trainer_state.json

{
  "best_metric": 0.9029529506581548,
  "best_model_checkpoint": "muppet_base_divergence_classifier_v1.0/checkpoint-4800",
  "epoch": 2.1248339973439574,
  "eval_steps": 480,
  "global_step": 4800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21248339973439576,
      "eval_accuracy": 0.8745519713261649,
      "eval_f1": 0.874329560354496,
      "eval_loss": 0.4335894286632538,
      "eval_runtime": 3.9801,
      "eval_samples_per_second": 140.196,
      "eval_steps_per_second": 17.587,
      "step": 480
    },
    {
      "epoch": 0.4249667994687915,
      "eval_accuracy": 0.8835125448028673,
      "eval_f1": 0.8827759092141159,
      "eval_loss": 0.37165558338165283,
      "eval_runtime": 4.0199,
      "eval_samples_per_second": 138.809,
      "eval_steps_per_second": 17.413,
      "step": 960
    },
    {
      "epoch": 0.6374501992031872,
      "eval_accuracy": 0.8853046594982079,
      "eval_f1": 0.8848186914136107,
      "eval_loss": 0.37070879340171814,
      "eval_runtime": 3.9732,
      "eval_samples_per_second": 140.44,
      "eval_steps_per_second": 17.618,
      "step": 1440
    },
    {
      "epoch": 0.849933598937583,
      "eval_accuracy": 0.8888888888888888,
      "eval_f1": 0.8877338353414772,
      "eval_loss": 0.43106743693351746,
      "eval_runtime": 4.0177,
      "eval_samples_per_second": 138.885,
      "eval_steps_per_second": 17.423,
      "step": 1920
    },
    {
      "epoch": 0.9065958388667552,
      "grad_norm": 0.671415388584137,
      "learning_rate": 1.804100815434448e-05,
      "loss": 0.5386,
      "step": 2048
    },
    {
      "epoch": 1.0624169986719787,
      "eval_accuracy": 0.8548387096774194,
      "eval_f1": 0.8556637971147828,
      "eval_loss": 0.6500999927520752,
      "eval_runtime": 4.0038,
      "eval_samples_per_second": 139.368,
      "eval_steps_per_second": 17.483,
      "step": 2400
    },
    {
      "epoch": 1.2749003984063745,
      "eval_accuracy": 0.8888888888888888,
      "eval_f1": 0.888992318806512,
      "eval_loss": 0.5748390555381775,
      "eval_runtime": 4.0052,
      "eval_samples_per_second": 139.319,
      "eval_steps_per_second": 17.477,
      "step": 2880
    },
    {
      "epoch": 1.4873837981407703,
      "eval_accuracy": 0.8924731182795699,
      "eval_f1": 0.8920627439279448,
      "eval_loss": 0.5230178236961365,
      "eval_runtime": 4.0216,
      "eval_samples_per_second": 138.749,
      "eval_steps_per_second": 17.406,
      "step": 3360
    },
    {
      "epoch": 1.699867197875166,
      "eval_accuracy": 0.8942652329749103,
      "eval_f1": 0.894461310705129,
      "eval_loss": 0.555411159992218,
      "eval_runtime": 3.9433,
      "eval_samples_per_second": 141.506,
      "eval_steps_per_second": 17.752,
      "step": 3840
    },
    {
      "epoch": 1.8131916777335104,
      "grad_norm": 8.14060115814209,
      "learning_rate": 1.1986926694634988e-05,
      "loss": 0.4507,
      "step": 4096
    },
    {
      "epoch": 1.9123505976095618,
      "eval_accuracy": 0.8906810035842294,
      "eval_f1": 0.8906975831458821,
      "eval_loss": 0.5315648317337036,
      "eval_runtime": 3.9793,
      "eval_samples_per_second": 140.227,
      "eval_steps_per_second": 17.591,
      "step": 4320
    },
    {
      "epoch": 2.1248339973439574,
      "eval_accuracy": 0.9032258064516129,
      "eval_f1": 0.9029529506581548,
      "eval_loss": 0.5185472369194031,
      "eval_runtime": 4.0159,
      "eval_samples_per_second": 138.949,
      "eval_steps_per_second": 17.431,
      "step": 4800
    }
  ],
  "logging_steps": 2048,
  "max_steps": 9036,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 480,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9462403342238616.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}