method2/final_model_deberta_weighted/checkpoint-1250/trainer_state.json

{
  "best_global_step": 1250,
  "best_metric": 0.39082855653986454,
  "best_model_checkpoint": "./final_model_deberta_weighted/checkpoint-1250",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "grad_norm": 2.7260196208953857,
      "learning_rate": 2.6329787234042553e-05,
      "loss": 0.6738,
      "step": 100
    },
    {
      "epoch": 0.32,
      "grad_norm": 2.201608657836914,
      "learning_rate": 4.967397747480736e-05,
      "loss": 0.6427,
      "step": 200
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.4532300233840942,
      "learning_rate": 4.6710136336692354e-05,
      "loss": 0.6344,
      "step": 300
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.2946869134902954,
      "learning_rate": 4.374629519857736e-05,
      "loss": 0.6146,
      "step": 400
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.2154576778411865,
      "learning_rate": 4.078245406046236e-05,
      "loss": 0.6375,
      "step": 500
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.464404344558716,
      "learning_rate": 3.781861292234736e-05,
      "loss": 0.6044,
      "step": 600
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7808,
      "eval_f1_fake": 0.3627906976744186,
      "eval_loss": 0.6771467924118042,
      "eval_precision_fake": 0.3035019455252918,
      "eval_recall_fake": 0.4508670520231214,
      "eval_runtime": 7.9328,
      "eval_samples_per_second": 630.297,
      "eval_steps_per_second": 5.042,
      "step": 625
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.8839420676231384,
      "learning_rate": 3.4854771784232364e-05,
      "loss": 0.5874,
      "step": 700
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.285341501235962,
      "learning_rate": 3.189093064611737e-05,
      "loss": 0.5764,
      "step": 800
    },
    {
      "epoch": 1.44,
      "grad_norm": 1.3300535678863525,
      "learning_rate": 2.8927089508002375e-05,
      "loss": 0.5671,
      "step": 900
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.5840404033660889,
      "learning_rate": 2.5963248369887376e-05,
      "loss": 0.562,
      "step": 1000
    },
    {
      "epoch": 1.76,
      "grad_norm": 1.9565057754516602,
      "learning_rate": 2.2999407231772376e-05,
      "loss": 0.5649,
      "step": 1100
    },
    {
      "epoch": 1.92,
      "grad_norm": 1.3821005821228027,
      "learning_rate": 2.003556609365738e-05,
      "loss": 0.5596,
      "step": 1200
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7662,
      "eval_f1_fake": 0.39082855653986454,
      "eval_loss": 0.6207814812660217,
      "eval_precision_fake": 0.3056234718826406,
      "eval_recall_fake": 0.541907514450867,
      "eval_runtime": 7.9513,
      "eval_samples_per_second": 628.832,
      "eval_steps_per_second": 5.031,
      "step": 1250
    }
  ],
  "logging_steps": 100,
  "max_steps": 1875,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1049262196391936e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}