_per_second": 332.435,
      "eval_steps_per_second": 16.662,
      "step": 2760
    },
    {
      "epoch": 59.67567567567568,
      "step": 2760,
      "total_flos": 1.4462122119191654e+17,
      "train_loss": 0.7543455040973166,
      "train_runtime": 3693.2507,
      "train_samples_per_second": 149.982,
      "train_steps_per_second": 0.747
    }
  ],
  "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": 1.4462122119191654e+17,
  "train_batch_size": 25,
  "trial_name": null,
  "trial_params": null
}