trainer_state.json

{
  "best_metric": 0.8237528800964355,
  "best_model_checkpoint": "./results/run-14f8e466/checkpoint-882",
  "epoch": 4.0,
  "global_step": 3528,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5.592891621465484e-08,
      "loss": 2.1634,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.5537101472531625,
      "eval_loss": 0.8237528800964355,
      "eval_runtime": 9.1428,
      "eval_samples_per_second": 257.253,
      "eval_steps_per_second": 32.157,
      "step": 882
    },
    {
      "epoch": 1.13,
      "learning_rate": 2.3346813109419988e-05,
      "loss": 1.1017,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.6235863967477107,
      "eval_loss": 0.8632814288139343,
      "eval_runtime": 9.1334,
      "eval_samples_per_second": 257.516,
      "eval_steps_per_second": 32.19,
      "step": 1764
    },
    {
      "epoch": 2.27,
      "learning_rate": 1.411152311360512e-05,
      "loss": 0.5733,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.6471466926746001,
      "eval_loss": 1.2097604274749756,
      "eval_runtime": 9.1352,
      "eval_samples_per_second": 257.466,
      "eval_steps_per_second": 32.183,
      "step": 2646
    },
    {
      "epoch": 3.4,
      "learning_rate": 4.87623311779025e-06,
      "loss": 0.2885,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.6374082978762154,
      "eval_loss": 1.457223653793335,
      "eval_runtime": 9.1351,
      "eval_samples_per_second": 257.47,
      "eval_steps_per_second": 32.184,
      "step": 3528
    }
  ],
  "max_steps": 3528,
  "num_train_epochs": 4,
  "total_flos": 5484511364643360.0,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.796445810732742e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 8,
    "warmup_steps": 500,
    "weight_decay": 0.21340361893154028
  }
}