run-4r5wbvc3/checkpoint-50/trainer_state.json

{
  "best_metric": 0.582995951417004,
  "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-4r5wbvc3/checkpoint-50",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 1.2984086275100708,
      "learning_rate": 2.8254267437810627e-05,
      "loss": 0.6014,
      "step": 10
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.4973173141479492,
      "learning_rate": 2.119070057835797e-05,
      "loss": 0.5197,
      "step": 20
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.660354733467102,
      "learning_rate": 1.4127133718905313e-05,
      "loss": 0.4239,
      "step": 30
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.9838685989379883,
      "learning_rate": 7.063566859452657e-06,
      "loss": 0.4782,
      "step": 40
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.0562429428100586,
      "learning_rate": 0.0,
      "loss": 0.4456,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7425,
      "eval_f1": 0.582995951417004,
      "eval_loss": 0.4402432143688202,
      "eval_precision": 0.5806451612903226,
      "eval_recall": 0.5853658536585366,
      "eval_runtime": 1.5859,
      "eval_samples_per_second": 252.225,
      "eval_steps_per_second": 15.764,
      "step": 50
    }
  ],
  "logging_steps": 10,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 211815370450944.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "_wandb": {},
    "assignments": {},
    "learning_rate": 3.531783429726328e-05,
    "metric": "eval/loss",
    "num_train_epochs": 1,
    "per_device_train_batch_size": 32,
    "seed": 5
  }
}