run-8g9rtuui/checkpoint-75/trainer_state.json

{
  "best_metric": 0.8062015503875969,
  "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-8g9rtuui/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": 0.947352409362793,
      "learning_rate": 6.387467631914562e-05,
      "loss": 0.5429,
      "step": 10
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.6765072345733643,
      "learning_rate": 5.83203566392199e-05,
      "loss": 0.4161,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8775,
      "eval_f1": 0.7983539094650206,
      "eval_loss": 0.3228859305381775,
      "eval_precision": 0.8083333333333333,
      "eval_recall": 0.7886178861788617,
      "eval_runtime": 1.5306,
      "eval_samples_per_second": 261.327,
      "eval_steps_per_second": 16.333,
      "step": 25
    },
    {
      "epoch": 1.2,
      "grad_norm": 5.193092346191406,
      "learning_rate": 5.27660369592942e-05,
      "loss": 0.2874,
      "step": 30
    },
    {
      "epoch": 1.6,
      "grad_norm": 4.044554710388184,
      "learning_rate": 4.72117172793685e-05,
      "loss": 0.2978,
      "step": 40
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.186301231384277,
      "learning_rate": 4.165739759944279e-05,
      "loss": 0.2679,
      "step": 50
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.87,
      "eval_f1": 0.7851239669421488,
      "eval_loss": 0.3004050850868225,
      "eval_precision": 0.7983193277310925,
      "eval_recall": 0.7723577235772358,
      "eval_runtime": 1.5237,
      "eval_samples_per_second": 262.521,
      "eval_steps_per_second": 16.408,
      "step": 50
    },
    {
      "epoch": 2.4,
      "grad_norm": 3.6902565956115723,
      "learning_rate": 3.610307791951709e-05,
      "loss": 0.1463,
      "step": 60
    },
    {
      "epoch": 2.8,
      "grad_norm": 5.377773284912109,
      "learning_rate": 3.054875823959138e-05,
      "loss": 0.2031,
      "step": 70
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.875,
      "eval_f1": 0.8062015503875969,
      "eval_loss": 0.3549833595752716,
      "eval_precision": 0.7703703703703704,
      "eval_recall": 0.8455284552845529,
      "eval_runtime": 1.5104,
      "eval_samples_per_second": 264.838,
      "eval_steps_per_second": 16.552,
      "step": 75
    }
  ],
  "logging_steps": 10,
  "max_steps": 125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 593189011181568.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "_wandb": {},
    "assignments": {},
    "learning_rate": 6.942899599907132e-05,
    "metric": "eval/loss",
    "num_train_epochs": 5,
    "per_device_train_batch_size": 64,
    "seed": 6
  }
}