last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0015160703456640388,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00015160703456640388,
      "grad_norm": 16.83426284790039,
      "learning_rate": 2e-05,
      "loss": 11.4822,
      "step": 1
    },
    {
      "epoch": 0.00015160703456640388,
      "eval_loss": 11.705659866333008,
      "eval_runtime": 94.653,
      "eval_samples_per_second": 29.349,
      "eval_steps_per_second": 14.675,
      "step": 1
    },
    {
      "epoch": 0.00030321406913280777,
      "grad_norm": 24.82701301574707,
      "learning_rate": 4e-05,
      "loss": 11.0969,
      "step": 2
    },
    {
      "epoch": 0.00045482110369921165,
      "grad_norm": 12.29438591003418,
      "learning_rate": 6e-05,
      "loss": 12.4466,
      "step": 3
    },
    {
      "epoch": 0.00045482110369921165,
      "eval_loss": 11.678789138793945,
      "eval_runtime": 91.2328,
      "eval_samples_per_second": 30.45,
      "eval_steps_per_second": 15.225,
      "step": 3
    },
    {
      "epoch": 0.0006064281382656155,
      "grad_norm": 21.273990631103516,
      "learning_rate": 8e-05,
      "loss": 11.3351,
      "step": 4
    },
    {
      "epoch": 0.0007580351728320194,
      "grad_norm": 16.648815155029297,
      "learning_rate": 0.0001,
      "loss": 11.359,
      "step": 5
    },
    {
      "epoch": 0.0009096422073984233,
      "grad_norm": 19.553115844726562,
      "learning_rate": 0.00012,
      "loss": 11.2096,
      "step": 6
    },
    {
      "epoch": 0.0009096422073984233,
      "eval_loss": 11.353554725646973,
      "eval_runtime": 91.1262,
      "eval_samples_per_second": 30.485,
      "eval_steps_per_second": 15.243,
      "step": 6
    },
    {
      "epoch": 0.001061249241964827,
      "grad_norm": 15.552982330322266,
      "learning_rate": 0.00014,
      "loss": 10.9988,
      "step": 7
    },
    {
      "epoch": 0.001212856276531231,
      "grad_norm": 14.931906700134277,
      "learning_rate": 0.00016,
      "loss": 11.3466,
      "step": 8
    },
    {
      "epoch": 0.0013644633110976349,
      "grad_norm": 15.584132194519043,
      "learning_rate": 0.00018,
      "loss": 10.5936,
      "step": 9
    },
    {
      "epoch": 0.0013644633110976349,
      "eval_loss": 10.818541526794434,
      "eval_runtime": 91.1349,
      "eval_samples_per_second": 30.482,
      "eval_steps_per_second": 15.241,
      "step": 9
    },
    {
      "epoch": 0.0015160703456640388,
      "grad_norm": 13.891613960266113,
      "learning_rate": 0.0002,
      "loss": 10.5964,
      "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": 5429904015360.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}