run-2/checkpoint-1000/trainer_state.json

{
  "best_global_step": 1000,
  "best_metric": 0.7839408748610676,
  "best_model_checkpoint": "./results/run-2/checkpoint-1000",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 3.6001648902893066,
      "learning_rate": 3.1887409174919234e-06,
      "loss": 0.0412,
      "step": 100
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6239789128303528,
      "learning_rate": 6.409691339200936e-06,
      "loss": 0.1091,
      "step": 200
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6479267477989197,
      "learning_rate": 9.63064176090995e-06,
      "loss": 0.0,
      "step": 300
    },
    {
      "epoch": 0.4,
      "grad_norm": 20.527082443237305,
      "learning_rate": 1.054959200884952e-05,
      "loss": 0.0,
      "step": 400
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.8853731751441956,
      "learning_rate": 1.0481970753745042e-05,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.839735507965088,
      "learning_rate": 1.0414349498640566e-05,
      "loss": 0.0,
      "step": 600
    },
    {
      "epoch": 0.7,
      "grad_norm": 2.7137391567230225,
      "learning_rate": 1.0346728243536088e-05,
      "loss": 0.0284,
      "step": 700
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.6026535630226135,
      "learning_rate": 1.027910698843161e-05,
      "loss": 0.1612,
      "step": 800
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5444966554641724,
      "learning_rate": 1.0211485733327134e-05,
      "loss": 0.0,
      "step": 900
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.2413469552993774,
      "learning_rate": 1.0143864478222655e-05,
      "loss": 0.0986,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_economic_inequality_accuracy": 0.8047858942065491,
      "eval_economic_inequality_f1": 0.7538941327354425,
      "eval_economic_policy_benefits_accuracy": 0.8282828282828283,
      "eval_economic_policy_benefits_f1": 0.764129475069552,
      "eval_ethnic_boundaries_accuracy": 0.9493464052287581,
      "eval_ethnic_boundaries_f1": 0.9281293325410972,
      "eval_language_policy_accuracy": 0.7058823529411765,
      "eval_language_policy_f1": 0.6071428571428571,
      "eval_loss": NaN,
      "eval_mother_tongue_education_accuracy": 0.8888888888888888,
      "eval_mother_tongue_education_f1": 0.837037037037037,
      "eval_overall_accuracy": 0.8390088178776338,
      "eval_overall_f1": 0.7839408748610676,
      "eval_religion_ethnic_identity_accuracy": 0.8568665377176016,
      "eval_religion_ethnic_identity_f1": 0.8133124146404199,
      "eval_runtime": 4.4593,
      "eval_samples_per_second": 897.004,
      "eval_steps_per_second": 56.063,
      "step": 1000
    }
  ],
  "logging_steps": 100,
  "max_steps": 16000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 16,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4210381651968000.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "gradient_accumulation_steps": 2,
    "learning_rate": 1.0596926887422654e-05,
    "num_train_epochs": 16,
    "per_device_train_batch_size": 8,
    "warmup_steps": 329,
    "weight_decay": 0.059185379025811405
  }
}