last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02820476660555634,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005640953321111268,
      "eval_loss": NaN,
      "eval_runtime": 51.6032,
      "eval_samples_per_second": 14.476,
      "eval_steps_per_second": 7.248,
      "step": 1
    },
    {
      "epoch": 0.0028204766605556338,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0056409533211112676,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0056409533211112676,
      "eval_loss": NaN,
      "eval_runtime": 52.8155,
      "eval_samples_per_second": 14.144,
      "eval_steps_per_second": 7.081,
      "step": 10
    },
    {
      "epoch": 0.008461429981666902,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.011281906642222535,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.011281906642222535,
      "eval_loss": NaN,
      "eval_runtime": 53.6606,
      "eval_samples_per_second": 13.921,
      "eval_steps_per_second": 6.97,
      "step": 20
    },
    {
      "epoch": 0.01410238330277817,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.016922859963333804,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.016922859963333804,
      "eval_loss": NaN,
      "eval_runtime": 53.926,
      "eval_samples_per_second": 13.852,
      "eval_steps_per_second": 6.935,
      "step": 30
    },
    {
      "epoch": 0.019743336623889437,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.02256381328444507,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.02256381328444507,
      "eval_loss": NaN,
      "eval_runtime": 54.0127,
      "eval_samples_per_second": 13.83,
      "eval_steps_per_second": 6.924,
      "step": 40
    },
    {
      "epoch": 0.025384289945000704,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.02820476660555634,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.02820476660555634,
      "eval_loss": NaN,
      "eval_runtime": 54.1966,
      "eval_samples_per_second": 13.783,
      "eval_steps_per_second": 6.901,
      "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": 8756746131603456.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}