last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003182432970005569,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003182432970005569,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.4965,
      "step": 1
    },
    {
      "epoch": 0.0003182432970005569,
      "eval_loss": NaN,
      "eval_runtime": 75.1726,
      "eval_samples_per_second": 17.613,
      "eval_steps_per_second": 8.806,
      "step": 1
    },
    {
      "epoch": 0.0006364865940011138,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.0009547298910016707,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0009547298910016707,
      "eval_loss": NaN,
      "eval_runtime": 74.1338,
      "eval_samples_per_second": 17.86,
      "eval_steps_per_second": 8.93,
      "step": 3
    },
    {
      "epoch": 0.0012729731880022277,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0015912164850027846,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0019094597820033415,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 1.06,
      "step": 6
    },
    {
      "epoch": 0.0019094597820033415,
      "eval_loss": NaN,
      "eval_runtime": 74.1717,
      "eval_samples_per_second": 17.85,
      "eval_steps_per_second": 8.925,
      "step": 6
    },
    {
      "epoch": 0.0022277030790038986,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 1.1596,
      "step": 7
    },
    {
      "epoch": 0.0025459463760044553,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0028641896730050125,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0028641896730050125,
      "eval_loss": NaN,
      "eval_runtime": 74.1463,
      "eval_samples_per_second": 17.857,
      "eval_steps_per_second": 8.928,
      "step": 9
    },
    {
      "epoch": 0.003182432970005569,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4283942672793600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}