last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0031257325935766197,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00031257325935766194,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 5.6661,
      "step": 1
    },
    {
      "epoch": 0.00031257325935766194,
      "eval_loss": 7.102509498596191,
      "eval_runtime": 153.2085,
      "eval_samples_per_second": 35.168,
      "eval_steps_per_second": 4.399,
      "step": 1
    },
    {
      "epoch": 0.0006251465187153239,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 5.1426,
      "step": 2
    },
    {
      "epoch": 0.0009377197780729859,
      "grad_norm": 3.7997171878814697,
      "learning_rate": 2e-05,
      "loss": 5.1073,
      "step": 3
    },
    {
      "epoch": 0.0009377197780729859,
      "eval_loss": 7.102509498596191,
      "eval_runtime": 147.6213,
      "eval_samples_per_second": 36.499,
      "eval_steps_per_second": 4.566,
      "step": 3
    },
    {
      "epoch": 0.0012502930374306478,
      "grad_norm": 3.34021258354187,
      "learning_rate": 4e-05,
      "loss": 5.0575,
      "step": 4
    },
    {
      "epoch": 0.0015628662967883099,
      "grad_norm": 3.7570230960845947,
      "learning_rate": 6e-05,
      "loss": 5.2018,
      "step": 5
    },
    {
      "epoch": 0.0018754395561459717,
      "grad_norm": 3.476313591003418,
      "learning_rate": 8e-05,
      "loss": 4.5102,
      "step": 6
    },
    {
      "epoch": 0.0018754395561459717,
      "eval_loss": 6.427522659301758,
      "eval_runtime": 147.6303,
      "eval_samples_per_second": 36.497,
      "eval_steps_per_second": 4.565,
      "step": 6
    },
    {
      "epoch": 0.002188012815503634,
      "grad_norm": 3.991544485092163,
      "learning_rate": 0.0001,
      "loss": 4.4218,
      "step": 7
    },
    {
      "epoch": 0.0025005860748612955,
      "grad_norm": 3.9677493572235107,
      "learning_rate": 0.00012,
      "loss": 3.3807,
      "step": 8
    },
    {
      "epoch": 0.0028131593342189576,
      "grad_norm": 6.058411598205566,
      "learning_rate": 0.00014,
      "loss": 4.0116,
      "step": 9
    },
    {
      "epoch": 0.0028131593342189576,
      "eval_loss": 3.9740207195281982,
      "eval_runtime": 147.5358,
      "eval_samples_per_second": 36.52,
      "eval_steps_per_second": 4.568,
      "step": 9
    },
    {
      "epoch": 0.0031257325935766197,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 2.9843,
      "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": 5.935098268483584e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}