trainer_state.json

{
  "best_metric": 0.8566413316413316,
  "best_model_checkpoint": "./splash_train_concept_continued/checkpoint-128",
  "epoch": 127.54739652870494,
  "global_step": 128,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 0.0026,
      "step": 1
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0001,
      "loss": 0.0011,
      "step": 4
    },
    {
      "epoch": 7.55,
      "learning_rate": 0.0001,
      "loss": 0.0022,
      "step": 8
    },
    {
      "epoch": 11.55,
      "learning_rate": 0.0001,
      "loss": 0.0018,
      "step": 12
    },
    {
      "epoch": 15.55,
      "learning_rate": 0.0001,
      "loss": 0.0003,
      "step": 16
    },
    {
      "epoch": 19.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 20
    },
    {
      "epoch": 23.55,
      "learning_rate": 0.0001,
      "loss": 0.0003,
      "step": 24
    },
    {
      "epoch": 27.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 28
    },
    {
      "epoch": 31.55,
      "learning_rate": 0.0001,
      "loss": 0.0018,
      "step": 32
    },
    {
      "epoch": 35.55,
      "learning_rate": 0.0001,
      "loss": 0.0022,
      "step": 36
    },
    {
      "epoch": 39.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 40
    },
    {
      "epoch": 43.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 44
    },
    {
      "epoch": 47.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 48
    },
    {
      "epoch": 51.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 52
    },
    {
      "epoch": 55.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 56
    },
    {
      "epoch": 59.55,
      "learning_rate": 0.0001,
      "loss": 0.0051,
      "step": 60
    },
    {
      "epoch": 63.55,
      "learning_rate": 0.0001,
      "loss": 0.0003,
      "step": 64
    },
    {
      "epoch": 63.55,
      "eval_f1_score": 0.8557729057729058,
      "eval_loss": 0.3384537696838379,
      "eval_runtime": 287.8554,
      "eval_samples_per_second": 3.342,
      "step": 64
    },
    {
      "epoch": 67.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 68
    },
    {
      "epoch": 71.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 72
    },
    {
      "epoch": 75.55,
      "learning_rate": 0.0001,
      "loss": 0.0002,
      "step": 76
    },
    {
      "epoch": 79.55,
      "learning_rate": 0.0001,
      "loss": 0.0046,
      "step": 80
    },
    {
      "epoch": 83.55,
      "learning_rate": 0.0001,
      "loss": 0.0007,
      "step": 84
    },
    {
      "epoch": 87.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 88
    },
    {
      "epoch": 91.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 92
    },
    {
      "epoch": 95.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 96
    },
    {
      "epoch": 99.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 100
    },
    {
      "epoch": 103.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 104
    },
    {
      "epoch": 107.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 108
    },
    {
      "epoch": 111.55,
      "learning_rate": 0.0001,
      "loss": 0.0006,
      "step": 112
    },
    {
      "epoch": 115.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 116
    },
    {
      "epoch": 119.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 120
    },
    {
      "epoch": 123.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 124
    },
    {
      "epoch": 127.55,
      "learning_rate": 0.0001,
      "loss": 0.0001,
      "step": 128
    },
    {
      "epoch": 127.55,
      "eval_f1_score": 0.8566413316413316,
      "eval_loss": 0.35904133319854736,
      "eval_runtime": 290.9961,
      "eval_samples_per_second": 3.306,
      "step": 128
    }
  ],
  "max_steps": 3072,
  "num_train_epochs": 3072,
  "total_flos": 1.897685960620032e+18,
  "trial_name": null,
  "trial_params": null
}