trainer_state.json

{
  "best_metric": 0.2616243362426758,
  "best_model_checkpoint": "./checkpoints/checkpoint-1964",
  "epoch": 4.0,
  "global_step": 1964,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8884644766997708,
      "eval_f1": 0.8884644766997708,
      "eval_loss": 0.4412623643875122,
      "eval_precision": 0.8884644766997708,
      "eval_recall": 0.8884644766997708,
      "eval_runtime": 43.6404,
      "eval_samples_per_second": 59.99,
      "eval_steps_per_second": 1.879,
      "step": 491
    },
    {
      "epoch": 1.02,
      "learning_rate": 2e-05,
      "loss": 1.3111,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9037433155080213,
      "eval_f1": 0.9037433155080212,
      "eval_loss": 0.31344300508499146,
      "eval_precision": 0.9037433155080213,
      "eval_recall": 0.9037433155080213,
      "eval_runtime": 43.7135,
      "eval_samples_per_second": 59.89,
      "eval_steps_per_second": 1.876,
      "step": 982
    },
    {
      "epoch": 2.04,
      "learning_rate": 1.591169255928046e-05,
      "loss": 0.3547,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9113827349121467,
      "eval_f1": 0.9113827349121467,
      "eval_loss": 0.2822723388671875,
      "eval_precision": 0.9113827349121467,
      "eval_recall": 0.9113827349121467,
      "eval_runtime": 43.7763,
      "eval_samples_per_second": 59.804,
      "eval_steps_per_second": 1.873,
      "step": 1473
    },
    {
      "epoch": 3.05,
      "learning_rate": 1.1823385118560918e-05,
      "loss": 0.2375,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9182582123758595,
      "eval_f1": 0.9182582123758595,
      "eval_loss": 0.2616243362426758,
      "eval_precision": 0.9182582123758595,
      "eval_recall": 0.9182582123758595,
      "eval_runtime": 43.7951,
      "eval_samples_per_second": 59.778,
      "eval_steps_per_second": 1.872,
      "step": 1964
    }
  ],
  "max_steps": 2946,
  "num_train_epochs": 6,
  "total_flos": 1100711158866000.0,
  "trial_name": null,
  "trial_params": null
}