last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05037783375314862,
  "eval_steps": 3,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010075566750629723,
      "grad_norm": 0.11719062924385071,
      "learning_rate": 2e-05,
      "loss": 11.7756,
      "step": 1
    },
    {
      "epoch": 0.010075566750629723,
      "eval_loss": 11.775632858276367,
      "eval_runtime": 1.4254,
      "eval_samples_per_second": 58.929,
      "eval_steps_per_second": 29.464,
      "step": 1
    },
    {
      "epoch": 0.020151133501259445,
      "grad_norm": 0.14791575074195862,
      "learning_rate": 4e-05,
      "loss": 11.7752,
      "step": 2
    },
    {
      "epoch": 0.030226700251889168,
      "grad_norm": 0.14653781056404114,
      "learning_rate": 6e-05,
      "loss": 11.7731,
      "step": 3
    },
    {
      "epoch": 0.030226700251889168,
      "eval_loss": 11.775500297546387,
      "eval_runtime": 1.4206,
      "eval_samples_per_second": 59.13,
      "eval_steps_per_second": 29.565,
      "step": 3
    },
    {
      "epoch": 0.04030226700251889,
      "grad_norm": 0.15554942190647125,
      "learning_rate": 8e-05,
      "loss": 11.7685,
      "step": 4
    },
    {
      "epoch": 0.05037783375314862,
      "grad_norm": 0.1331409513950348,
      "learning_rate": 0.0001,
      "loss": 11.7752,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2037000437760.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}