trainer_state.json

{
  "best_metric": 0.936571294401771,
  "best_model_checkpoint": "model_saves/deberta-v3-large_spell_5k_2_p3/checkpoint-536",
  "epoch": 5.0,
  "global_step": 1340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9365489327921824,
      "eval_loss": 0.47526291012763977,
      "eval_runtime": 4.5624,
      "eval_samples_per_second": 960.453,
      "eval_steps_per_second": 7.671,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.936571294401771,
      "eval_loss": 0.4809601604938507,
      "eval_runtime": 4.5593,
      "eval_samples_per_second": 961.109,
      "eval_steps_per_second": 7.677,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9361072910028064,
      "eval_loss": 0.49737367033958435,
      "eval_runtime": 4.5689,
      "eval_samples_per_second": 959.09,
      "eval_steps_per_second": 7.66,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.3572,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9352519594360402,
      "eval_loss": 0.52593594789505,
      "eval_runtime": 4.5613,
      "eval_samples_per_second": 960.701,
      "eval_steps_per_second": 7.673,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9338823108487349,
      "eval_loss": 0.5405715703964233,
      "eval_runtime": 4.5746,
      "eval_samples_per_second": 957.899,
      "eval_steps_per_second": 7.651,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.6964926941691904e+16,
      "train_loss": 0.332953547008002,
      "train_runtime": 581.5265,
      "train_samples_per_second": 884.844,
      "train_steps_per_second": 6.913
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.6964926941691904e+16,
  "trial_name": null,
  "trial_params": null
}