checkpoint-1000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.999466666666667e-05,
      "loss": 2.308,
      "step": 1
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.8997333333333335e-05,
      "loss": 1.6916,
      "step": 188
    },
    {
      "epoch": 0.75,
      "eval_accuracy": 0.6755,
      "eval_disk_space_total": 78.1898422241211,
      "eval_disk_space_used": 24.806385040283203,
      "eval_f1": 0.6708054417489328,
      "eval_gpu_ram_allocated": 2.089780330657959,
      "eval_gpu_ram_cached": 25.85546875,
      "eval_gpu_ram_total": 39.56402587890625,
      "eval_gpu_utilization": 50,
      "eval_loss": 1.1062816381454468,
      "eval_precision": 0.690043017889279,
      "eval_recall": 0.6755,
      "eval_runtime": 2.3912,
      "eval_samples_per_second": 836.398,
      "eval_steps_per_second": 26.347,
      "eval_system_ram_total": 83.48074722290039,
      "eval_system_ram_used": 4.019077301025391,
      "step": 188
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.8e-05,
      "loss": 0.9694,
      "step": 376
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.7195,
      "eval_disk_space_total": 78.1898422241211,
      "eval_disk_space_used": 29.64177703857422,
      "eval_f1": 0.7181081417115642,
      "eval_gpu_ram_allocated": 2.0897774696350098,
      "eval_gpu_ram_cached": 25.85546875,
      "eval_gpu_ram_total": 39.56402587890625,
      "eval_gpu_utilization": 50,
      "eval_loss": 0.9585903286933899,
      "eval_precision": 0.719758443061289,
      "eval_recall": 0.7195,
      "eval_runtime": 2.3693,
      "eval_samples_per_second": 844.141,
      "eval_steps_per_second": 26.59,
      "eval_system_ram_total": 83.48074722290039,
      "eval_system_ram_used": 4.2536163330078125,
      "step": 376
    },
    {
      "epoch": 2.26,
      "learning_rate": 1.6997333333333334e-05,
      "loss": 0.8509,
      "step": 564
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.712,
      "eval_disk_space_total": 78.1898422241211,
      "eval_disk_space_used": 29.641841888427734,
      "eval_f1": 0.7070168337920522,
      "eval_gpu_ram_allocated": 2.089791774749756,
      "eval_gpu_ram_cached": 25.85546875,
      "eval_gpu_ram_total": 39.56402587890625,
      "eval_gpu_utilization": 46,
      "eval_loss": 0.9747923016548157,
      "eval_precision": 0.7160570316458433,
      "eval_recall": 0.712,
      "eval_runtime": 2.4432,
      "eval_samples_per_second": 818.586,
      "eval_steps_per_second": 25.785,
      "eval_system_ram_total": 83.48074722290039,
      "eval_system_ram_used": 4.160213470458984,
      "step": 564
    },
    {
      "epoch": 3.01,
      "learning_rate": 1.5994666666666668e-05,
      "loss": 0.7475,
      "step": 752
    },
    {
      "epoch": 3.01,
      "eval_accuracy": 0.714,
      "eval_disk_space_total": 78.1898422241211,
      "eval_disk_space_used": 29.641963958740234,
      "eval_f1": 0.7122032912823338,
      "eval_gpu_ram_allocated": 2.089776039123535,
      "eval_gpu_ram_cached": 25.85546875,
      "eval_gpu_ram_total": 39.56402587890625,
      "eval_gpu_utilization": 50,
      "eval_loss": 0.9446640014648438,
      "eval_precision": 0.7148157467744413,
      "eval_recall": 0.714,
      "eval_runtime": 2.5063,
      "eval_samples_per_second": 798.001,
      "eval_steps_per_second": 25.137,
      "eval_system_ram_total": 83.48074722290039,
      "eval_system_ram_used": 4.160709381103516,
      "step": 752
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.4997333333333335e-05,
      "loss": 0.5841,
      "step": 940
    },
    {
      "epoch": 3.76,
      "eval_accuracy": 0.711,
      "eval_disk_space_total": 78.1898422241211,
      "eval_disk_space_used": 29.642024993896484,
      "eval_f1": 0.7076606604060025,
      "eval_gpu_ram_allocated": 2.089787483215332,
      "eval_gpu_ram_cached": 25.85546875,
      "eval_gpu_ram_total": 39.56402587890625,
      "eval_gpu_utilization": 47,
      "eval_loss": 1.0064291954040527,
      "eval_precision": 0.7225290812411572,
      "eval_recall": 0.711,
      "eval_runtime": 2.4755,
      "eval_samples_per_second": 807.933,
      "eval_steps_per_second": 25.45,
      "eval_system_ram_total": 83.48074722290039,
      "eval_system_ram_used": 4.188880920410156,
      "step": 940
    }
  ],
  "max_steps": 3750,
  "num_train_epochs": 15,
  "total_flos": 1921836975330048.0,
  "trial_name": null,
  "trial_params": null
}