trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0002304099201943822,
  "eval_steps": 500,
  "global_step": 11,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0946356381307473e-05,
      "grad_norm": 0.1689453125,
      "learning_rate": 5e-06,
      "loss": 2.9393,
      "step": 1
    },
    {
      "epoch": 4.1892712762614946e-05,
      "grad_norm": 0.16015625,
      "learning_rate": 1e-05,
      "loss": 2.731,
      "step": 2
    },
    {
      "epoch": 6.283906914392241e-05,
      "grad_norm": 0.18359375,
      "learning_rate": 1.5e-05,
      "loss": 3.0316,
      "step": 3
    },
    {
      "epoch": 8.378542552522989e-05,
      "grad_norm": 0.142578125,
      "learning_rate": 2e-05,
      "loss": 2.6462,
      "step": 4
    },
    {
      "epoch": 0.00010473178190653736,
      "grad_norm": 0.1484375,
      "learning_rate": 2.5e-05,
      "loss": 2.539,
      "step": 5
    },
    {
      "epoch": 0.00012567813828784482,
      "grad_norm": 0.1767578125,
      "learning_rate": 3e-05,
      "loss": 2.6637,
      "step": 6
    },
    {
      "epoch": 0.0001466244946691523,
      "grad_norm": 0.154296875,
      "learning_rate": 3.5000000000000004e-05,
      "loss": 2.7705,
      "step": 7
    },
    {
      "epoch": 0.00016757085105045978,
      "grad_norm": 0.1572265625,
      "learning_rate": 4e-05,
      "loss": 2.7461,
      "step": 8
    },
    {
      "epoch": 0.00018851720743176724,
      "grad_norm": 0.1650390625,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 2.8261,
      "step": 9
    },
    {
      "epoch": 0.00020946356381307472,
      "grad_norm": 0.166015625,
      "learning_rate": 5e-05,
      "loss": 3.0726,
      "step": 10
    },
    {
      "epoch": 0.0002304099201943822,
      "grad_norm": 0.1474609375,
      "learning_rate": 5.5e-05,
      "loss": 2.4639,
      "step": 11
    }
  ],
  "logging_steps": 1,
  "max_steps": 47741,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2842846939840512.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}