{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 8910,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.20035457557497433,
      "eval_loss": 5.137700080871582,
      "eval_runtime": 4.7614,
      "eval_samples_per_second": 42.424,
      "eval_steps_per_second": 1.47,
      "step": 891
    },
    {
      "epoch": 1.122334455667789,
      "grad_norm": 12257.177734375,
      "learning_rate": 0.0005993999999999999,
      "loss": 1.2877,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.22360543295034005,
      "eval_loss": 4.890064239501953,
      "eval_runtime": 4.3477,
      "eval_samples_per_second": 46.461,
      "eval_steps_per_second": 1.61,
      "step": 1782
    },
    {
      "epoch": 2.244668911335578,
      "grad_norm": 16755.361328125,
      "learning_rate": 0.0005242225031605562,
      "loss": 1.1355,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2355505609269342,
      "eval_loss": 4.7771735191345215,
      "eval_runtime": 4.3663,
      "eval_samples_per_second": 46.264,
      "eval_steps_per_second": 1.603,
      "step": 2673
    },
    {
      "epoch": 3.3670033670033668,
      "grad_norm": 27167.396484375,
      "learning_rate": 0.0004483691529709229,
      "loss": 1.0964,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2399488481137742,
      "eval_loss": 4.702260971069336,
      "eval_runtime": 4.4874,
      "eval_samples_per_second": 45.014,
      "eval_steps_per_second": 1.56,
      "step": 3564
    },
    {
      "epoch": 4.489337822671156,
      "grad_norm": 20150.478515625,
      "learning_rate": 0.00037251580278128944,
      "loss": 1.0755,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.2460231346030885,
      "eval_loss": 4.61249303817749,
      "eval_runtime": 4.3579,
      "eval_samples_per_second": 46.353,
      "eval_steps_per_second": 1.606,
      "step": 4455
    },
    {
      "epoch": 5.611672278338945,
      "grad_norm": 24080.505859375,
      "learning_rate": 0.0002966624525916561,
      "loss": 1.0502,
      "step": 5000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.2551200325512003,
      "eval_loss": 4.534069538116455,
      "eval_runtime": 4.3651,
      "eval_samples_per_second": 46.276,
      "eval_steps_per_second": 1.604,
      "step": 5346
    },
    {
      "epoch": 6.7340067340067336,
      "grad_norm": 19178.21484375,
      "learning_rate": 0.00022080910240202274,
      "loss": 1.0248,
      "step": 6000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.2642556819282711,
      "eval_loss": 4.448388576507568,
      "eval_runtime": 4.4493,
      "eval_samples_per_second": 45.401,
      "eval_steps_per_second": 1.573,
      "step": 6237
    },
    {
      "epoch": 7.856341189674523,
      "grad_norm": 13593.7294921875,
      "learning_rate": 0.00014495575221238938,
      "loss": 1.0035,
      "step": 7000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.2731878863033074,
      "eval_loss": 4.369141578674316,
      "eval_runtime": 4.3376,
      "eval_samples_per_second": 46.569,
      "eval_steps_per_second": 1.614,
      "step": 7128
    },
    {
      "epoch": 8.978675645342312,
      "grad_norm": 12907.32421875,
      "learning_rate": 6.9102402022756e-05,
      "loss": 0.9814,
      "step": 8000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.27882621921683365,
      "eval_loss": 4.313453197479248,
      "eval_runtime": 4.3378,
      "eval_samples_per_second": 46.567,
      "eval_steps_per_second": 1.614,
      "step": 8019
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.28208133924938483,
      "eval_loss": 4.283874988555908,
      "eval_runtime": 4.4601,
      "eval_samples_per_second": 45.29,
      "eval_steps_per_second": 1.569,
      "step": 8910
    },
    {
      "epoch": 10.0,
      "step": 8910,
      "total_flos": 7.44290353152e+16,
      "train_loss": 1.0702610208530619,
      "train_runtime": 9181.6961,
      "train_samples_per_second": 31.024,
      "train_steps_per_second": 0.97
    }
  ],
  "logging_steps": 1000,
  "max_steps": 8910,
  "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": 7.44290353152e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}