last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004943398091848336,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.886796183696674e-05,
      "eval_loss": NaN,
      "eval_runtime": 184.2252,
      "eval_samples_per_second": 23.118,
      "eval_steps_per_second": 11.562,
      "step": 1
    },
    {
      "epoch": 0.0004943398091848336,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0009886796183696672,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0009886796183696672,
      "eval_loss": NaN,
      "eval_runtime": 183.0926,
      "eval_samples_per_second": 23.261,
      "eval_steps_per_second": 11.633,
      "step": 10
    },
    {
      "epoch": 0.001483019427554501,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0019773592367393344,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0019773592367393344,
      "eval_loss": NaN,
      "eval_runtime": 183.9555,
      "eval_samples_per_second": 23.152,
      "eval_steps_per_second": 11.579,
      "step": 20
    },
    {
      "epoch": 0.002471699045924168,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.002966038855109002,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.002966038855109002,
      "eval_loss": NaN,
      "eval_runtime": 177.7137,
      "eval_samples_per_second": 23.966,
      "eval_steps_per_second": 11.986,
      "step": 30
    },
    {
      "epoch": 0.0034603786642938357,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.003954718473478669,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.003954718473478669,
      "eval_loss": NaN,
      "eval_runtime": 184.3478,
      "eval_samples_per_second": 23.103,
      "eval_steps_per_second": 11.554,
      "step": 40
    },
    {
      "epoch": 0.0044490582826635025,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.004943398091848336,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.004943398091848336,
      "eval_loss": NaN,
      "eval_runtime": 182.2139,
      "eval_samples_per_second": 23.374,
      "eval_steps_per_second": 11.69,
      "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": 1621491292569600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}