last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0030698388334612432,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003069838833461243,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0003069838833461243,
      "eval_loss": NaN,
      "eval_runtime": 80.6231,
      "eval_samples_per_second": 17.017,
      "eval_steps_per_second": 8.509,
      "step": 1
    },
    {
      "epoch": 0.0006139677666922486,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 1.7251,
      "step": 2
    },
    {
      "epoch": 0.000920951650038373,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.000920951650038373,
      "eval_loss": NaN,
      "eval_runtime": 79.5625,
      "eval_samples_per_second": 17.244,
      "eval_steps_per_second": 8.622,
      "step": 3
    },
    {
      "epoch": 0.0012279355333844973,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 4.0081,
      "step": 4
    },
    {
      "epoch": 0.0015349194167306216,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 8.9374,
      "step": 5
    },
    {
      "epoch": 0.001841903300076746,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.001841903300076746,
      "eval_loss": NaN,
      "eval_runtime": 79.6113,
      "eval_samples_per_second": 17.234,
      "eval_steps_per_second": 8.617,
      "step": 6
    },
    {
      "epoch": 0.00214888718342287,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 8.0263,
      "step": 7
    },
    {
      "epoch": 0.0024558710667689945,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 4.4873,
      "step": 8
    },
    {
      "epoch": 0.002762854950115119,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 5.249,
      "step": 9
    },
    {
      "epoch": 0.002762854950115119,
      "eval_loss": NaN,
      "eval_runtime": 79.5838,
      "eval_samples_per_second": 17.24,
      "eval_steps_per_second": 8.62,
      "step": 9
    },
    {
      "epoch": 0.0030698388334612432,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 1.3396,
      "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": 3505345611694080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}