trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4,
  "eval_steps": 25,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 0.2926064431667328,
      "learning_rate": 0.00015151515151515152,
      "loss": 1.6323,
      "step": 25
    },
    {
      "epoch": 0.1,
      "eval_loss": 1.3890495300292969,
      "eval_runtime": 172.9903,
      "eval_samples_per_second": 2.885,
      "eval_steps_per_second": 0.185,
      "step": 25
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.34952306747436523,
      "learning_rate": 0.00010101010101010102,
      "loss": 1.2079,
      "step": 50
    },
    {
      "epoch": 0.2,
      "eval_loss": 1.3851372003555298,
      "eval_runtime": 172.917,
      "eval_samples_per_second": 2.886,
      "eval_steps_per_second": 0.185,
      "step": 50
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.23053358495235443,
      "learning_rate": 5.050505050505051e-05,
      "loss": 1.4311,
      "step": 75
    },
    {
      "epoch": 0.3,
      "eval_loss": 1.3527110815048218,
      "eval_runtime": 173.0035,
      "eval_samples_per_second": 2.884,
      "eval_steps_per_second": 0.185,
      "step": 75
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5876684188842773,
      "learning_rate": 0.0,
      "loss": 1.1624,
      "step": 100
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.3475642204284668,
      "eval_runtime": 172.9239,
      "eval_samples_per_second": 2.886,
      "eval_steps_per_second": 0.185,
      "step": 100
    }
  ],
  "logging_steps": 25,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3660831696568320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}