last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00018127846638417438,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.812784663841744e-05,
      "grad_norm": 1.2593621015548706,
      "learning_rate": 2e-05,
      "loss": 2.7437,
      "step": 1
    },
    {
      "epoch": 1.812784663841744e-05,
      "eval_loss": 3.025801420211792,
      "eval_runtime": 794.4669,
      "eval_samples_per_second": 29.236,
      "eval_steps_per_second": 14.619,
      "step": 1
    },
    {
      "epoch": 3.625569327683488e-05,
      "grad_norm": 2.2349255084991455,
      "learning_rate": 4e-05,
      "loss": 2.9811,
      "step": 2
    },
    {
      "epoch": 5.4383539915252315e-05,
      "grad_norm": 2.0108678340911865,
      "learning_rate": 6e-05,
      "loss": 2.9677,
      "step": 3
    },
    {
      "epoch": 5.4383539915252315e-05,
      "eval_loss": 3.0133819580078125,
      "eval_runtime": 832.8088,
      "eval_samples_per_second": 27.89,
      "eval_steps_per_second": 13.946,
      "step": 3
    },
    {
      "epoch": 7.251138655366975e-05,
      "grad_norm": 1.0544321537017822,
      "learning_rate": 8e-05,
      "loss": 2.7406,
      "step": 4
    },
    {
      "epoch": 9.063923319208719e-05,
      "grad_norm": 1.504389762878418,
      "learning_rate": 0.0001,
      "loss": 2.7452,
      "step": 5
    },
    {
      "epoch": 0.00010876707983050463,
      "grad_norm": 1.5998446941375732,
      "learning_rate": 0.00012,
      "loss": 2.7127,
      "step": 6
    },
    {
      "epoch": 0.00010876707983050463,
      "eval_loss": 2.7751402854919434,
      "eval_runtime": 798.7493,
      "eval_samples_per_second": 29.079,
      "eval_steps_per_second": 14.54,
      "step": 6
    },
    {
      "epoch": 0.00012689492646892207,
      "grad_norm": 1.0239311456680298,
      "learning_rate": 0.00014,
      "loss": 2.1336,
      "step": 7
    },
    {
      "epoch": 0.0001450227731073395,
      "grad_norm": 0.9778740406036377,
      "learning_rate": 0.00016,
      "loss": 2.3313,
      "step": 8
    },
    {
      "epoch": 0.00016315061974575695,
      "grad_norm": 0.8723558187484741,
      "learning_rate": 0.00018,
      "loss": 2.0553,
      "step": 9
    },
    {
      "epoch": 0.00016315061974575695,
      "eval_loss": 2.4100735187530518,
      "eval_runtime": 782.4358,
      "eval_samples_per_second": 29.686,
      "eval_steps_per_second": 14.843,
      "step": 9
    },
    {
      "epoch": 0.00018127846638417438,
      "grad_norm": 1.1135058403015137,
      "learning_rate": 0.0002,
      "loss": 2.31,
      "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": 1131250949554176.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}