trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 16290,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.37,
      "learning_rate": 2.6316758747697975e-05,
      "loss": 1.7533,
      "step": 2000
    },
    {
      "epoch": 0.37,
      "eval_loss": 1.023402452468872,
      "eval_runtime": 20.2636,
      "eval_samples_per_second": 57.69,
      "eval_steps_per_second": 3.652,
      "step": 2000
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.2633517495395946e-05,
      "loss": 1.2356,
      "step": 4000
    },
    {
      "epoch": 0.74,
      "eval_loss": 0.9132137298583984,
      "eval_runtime": 20.2624,
      "eval_samples_per_second": 57.693,
      "eval_steps_per_second": 3.652,
      "step": 4000
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.8950276243093924e-05,
      "loss": 1.0926,
      "step": 6000
    },
    {
      "epoch": 1.1,
      "eval_loss": 0.9214878082275391,
      "eval_runtime": 20.2606,
      "eval_samples_per_second": 57.698,
      "eval_steps_per_second": 3.652,
      "step": 6000
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.52670349907919e-05,
      "loss": 0.8857,
      "step": 8000
    },
    {
      "epoch": 1.47,
      "eval_loss": 0.948037326335907,
      "eval_runtime": 20.263,
      "eval_samples_per_second": 57.691,
      "eval_steps_per_second": 3.652,
      "step": 8000
    },
    {
      "epoch": 1.84,
      "learning_rate": 1.1583793738489871e-05,
      "loss": 0.8726,
      "step": 10000
    },
    {
      "epoch": 1.84,
      "eval_loss": 0.9162865877151489,
      "eval_runtime": 20.2461,
      "eval_samples_per_second": 57.739,
      "eval_steps_per_second": 3.655,
      "step": 10000
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.900552486187846e-06,
      "loss": 0.7355,
      "step": 12000
    },
    {
      "epoch": 2.21,
      "eval_loss": 0.9980245232582092,
      "eval_runtime": 20.2431,
      "eval_samples_per_second": 57.748,
      "eval_steps_per_second": 3.656,
      "step": 12000
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.21731123388582e-06,
      "loss": 0.6529,
      "step": 14000
    },
    {
      "epoch": 2.58,
      "eval_loss": 1.0005152225494385,
      "eval_runtime": 20.2501,
      "eval_samples_per_second": 57.728,
      "eval_steps_per_second": 3.654,
      "step": 14000
    },
    {
      "epoch": 2.95,
      "learning_rate": 5.340699815837937e-07,
      "loss": 0.6474,
      "step": 16000
    },
    {
      "epoch": 2.95,
      "eval_loss": 0.9974539875984192,
      "eval_runtime": 20.2537,
      "eval_samples_per_second": 57.718,
      "eval_steps_per_second": 3.654,
      "step": 16000
    }
  ],
  "max_steps": 16290,
  "num_train_epochs": 3,
  "total_flos": 6.809654777296896e+16,
  "trial_name": null,
  "trial_params": null
}