trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9846153846153847,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 0.0,
      "loss": 1.9361,
      "step": 2
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0,
      "loss": 2.1523,
      "step": 4
    },
    {
      "epoch": 0.12,
      "learning_rate": 1e-05,
      "loss": 2.2792,
      "step": 6
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9976687691905394e-05,
      "loss": 2.2289,
      "step": 8
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.9906859460363307e-05,
      "loss": 2.2699,
      "step": 10
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.979084087682323e-05,
      "loss": 2.5859,
      "step": 12
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.9422609221188208e-05,
      "loss": 1.8846,
      "step": 14
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.8878852184023754e-05,
      "loss": 2.2629,
      "step": 16
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.816969893010442e-05,
      "loss": 2.1982,
      "step": 18
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.730835964278124e-05,
      "loss": 2.2893,
      "step": 20
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.631087944326053e-05,
      "loss": 2.1624,
      "step": 22
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.5195839500354337e-05,
      "loss": 2.5566,
      "step": 24
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.3984010898462417e-05,
      "loss": 1.9081,
      "step": 26
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.2697967711570243e-05,
      "loss": 1.9444,
      "step": 28
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.1361666490962468e-05,
      "loss": 2.1857,
      "step": 30
    },
    {
      "epoch": 0.66,
      "learning_rate": 1e-05,
      "loss": 2.1617,
      "step": 32
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.638333509037537e-06,
      "loss": 2.3324,
      "step": 34
    },
    {
      "epoch": 0.74,
      "learning_rate": 7.3020322884297565e-06,
      "loss": 2.5691,
      "step": 36
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.015989101537586e-06,
      "loss": 1.7745,
      "step": 38
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.804160499645667e-06,
      "loss": 2.1199,
      "step": 40
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.689120556739475e-06,
      "loss": 2.3115,
      "step": 42
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.691640357218759e-06,
      "loss": 2.1067,
      "step": 44
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.8303010698955803e-06,
      "loss": 2.1117,
      "step": 46
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.121147815976248e-06,
      "loss": 2.3375,
      "step": 48
    }
  ],
  "logging_steps": 2,
  "max_steps": 48,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 4218566835981312.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}