run-d67sabor/checkpoint-100/trainer_state.json

{
  "best_metric": 0.75,
  "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-d67sabor/checkpoint-100",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 2.484299421310425,
      "learning_rate": 1.9694903334952482e-05,
      "loss": 0.6039,
      "step": 10
    },
    {
      "epoch": 0.4,
      "grad_norm": 3.1018054485321045,
      "learning_rate": 1.7506580742179987e-05,
      "loss": 0.5167,
      "step": 20
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.458459377288818,
      "learning_rate": 1.5318258149407485e-05,
      "loss": 0.4812,
      "step": 30
    },
    {
      "epoch": 0.8,
      "grad_norm": 3.5263173580169678,
      "learning_rate": 1.3129935556634988e-05,
      "loss": 0.4425,
      "step": 40
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.9599881172180176,
      "learning_rate": 1.094161296386249e-05,
      "loss": 0.476,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.77,
      "eval_f1": 0.6642335766423357,
      "eval_loss": 0.43119922280311584,
      "eval_precision": 0.6026490066225165,
      "eval_recall": 0.7398373983739838,
      "eval_runtime": 1.5174,
      "eval_samples_per_second": 263.61,
      "eval_steps_per_second": 16.476,
      "step": 50
    },
    {
      "epoch": 1.2,
      "grad_norm": 4.448625564575195,
      "learning_rate": 8.753290371089993e-06,
      "loss": 0.3837,
      "step": 60
    },
    {
      "epoch": 1.4,
      "grad_norm": 1.8515591621398926,
      "learning_rate": 6.564967778317494e-06,
      "loss": 0.3748,
      "step": 70
    },
    {
      "epoch": 1.6,
      "grad_norm": 3.0516843795776367,
      "learning_rate": 4.376645185544997e-06,
      "loss": 0.3408,
      "step": 80
    },
    {
      "epoch": 1.8,
      "grad_norm": 2.1405086517333984,
      "learning_rate": 2.1883225927724984e-06,
      "loss": 0.3669,
      "step": 90
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.383695363998413,
      "learning_rate": 0.0,
      "loss": 0.3174,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.84,
      "eval_f1": 0.75,
      "eval_loss": 0.34978073835372925,
      "eval_precision": 0.7218045112781954,
      "eval_recall": 0.7804878048780488,
      "eval_runtime": 1.5618,
      "eval_samples_per_second": 256.109,
      "eval_steps_per_second": 16.007,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 423630740901888.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "_wandb": {},
    "assignments": {},
    "learning_rate": 2.188322592772498e-05,
    "metric": "eval/loss",
    "num_train_epochs": 2,
    "per_device_train_batch_size": 32,
    "seed": 18
  }
}