last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008090614886731391,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016181229773462783,
      "eval_loss": NaN,
      "eval_runtime": 89.4414,
      "eval_samples_per_second": 29.103,
      "eval_steps_per_second": 14.557,
      "step": 1
    },
    {
      "epoch": 0.0008090614886731392,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0016181229773462784,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0016181229773462784,
      "eval_loss": NaN,
      "eval_runtime": 87.4294,
      "eval_samples_per_second": 29.773,
      "eval_steps_per_second": 14.892,
      "step": 10
    },
    {
      "epoch": 0.0024271844660194173,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.003236245954692557,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.003236245954692557,
      "eval_loss": NaN,
      "eval_runtime": 87.5415,
      "eval_samples_per_second": 29.734,
      "eval_steps_per_second": 14.873,
      "step": 20
    },
    {
      "epoch": 0.0040453074433656954,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.0048543689320388345,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0048543689320388345,
      "eval_loss": NaN,
      "eval_runtime": 87.4413,
      "eval_samples_per_second": 29.769,
      "eval_steps_per_second": 14.89,
      "step": 30
    },
    {
      "epoch": 0.0056634304207119745,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.006472491909385114,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.006472491909385114,
      "eval_loss": NaN,
      "eval_runtime": 88.8162,
      "eval_samples_per_second": 29.308,
      "eval_steps_per_second": 14.659,
      "step": 40
    },
    {
      "epoch": 0.007281553398058253,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.008090614886731391,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.008090614886731391,
      "eval_loss": NaN,
      "eval_runtime": 88.5774,
      "eval_samples_per_second": 29.387,
      "eval_steps_per_second": 14.699,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 643300733485056.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}