{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 40,
  "global_step": 384,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3125,
      "grad_norm": 2.4588866233825684,
      "learning_rate": 6.578947368421053e-06,
      "loss": 0.3311,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 40,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.3125,
      "eval_accuracy": 0.9449360865290068,
      "eval_auroc": 0.9216382070437567,
      "eval_loss": 0.16689223051071167,
      "eval_runtime": 4.7738,
      "eval_samples_per_second": 213.038,
      "eval_steps_per_second": 6.703,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 40,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.625,
      "grad_norm": 6.371248245239258,
      "learning_rate": 1.3157894736842106e-05,
      "loss": 0.1336,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 80,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.625,
      "eval_accuracy": 0.951819075712881,
      "eval_auroc": 0.9588580576307363,
      "eval_loss": 0.14572063088417053,
      "eval_runtime": 1.6415,
      "eval_samples_per_second": 619.571,
      "eval_steps_per_second": 19.495,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 80,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.8174734115600586,
      "learning_rate": 1.9736842105263158e-05,
      "loss": 0.092,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.9375,
      "eval_accuracy": 0.9587020648967551,
      "eval_auroc": 0.9663153681963713,
      "eval_loss": 0.13402177393436432,
      "eval_runtime": 2.0205,
      "eval_samples_per_second": 503.333,
      "eval_steps_per_second": 15.837,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "grad_norm": 10.359611511230469,
      "learning_rate": 2.6315789473684212e-05,
      "loss": 0.0683,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.25,
      "eval_accuracy": 0.9616519174041298,
      "eval_auroc": 0.9728054962646745,
      "eval_loss": 0.15004919469356537,
      "eval_runtime": 1.9839,
      "eval_samples_per_second": 512.639,
      "eval_steps_per_second": 16.13,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 160,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5625,
      "grad_norm": 0.29452481865882874,
      "learning_rate": 3.289473684210527e-05,
      "loss": 0.0778,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.5625,
      "eval_accuracy": 0.9567354965585054,
      "eval_auroc": 0.9560165421558164,
      "eval_loss": 0.19474196434020996,
      "eval_runtime": 1.8404,
      "eval_samples_per_second": 552.609,
      "eval_steps_per_second": 17.388,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 200,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.4737164378166199,
      "learning_rate": 3.9473684210526316e-05,
      "loss": 0.1011,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 1.875,
      "eval_accuracy": 0.967551622418879,
      "eval_auroc": 0.9674493062966916,
      "eval_loss": 0.13491444289684296,
      "eval_runtime": 1.7786,
      "eval_samples_per_second": 571.791,
      "eval_steps_per_second": 17.991,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 240,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.1875,
      "grad_norm": 0.6345912218093872,
      "learning_rate": 4.605263157894737e-05,
      "loss": 0.06,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.1875,
      "eval_accuracy": 0.9655850540806293,
      "eval_auroc": 0.9698305763073639,
      "eval_loss": 0.1612750142812729,
      "eval_runtime": 1.893,
      "eval_samples_per_second": 537.248,
      "eval_steps_per_second": 16.905,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 280,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.5,
      "grad_norm": 13.752758026123047,
      "learning_rate": 4e-05,
      "loss": 0.0508,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.5,
      "eval_accuracy": 0.9685349065880039,
      "eval_auroc": 0.9268209711846318,
      "eval_loss": 0.19579041004180908,
      "eval_runtime": 1.7675,
      "eval_samples_per_second": 575.401,
      "eval_steps_per_second": 18.105,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 320,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.8125,
      "grad_norm": 15.928082466125488,
      "learning_rate": 1.5e-05,
      "loss": 0.0722,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 360,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 2.8125,
      "eval_accuracy": 0.967551622418879,
      "eval_auroc": 0.9649679829242261,
      "eval_loss": 0.15064473450183868,
      "eval_runtime": 1.906,
      "eval_samples_per_second": 533.581,
      "eval_steps_per_second": 16.789,
      "max_memory_allocated (GB)": 13.36,
      "memory_allocated (GB)": 11.92,
      "step": 360,
      "total_memory_available (GB)": 94.62
    }
  ],
  "logging_steps": 40,
  "max_steps": 384,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3233108648263680.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}