trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.78048780487805,
  "eval_steps": 500,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.49,
      "learning_rate": 0.00019864823963335033,
      "loss": 2.6568,
      "step": 5
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00019428459070078416,
      "loss": 2.4777,
      "step": 10
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001870370067496355,
      "loss": 2.3358,
      "step": 15
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00017712717364608328,
      "loss": 2.2961,
      "step": 20
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0001648582089462756,
      "loss": 2.231,
      "step": 25
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.00015060539027168316,
      "loss": 2.1653,
      "step": 30
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0001348046764845511,
      "loss": 2.0799,
      "step": 35
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.000117939372772616,
      "loss": 2.1272,
      "step": 40
    },
    {
      "epoch": 4.39,
      "learning_rate": 0.00010052534752553063,
      "loss": 2.0387,
      "step": 45
    },
    {
      "epoch": 4.88,
      "learning_rate": 8.309525318258826e-05,
      "loss": 1.9852,
      "step": 50
    },
    {
      "epoch": 5.37,
      "learning_rate": 6.618223369742117e-05,
      "loss": 1.9507,
      "step": 55
    },
    {
      "epoch": 5.85,
      "learning_rate": 5.030361696847705e-05,
      "loss": 1.9503,
      "step": 60
    },
    {
      "epoch": 6.34,
      "learning_rate": 3.594509104393951e-05,
      "loss": 1.9554,
      "step": 65
    },
    {
      "epoch": 6.83,
      "learning_rate": 2.3545848112279833e-05,
      "loss": 1.8693,
      "step": 70
    },
    {
      "epoch": 7.32,
      "learning_rate": 1.3485150687433167e-05,
      "loss": 1.8875,
      "step": 75
    },
    {
      "epoch": 7.8,
      "learning_rate": 6.070730896136479e-06,
      "loss": 1.8837,
      "step": 80
    },
    {
      "epoch": 8.29,
      "learning_rate": 1.5293777048546509e-06,
      "loss": 1.889,
      "step": 85
    },
    {
      "epoch": 8.78,
      "learning_rate": 0.0,
      "loss": 1.8582,
      "step": 90
    },
    {
      "epoch": 8.78,
      "step": 90,
      "total_flos": 1.573518009434112e+16,
      "train_loss": 2.090980339050293,
      "train_runtime": 141.8274,
      "train_samples_per_second": 2.602,
      "train_steps_per_second": 0.635
    }
  ],
  "logging_steps": 5,
  "max_steps": 90,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 10,
  "total_flos": 1.573518009434112e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}