trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 8970,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.56,
      "learning_rate": 2.8327759197324414e-05,
      "loss": 0.668,
      "step": 500
    },
    {
      "epoch": 1.11,
      "learning_rate": 2.665551839464883e-05,
      "loss": 0.2947,
      "step": 1000
    },
    {
      "epoch": 1.67,
      "learning_rate": 2.4983277591973243e-05,
      "loss": 0.2293,
      "step": 1500
    },
    {
      "epoch": 2.23,
      "learning_rate": 2.331103678929766e-05,
      "loss": 0.2054,
      "step": 2000
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.1638795986622073e-05,
      "loss": 0.1899,
      "step": 2500
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.996655518394649e-05,
      "loss": 0.1676,
      "step": 3000
    },
    {
      "epoch": 3.9,
      "learning_rate": 1.8294314381270902e-05,
      "loss": 0.1608,
      "step": 3500
    },
    {
      "epoch": 4.46,
      "learning_rate": 1.662207357859532e-05,
      "loss": 0.1571,
      "step": 4000
    },
    {
      "epoch": 5.02,
      "learning_rate": 1.4949832775919733e-05,
      "loss": 0.1482,
      "step": 4500
    },
    {
      "epoch": 5.57,
      "learning_rate": 1.3277591973244148e-05,
      "loss": 0.141,
      "step": 5000
    },
    {
      "epoch": 6.13,
      "learning_rate": 1.1605351170568563e-05,
      "loss": 0.1377,
      "step": 5500
    },
    {
      "epoch": 6.69,
      "learning_rate": 9.933110367892978e-06,
      "loss": 0.1366,
      "step": 6000
    },
    {
      "epoch": 7.25,
      "learning_rate": 8.260869565217392e-06,
      "loss": 0.1257,
      "step": 6500
    },
    {
      "epoch": 7.8,
      "learning_rate": 6.588628762541806e-06,
      "loss": 0.1293,
      "step": 7000
    },
    {
      "epoch": 8.36,
      "learning_rate": 4.916387959866221e-06,
      "loss": 0.1217,
      "step": 7500
    },
    {
      "epoch": 8.92,
      "learning_rate": 3.2441471571906356e-06,
      "loss": 0.1231,
      "step": 8000
    },
    {
      "epoch": 9.48,
      "learning_rate": 1.5719063545150504e-06,
      "loss": 0.1221,
      "step": 8500
    },
    {
      "epoch": 10.0,
      "step": 8970,
      "total_flos": 6.55421377609728e+16,
      "train_loss": 0.18795312819805166,
      "train_runtime": 5486.4328,
      "train_samples_per_second": 19.617,
      "train_steps_per_second": 1.635
    }
  ],
  "max_steps": 8970,
  "num_train_epochs": 10,
  "total_flos": 6.55421377609728e+16,
  "trial_name": null,
  "trial_params": null
}