last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.055401662049861494,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00554016620498615,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 10.7803,
      "step": 1
    },
    {
      "epoch": 0.00554016620498615,
      "eval_loss": 10.876238822937012,
      "eval_runtime": 1.1301,
      "eval_samples_per_second": 67.253,
      "eval_steps_per_second": 33.627,
      "step": 1
    },
    {
      "epoch": 0.0110803324099723,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 10.5694,
      "step": 2
    },
    {
      "epoch": 0.01662049861495845,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 10.5919,
      "step": 3
    },
    {
      "epoch": 0.01662049861495845,
      "eval_loss": 10.876238822937012,
      "eval_runtime": 1.0865,
      "eval_samples_per_second": 69.949,
      "eval_steps_per_second": 34.975,
      "step": 3
    },
    {
      "epoch": 0.0221606648199446,
      "grad_norm": 17.566349029541016,
      "learning_rate": 2e-05,
      "loss": 11.1469,
      "step": 4
    },
    {
      "epoch": 0.027700831024930747,
      "grad_norm": 17.940860748291016,
      "learning_rate": 4e-05,
      "loss": 10.8969,
      "step": 5
    },
    {
      "epoch": 0.0332409972299169,
      "grad_norm": 20.539535522460938,
      "learning_rate": 6e-05,
      "loss": 10.8141,
      "step": 6
    },
    {
      "epoch": 0.0332409972299169,
      "eval_loss": 10.789344787597656,
      "eval_runtime": 1.1219,
      "eval_samples_per_second": 67.744,
      "eval_steps_per_second": 33.872,
      "step": 6
    },
    {
      "epoch": 0.038781163434903045,
      "grad_norm": 23.39402961730957,
      "learning_rate": 8e-05,
      "loss": 10.8841,
      "step": 7
    },
    {
      "epoch": 0.0443213296398892,
      "grad_norm": 17.9848690032959,
      "learning_rate": 0.0001,
      "loss": 10.4167,
      "step": 8
    },
    {
      "epoch": 0.04986149584487535,
      "grad_norm": 16.139219284057617,
      "learning_rate": 0.00012,
      "loss": 10.8235,
      "step": 9
    },
    {
      "epoch": 0.04986149584487535,
      "eval_loss": 10.459603309631348,
      "eval_runtime": 1.1553,
      "eval_samples_per_second": 65.786,
      "eval_steps_per_second": 32.893,
      "step": 9
    },
    {
      "epoch": 0.055401662049861494,
      "grad_norm": 20.555219650268555,
      "learning_rate": 0.00014,
      "loss": 10.6594,
      "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
}