run-4/checkpoint-1000/trainer_state.json

{
  "best_global_step": 1000,
  "best_metric": 0.7331392486819855,
  "best_model_checkpoint": "./results/run-4/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": 3.002380609512329,
      "learning_rate": 4.572943108695439e-06,
      "loss": 0.296,
      "step": 100
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9367781281471252,
      "learning_rate": 9.192077561923155e-06,
      "loss": 1.0274,
      "step": 200
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.7274576425552368,
      "learning_rate": 1.3811212015150872e-05,
      "loss": 0.0,
      "step": 300
    },
    {
      "epoch": 0.4,
      "grad_norm": 44.297061920166016,
      "learning_rate": 1.8430346468378586e-05,
      "loss": 0.0,
      "step": 400
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.0151411294937134,
      "learning_rate": 2.3049480921606306e-05,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.208032488822937,
      "learning_rate": 2.766861537483402e-05,
      "loss": 0.0,
      "step": 600
    },
    {
      "epoch": 0.7,
      "grad_norm": 6.984821319580078,
      "learning_rate": 3.2287749828061733e-05,
      "loss": 0.0342,
      "step": 700
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.5860588550567627,
      "learning_rate": 3.690688428128945e-05,
      "loss": 0.0488,
      "step": 800
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.7841507196426392,
      "learning_rate": 4.152601873451717e-05,
      "loss": 0.0314,
      "step": 900
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9344207644462585,
      "learning_rate": 4.414230747586468e-05,
      "loss": 0.0356,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_economic_inequality_accuracy": 0.7783375314861462,
      "eval_economic_inequality_f1": 0.6829252350399464,
      "eval_economic_policy_benefits_accuracy": 0.8215488215488216,
      "eval_economic_policy_benefits_f1": 0.752833443442648,
      "eval_ethnic_boundaries_accuracy": 0.9493464052287581,
      "eval_ethnic_boundaries_f1": 0.9305506350938352,
      "eval_language_policy_accuracy": 0.6764705882352942,
      "eval_language_policy_f1": 0.555672268907563,
      "eval_loss": NaN,
      "eval_mother_tongue_education_accuracy": 0.7777777777777778,
      "eval_mother_tongue_education_f1": 0.6805555555555556,
      "eval_overall_accuracy": 0.8090908256773385,
      "eval_overall_f1": 0.7331392486819855,
      "eval_religion_ethnic_identity_accuracy": 0.851063829787234,
      "eval_religion_ethnic_identity_f1": 0.7962983540523647,
      "eval_runtime": 4.4716,
      "eval_samples_per_second": 894.535,
      "eval_steps_per_second": 55.908,
      "step": 1000
    }
  ],
  "logging_steps": 100,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 16,
  "trial_name": null,
  "trial_params": {
    "gradient_accumulation_steps": 1,
    "learning_rate": 4.452845612911518e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 16,
    "warmup_steps": 964,
    "weight_decay": 0.08427803271887896
  }
}