trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5,
  "eval_steps": 3,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008333333333333333,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.1097,
      "step": 1
    },
    {
      "epoch": 0.025,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.1424,
      "step": 3
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.4957225914582386,
      "learning_rate": 6.000000000000001e-07,
      "loss": 1.1714,
      "step": 6
    },
    {
      "epoch": 0.075,
      "grad_norm": 1.0203233450225075,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 1.1459,
      "step": 9
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.1952212091375145,
      "learning_rate": 3e-06,
      "loss": 1.1465,
      "step": 12
    },
    {
      "epoch": 0.125,
      "grad_norm": 1.832578954318156,
      "learning_rate": 2.839090909090909e-06,
      "loss": 1.1517,
      "step": 15
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.0877339332216878,
      "learning_rate": 2.678181818181818e-06,
      "loss": 1.0991,
      "step": 18
    },
    {
      "epoch": 0.175,
      "grad_norm": 1.4203314600042332,
      "learning_rate": 2.5172727272727275e-06,
      "loss": 1.1481,
      "step": 21
    },
    {
      "epoch": 0.2,
      "grad_norm": 2.543863444651869,
      "learning_rate": 2.3563636363636366e-06,
      "loss": 1.151,
      "step": 24
    },
    {
      "epoch": 0.225,
      "grad_norm": 2.2672694357791805,
      "learning_rate": 2.1954545454545456e-06,
      "loss": 1.1132,
      "step": 27
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7570282713584158,
      "learning_rate": 2.0345454545454546e-06,
      "loss": 1.1043,
      "step": 30
    },
    {
      "epoch": 0.275,
      "grad_norm": 0.7556695763204282,
      "learning_rate": 1.9272727272727273e-06,
      "loss": 1.0492,
      "step": 33
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.5250919282965105,
      "learning_rate": 1.8199999999999997e-06,
      "loss": 1.1206,
      "step": 36
    },
    {
      "epoch": 0.325,
      "grad_norm": 2.402860972914852,
      "learning_rate": 1.659090909090909e-06,
      "loss": 1.0908,
      "step": 39
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.9250227812759156,
      "learning_rate": 1.4981818181818184e-06,
      "loss": 1.0982,
      "step": 42
    },
    {
      "epoch": 0.375,
      "grad_norm": 1.0119045287556745,
      "learning_rate": 1.3372727272727274e-06,
      "loss": 1.1147,
      "step": 45
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.7972164809866773,
      "learning_rate": 1.1763636363636364e-06,
      "loss": 1.1004,
      "step": 48
    },
    {
      "epoch": 0.425,
      "grad_norm": 1.4489209555210352,
      "learning_rate": 1.0154545454545454e-06,
      "loss": 1.0982,
      "step": 51
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.7531358827177462,
      "learning_rate": 8.545454545454544e-07,
      "loss": 1.0794,
      "step": 54
    },
    {
      "epoch": 0.475,
      "grad_norm": 0.7368058065522421,
      "learning_rate": 6.936363636363635e-07,
      "loss": 1.0948,
      "step": 57
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.2350079863263905,
      "learning_rate": 5.327272727272729e-07,
      "loss": 1.1091,
      "step": 60
    },
    {
      "epoch": 0.5,
      "step": 60,
      "total_flos": 238563958456320.0,
      "train_loss": 1.1158998648325602,
      "train_runtime": 19061.2793,
      "train_samples_per_second": 0.402,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 3,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 12,
  "total_flos": 238563958456320.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}