last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0002584296522182955,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.5842965221829554e-05,
      "grad_norm": 3.435499906539917,
      "learning_rate": 2e-05,
      "loss": 1.4246,
      "step": 1
    },
    {
      "epoch": 2.5842965221829554e-05,
      "eval_loss": 2.1122167110443115,
      "eval_runtime": 633.4485,
      "eval_samples_per_second": 25.721,
      "eval_steps_per_second": 12.861,
      "step": 1
    },
    {
      "epoch": 5.168593044365911e-05,
      "grad_norm": 2.8410768508911133,
      "learning_rate": 4e-05,
      "loss": 1.687,
      "step": 2
    },
    {
      "epoch": 7.752889566548866e-05,
      "grad_norm": 2.6962289810180664,
      "learning_rate": 6e-05,
      "loss": 1.2136,
      "step": 3
    },
    {
      "epoch": 7.752889566548866e-05,
      "eval_loss": 2.083096742630005,
      "eval_runtime": 631.5229,
      "eval_samples_per_second": 25.8,
      "eval_steps_per_second": 12.901,
      "step": 3
    },
    {
      "epoch": 0.00010337186088731822,
      "grad_norm": 2.957984447479248,
      "learning_rate": 8e-05,
      "loss": 1.3846,
      "step": 4
    },
    {
      "epoch": 0.00012921482610914775,
      "grad_norm": 3.768646240234375,
      "learning_rate": 0.0001,
      "loss": 3.3657,
      "step": 5
    },
    {
      "epoch": 0.0001550577913309773,
      "grad_norm": 2.4247074127197266,
      "learning_rate": 0.00012,
      "loss": 1.0722,
      "step": 6
    },
    {
      "epoch": 0.0001550577913309773,
      "eval_loss": 1.7299082279205322,
      "eval_runtime": 631.1001,
      "eval_samples_per_second": 25.817,
      "eval_steps_per_second": 12.909,
      "step": 6
    },
    {
      "epoch": 0.00018090075655280687,
      "grad_norm": 3.7721331119537354,
      "learning_rate": 0.00014,
      "loss": 1.0112,
      "step": 7
    },
    {
      "epoch": 0.00020674372177463643,
      "grad_norm": 2.6731598377227783,
      "learning_rate": 0.00016,
      "loss": 0.8786,
      "step": 8
    },
    {
      "epoch": 0.00023258668699646597,
      "grad_norm": 2.0309290885925293,
      "learning_rate": 0.00018,
      "loss": 0.7873,
      "step": 9
    },
    {
      "epoch": 0.00023258668699646597,
      "eval_loss": 1.1887303590774536,
      "eval_runtime": 631.1223,
      "eval_samples_per_second": 25.816,
      "eval_steps_per_second": 12.909,
      "step": 9
    },
    {
      "epoch": 0.0002584296522182955,
      "grad_norm": 2.100818634033203,
      "learning_rate": 0.0002,
      "loss": 0.6941,
      "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": 2450672628793344.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}