run-q6uzrht1/checkpoint-75/trainer_state.json

{
  "best_metric": 0.8031496062992126,
  "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-q6uzrht1/checkpoint-75",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 75,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4,
      "grad_norm": 5.688959121704102,
      "learning_rate": 7.511886794780784e-05,
      "loss": 0.5399,
      "step": 10
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.843003273010254,
      "learning_rate": 6.356211903276047e-05,
      "loss": 0.4557,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.875,
      "eval_f1": 0.7916666666666666,
      "eval_loss": 0.3184812068939209,
      "eval_precision": 0.811965811965812,
      "eval_recall": 0.7723577235772358,
      "eval_runtime": 1.527,
      "eval_samples_per_second": 261.959,
      "eval_steps_per_second": 16.372,
      "step": 25
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.2531931400299072,
      "learning_rate": 5.200537011771312e-05,
      "loss": 0.3137,
      "step": 30
    },
    {
      "epoch": 1.6,
      "grad_norm": 4.9807209968566895,
      "learning_rate": 4.044862120266576e-05,
      "loss": 0.288,
      "step": 40
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.3428285121917725,
      "learning_rate": 2.8891872287618396e-05,
      "loss": 0.2689,
      "step": 50
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.87,
      "eval_f1": 0.7739130434782608,
      "eval_loss": 0.30559664964675903,
      "eval_precision": 0.8317757009345794,
      "eval_recall": 0.7235772357723578,
      "eval_runtime": 1.5419,
      "eval_samples_per_second": 259.427,
      "eval_steps_per_second": 16.214,
      "step": 50
    },
    {
      "epoch": 2.4,
      "grad_norm": 3.228431463241577,
      "learning_rate": 1.733512337257104e-05,
      "loss": 0.2051,
      "step": 60
    },
    {
      "epoch": 2.8,
      "grad_norm": 3.0721919536590576,
      "learning_rate": 5.778374457523679e-06,
      "loss": 0.1983,
      "step": 70
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.875,
      "eval_f1": 0.8031496062992126,
      "eval_loss": 0.3154485821723938,
      "eval_precision": 0.7786259541984732,
      "eval_recall": 0.8292682926829268,
      "eval_runtime": 1.5916,
      "eval_samples_per_second": 251.317,
      "eval_steps_per_second": 15.707,
      "step": 75
    }
  ],
  "logging_steps": 10,
  "max_steps": 75,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 593189011181568.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "_wandb": {},
    "assignments": {},
    "learning_rate": 8.66756168628552e-05,
    "metric": "eval/loss",
    "num_train_epochs": 3,
    "per_device_train_batch_size": 64,
    "seed": 20
  }
}