trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8710801393728222,
  "eval_steps": 100,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "grad_norm": 0.18273070454597473,
      "learning_rate": 1e-05,
      "loss": 1.1276,
      "step": 100
    },
    {
      "epoch": 0.17,
      "eval_acc": 0.9980372914622179,
      "eval_f1 score": 0.9988269794721408,
      "eval_loss": 0.009664908051490784,
      "eval_precision": 0.9994131455399061,
      "eval_recall": 0.9982415005861665,
      "eval_runtime": 35.543,
      "eval_samples_per_second": 57.339,
      "eval_steps_per_second": 1.801,
      "step": 100
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.015436016023159027,
      "learning_rate": 2e-05,
      "loss": 0.0073,
      "step": 200
    },
    {
      "epoch": 0.35,
      "eval_acc": 0.9980372914622179,
      "eval_f1 score": 0.9988283538371412,
      "eval_loss": 0.004532838240265846,
      "eval_precision": 0.9982435597189696,
      "eval_recall": 0.9994138335287222,
      "eval_runtime": 35.7648,
      "eval_samples_per_second": 56.983,
      "eval_steps_per_second": 1.789,
      "step": 200
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.009873462840914726,
      "learning_rate": 3e-05,
      "loss": 0.0029,
      "step": 300
    },
    {
      "epoch": 0.52,
      "eval_acc": 0.9995093228655545,
      "eval_f1 score": 0.9997068308413954,
      "eval_loss": 0.0031382227316498756,
      "eval_precision": 1.0,
      "eval_recall": 0.9994138335287222,
      "eval_runtime": 35.8701,
      "eval_samples_per_second": 56.816,
      "eval_steps_per_second": 1.784,
      "step": 300
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.00707666901871562,
      "learning_rate": 4e-05,
      "loss": 0.008,
      "step": 400
    },
    {
      "epoch": 0.7,
      "eval_acc": 1.0,
      "eval_f1 score": 1.0,
      "eval_loss": 0.00025456614093855023,
      "eval_precision": 1.0,
      "eval_recall": 1.0,
      "eval_runtime": 35.7717,
      "eval_samples_per_second": 56.972,
      "eval_steps_per_second": 1.789,
      "step": 400
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.05433596670627594,
      "learning_rate": 5e-05,
      "loss": 0.0103,
      "step": 500
    },
    {
      "epoch": 0.87,
      "eval_acc": 0.9985279685966634,
      "eval_f1 score": 0.9991199765327076,
      "eval_loss": 0.005017507821321487,
      "eval_precision": 1.0,
      "eval_recall": 0.9982415005861665,
      "eval_runtime": 35.9127,
      "eval_samples_per_second": 56.749,
      "eval_steps_per_second": 1.782,
      "step": 500
    }
  ],
  "logging_steps": 100,
  "max_steps": 574,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 2119478378496000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}