{
  "best_metric": 0.5134699516463275,
  "best_model_checkpoint": "model-22\\checkpoint-5214",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 8690,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29,
      "grad_norm": 4.971073150634766,
      "learning_rate": 1.885155350978136e-05,
      "loss": 1.058,
      "step": 500
    },
    {
      "epoch": 0.58,
      "grad_norm": 4.795872211456299,
      "learning_rate": 1.7700805523590335e-05,
      "loss": 1.0386,
      "step": 1000
    },
    {
      "epoch": 0.86,
      "grad_norm": 3.342405319213867,
      "learning_rate": 1.655005753739931e-05,
      "loss": 1.0355,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.49988487220815103,
      "eval_loss": 1.0105576515197754,
      "eval_runtime": 28.2378,
      "eval_samples_per_second": 153.801,
      "eval_steps_per_second": 38.459,
      "step": 1738
    },
    {
      "epoch": 1.15,
      "grad_norm": 2.335313081741333,
      "learning_rate": 1.5399309551208284e-05,
      "loss": 1.0264,
      "step": 2000
    },
    {
      "epoch": 1.44,
      "grad_norm": 6.163393974304199,
      "learning_rate": 1.4250863060989643e-05,
      "loss": 1.016,
      "step": 2500
    },
    {
      "epoch": 1.73,
      "grad_norm": 3.8983561992645264,
      "learning_rate": 1.3100115074798621e-05,
      "loss": 1.0206,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4616624453142989,
      "eval_loss": 1.052566647529602,
      "eval_runtime": 28.0643,
      "eval_samples_per_second": 154.752,
      "eval_steps_per_second": 38.697,
      "step": 3476
    },
    {
      "epoch": 2.01,
      "grad_norm": 2.8223962783813477,
      "learning_rate": 1.1949367088607596e-05,
      "loss": 1.0087,
      "step": 3500
    },
    {
      "epoch": 2.3,
      "grad_norm": 2.9344329833984375,
      "learning_rate": 1.0798619102416572e-05,
      "loss": 1.0169,
      "step": 4000
    },
    {
      "epoch": 2.59,
      "grad_norm": 3.9112868309020996,
      "learning_rate": 9.65017261219793e-06,
      "loss": 1.0032,
      "step": 4500
    },
    {
      "epoch": 2.88,
      "grad_norm": 4.701815128326416,
      "learning_rate": 8.499424626006904e-06,
      "loss": 0.9986,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5134699516463275,
      "eval_loss": 1.014447808265686,
      "eval_runtime": 27.9246,
      "eval_samples_per_second": 155.526,
      "eval_steps_per_second": 38.89,
      "step": 5214
    },
    {
      "epoch": 3.16,
      "grad_norm": 5.6566972732543945,
      "learning_rate": 7.350978135788263e-06,
      "loss": 1.0088,
      "step": 5500
    },
    {
      "epoch": 3.45,
      "grad_norm": 2.4693961143493652,
      "learning_rate": 6.200230149597238e-06,
      "loss": 1.0043,
      "step": 6000
    },
    {
      "epoch": 3.74,
      "grad_norm": 6.023201942443848,
      "learning_rate": 5.0494821634062144e-06,
      "loss": 0.9943,
      "step": 6500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5010361501266406,
      "eval_loss": 1.0094102621078491,
      "eval_runtime": 27.751,
      "eval_samples_per_second": 156.499,
      "eval_steps_per_second": 39.134,
      "step": 6952
    },
    {
      "epoch": 4.03,
      "grad_norm": 3.652107000350952,
      "learning_rate": 3.898734177215191e-06,
      "loss": 0.991,
      "step": 7000
    },
    {
      "epoch": 4.32,
      "grad_norm": 12.579499244689941,
      "learning_rate": 2.747986191024166e-06,
      "loss": 0.9881,
      "step": 7500
    },
    {
      "epoch": 4.6,
      "grad_norm": 5.4077372550964355,
      "learning_rate": 1.5972382048331416e-06,
      "loss": 0.9889,
      "step": 8000
    },
    {
      "epoch": 4.89,
      "grad_norm": 7.190982341766357,
      "learning_rate": 4.4649021864211743e-07,
      "loss": 1.0025,
      "step": 8500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4978125719548699,
      "eval_loss": 1.0031945705413818,
      "eval_runtime": 28.3493,
      "eval_samples_per_second": 153.196,
      "eval_steps_per_second": 38.308,
      "step": 8690
    },
    {
      "epoch": 5.0,
      "step": 8690,
      "total_flos": 0.0,
      "train_loss": 1.011234805828134,
      "train_runtime": 967.656,
      "train_samples_per_second": 89.763,
      "train_steps_per_second": 8.98
    }
  ],
  "logging_steps": 500,
  "max_steps": 8690,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 10,
  "trial_name": null,
  "trial_params": null
}