last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.07507507507507508,
  "eval_steps": 5,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003003003003003003,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.003003003003003003,
      "eval_loss": NaN,
      "eval_runtime": 62.1042,
      "eval_samples_per_second": 4.525,
      "eval_steps_per_second": 2.27,
      "step": 1
    },
    {
      "epoch": 0.006006006006006006,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.009009009009009009,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.012012012012012012,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.015015015015015015,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.015015015015015015,
      "eval_loss": NaN,
      "eval_runtime": 62.1167,
      "eval_samples_per_second": 4.524,
      "eval_steps_per_second": 2.27,
      "step": 5
    },
    {
      "epoch": 0.018018018018018018,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.021021021021021023,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.024024024024024024,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.02702702702702703,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.03003003003003003,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.03003003003003003,
      "eval_loss": NaN,
      "eval_runtime": 62.2079,
      "eval_samples_per_second": 4.517,
      "eval_steps_per_second": 2.267,
      "step": 10
    },
    {
      "epoch": 0.03303303303303303,
      "grad_norm": NaN,
      "learning_rate": 0.0001996917333733128,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.036036036036036036,
      "grad_norm": NaN,
      "learning_rate": 0.00019876883405951377,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.03903903903903904,
      "grad_norm": NaN,
      "learning_rate": 0.00019723699203976766,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.042042042042042045,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.04504504504504504,
      "grad_norm": NaN,
      "learning_rate": 0.0001923879532511287,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.04504504504504504,
      "eval_loss": NaN,
      "eval_runtime": 62.3137,
      "eval_samples_per_second": 4.509,
      "eval_steps_per_second": 2.263,
      "step": 15
    },
    {
      "epoch": 0.04804804804804805,
      "grad_norm": NaN,
      "learning_rate": 0.0001891006524188368,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.05105105105105105,
      "grad_norm": NaN,
      "learning_rate": 0.00018526401643540922,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.05405405405405406,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.057057057057057055,
      "grad_norm": NaN,
      "learning_rate": 0.0001760405965600031,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.06006006006006006,
      "grad_norm": NaN,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.06006006006006006,
      "eval_loss": NaN,
      "eval_runtime": 62.11,
      "eval_samples_per_second": 4.524,
      "eval_steps_per_second": 2.27,
      "step": 20
    },
    {
      "epoch": 0.06306306306306306,
      "grad_norm": NaN,
      "learning_rate": 0.00016494480483301836,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.06606606606606606,
      "grad_norm": NaN,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.06906906906906907,
      "grad_norm": NaN,
      "learning_rate": 0.0001522498564715949,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.07207207207207207,
      "grad_norm": NaN,
      "learning_rate": 0.00014539904997395468,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.07507507507507508,
      "grad_norm": NaN,
      "learning_rate": 0.000138268343236509,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.07507507507507508,
      "eval_loss": NaN,
      "eval_runtime": 62.2445,
      "eval_samples_per_second": 4.514,
      "eval_steps_per_second": 2.265,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 50,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.51719368163328e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}