checkpoint-2250/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.20536756126021,
  "eval_steps": 250,
  "global_step": 2250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.46674445740956827,
      "grad_norm": 6.355308532714844,
      "learning_rate": 9.899294610497411e-05,
      "loss": 3.4363,
      "step": 250
    },
    {
      "epoch": 0.46674445740956827,
      "eval_loss": 0.2021917849779129,
      "eval_runtime": 11.2146,
      "eval_samples_per_second": 7.758,
      "eval_steps_per_second": 7.758,
      "step": 250
    },
    {
      "epoch": 0.9334889148191365,
      "grad_norm": 2.1196815967559814,
      "learning_rate": 9.378729871108362e-05,
      "loss": 1.1086,
      "step": 500
    },
    {
      "epoch": 0.9334889148191365,
      "eval_loss": 0.13436643779277802,
      "eval_runtime": 11.2216,
      "eval_samples_per_second": 7.753,
      "eval_steps_per_second": 7.753,
      "step": 500
    },
    {
      "epoch": 1.4014002333722286,
      "grad_norm": 0.9552448391914368,
      "learning_rate": 8.459812499901582e-05,
      "loss": 0.817,
      "step": 750
    },
    {
      "epoch": 1.4014002333722286,
      "eval_loss": 0.12236687541007996,
      "eval_runtime": 11.2082,
      "eval_samples_per_second": 7.762,
      "eval_steps_per_second": 7.762,
      "step": 750
    },
    {
      "epoch": 1.8681446907817971,
      "grad_norm": 0.8024685978889465,
      "learning_rate": 7.226140518225348e-05,
      "loss": 0.7757,
      "step": 1000
    },
    {
      "epoch": 1.8681446907817971,
      "eval_loss": 0.11266180872917175,
      "eval_runtime": 11.2333,
      "eval_samples_per_second": 7.745,
      "eval_steps_per_second": 7.745,
      "step": 1000
    },
    {
      "epoch": 2.336056009334889,
      "grad_norm": 1.2465705871582031,
      "learning_rate": 5.7899465814766306e-05,
      "loss": 0.6768,
      "step": 1250
    },
    {
      "epoch": 2.336056009334889,
      "eval_loss": 0.11375312507152557,
      "eval_runtime": 11.2449,
      "eval_samples_per_second": 7.737,
      "eval_steps_per_second": 7.737,
      "step": 1250
    },
    {
      "epoch": 2.8028004667444573,
      "grad_norm": 1.350077748298645,
      "learning_rate": 4.281887672695056e-05,
      "loss": 0.6549,
      "step": 1500
    },
    {
      "epoch": 2.8028004667444573,
      "eval_loss": 0.10809295624494553,
      "eval_runtime": 11.217,
      "eval_samples_per_second": 7.756,
      "eval_steps_per_second": 7.756,
      "step": 1500
    },
    {
      "epoch": 3.2707117852975496,
      "grad_norm": 1.6034570932388306,
      "learning_rate": 2.8391586527038743e-05,
      "loss": 0.627,
      "step": 1750
    },
    {
      "epoch": 3.2707117852975496,
      "eval_loss": 0.10999161005020142,
      "eval_runtime": 11.2004,
      "eval_samples_per_second": 7.768,
      "eval_steps_per_second": 7.768,
      "step": 1750
    },
    {
      "epoch": 3.737456242707118,
      "grad_norm": 1.894575595855713,
      "learning_rate": 1.5930110302808637e-05,
      "loss": 0.5768,
      "step": 2000
    },
    {
      "epoch": 3.737456242707118,
      "eval_loss": 0.10841962695121765,
      "eval_runtime": 11.2,
      "eval_samples_per_second": 7.768,
      "eval_steps_per_second": 7.768,
      "step": 2000
    },
    {
      "epoch": 4.20536756126021,
      "grad_norm": 0.6385487914085388,
      "learning_rate": 6.5681242563502555e-06,
      "loss": 0.5417,
      "step": 2250
    },
    {
      "epoch": 4.20536756126021,
      "eval_loss": 0.1111413985490799,
      "eval_runtime": 11.2055,
      "eval_samples_per_second": 7.764,
      "eval_steps_per_second": 7.764,
      "step": 2250
    }
  ],
  "logging_steps": 250,
  "max_steps": 2675,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 250,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.5621221144725504e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}