trainer_state.json

{
  "best_metric": 0.11723319441080093,
  "best_model_checkpoint": "./trained_models/bertweet-emotion-base/checkpoint-500",
  "epoch": 6.0,
  "global_step": 1200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "eval_accuracy": 0.849,
      "eval_loss": 0.4884544014930725,
      "eval_runtime": 3.0562,
      "eval_samples_per_second": 654.404,
      "eval_steps_per_second": 8.18,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9195,
      "eval_loss": 0.2541632056236267,
      "eval_runtime": 3.0641,
      "eval_samples_per_second": 652.729,
      "eval_steps_per_second": 8.159,
      "step": 200
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.9305,
      "eval_loss": 0.17924615740776062,
      "eval_runtime": 3.0615,
      "eval_samples_per_second": 653.282,
      "eval_steps_per_second": 8.166,
      "step": 300
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.94,
      "eval_loss": 0.12382319569587708,
      "eval_runtime": 3.0806,
      "eval_samples_per_second": 649.223,
      "eval_steps_per_second": 8.115,
      "step": 400
    },
    {
      "epoch": 2.5,
      "learning_rate": 2.916666666666667e-05,
      "loss": 0.3822,
      "step": 500
    },
    {
      "epoch": 2.5,
      "eval_accuracy": 0.945,
      "eval_loss": 0.11723319441080093,
      "eval_runtime": 3.0736,
      "eval_samples_per_second": 650.705,
      "eval_steps_per_second": 8.134,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9455,
      "eval_loss": 0.11559800058603287,
      "eval_runtime": 3.0729,
      "eval_samples_per_second": 650.855,
      "eval_steps_per_second": 8.136,
      "step": 600
    },
    {
      "epoch": 3.5,
      "eval_accuracy": 0.942,
      "eval_loss": 0.11905854940414429,
      "eval_runtime": 3.0704,
      "eval_samples_per_second": 651.375,
      "eval_steps_per_second": 8.142,
      "step": 700
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9485,
      "eval_loss": 0.1258220076560974,
      "eval_runtime": 3.081,
      "eval_samples_per_second": 649.144,
      "eval_steps_per_second": 8.114,
      "step": 800
    },
    {
      "epoch": 4.5,
      "eval_accuracy": 0.9375,
      "eval_loss": 0.14012224972248077,
      "eval_runtime": 3.1002,
      "eval_samples_per_second": 645.114,
      "eval_steps_per_second": 8.064,
      "step": 900
    },
    {
      "epoch": 5.0,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0932,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.94,
      "eval_loss": 0.13431893289089203,
      "eval_runtime": 3.0988,
      "eval_samples_per_second": 645.402,
      "eval_steps_per_second": 8.068,
      "step": 1000
    },
    {
      "epoch": 5.5,
      "eval_accuracy": 0.944,
      "eval_loss": 0.14152021706104279,
      "eval_runtime": 3.0916,
      "eval_samples_per_second": 646.924,
      "eval_steps_per_second": 8.087,
      "step": 1100
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9435,
      "eval_loss": 0.13345251977443695,
      "eval_runtime": 3.0757,
      "eval_samples_per_second": 650.25,
      "eval_steps_per_second": 8.128,
      "step": 1200
    },
    {
      "epoch": 6.0,
      "step": 1200,
      "total_flos": 6906913251840000.0,
      "train_loss": 0.20971180438995363,
      "train_runtime": 446.3991,
      "train_samples_per_second": 215.054,
      "train_steps_per_second": 2.688
    }
  ],
  "max_steps": 1200,
  "num_train_epochs": 6,
  "total_flos": 6906913251840000.0,
  "trial_name": null,
  "trial_params": null
}