trainer_state.json

{
  "best_metric": 0.15345074236392975,
  "best_model_checkpoint": "checkpoints/checkpoint-36",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 36,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "embedding_loss": 0.4647,
      "epoch": 0.1111111111111111,
      "grad_norm": 3.230494976043701,
      "learning_rate": 5e-06,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_embedding_loss": 0.17025126516819,
      "eval_embedding_runtime": 0.1505,
      "eval_embedding_samples_per_second": 398.721,
      "eval_embedding_steps_per_second": 26.581,
      "step": 9
    },
    {
      "epoch": 2.0,
      "eval_embedding_loss": 0.16353432834148407,
      "eval_embedding_runtime": 0.1844,
      "eval_embedding_samples_per_second": 325.435,
      "eval_embedding_steps_per_second": 21.696,
      "step": 18
    },
    {
      "epoch": 3.0,
      "eval_embedding_loss": 0.15655823051929474,
      "eval_embedding_runtime": 0.1458,
      "eval_embedding_samples_per_second": 411.565,
      "eval_embedding_steps_per_second": 27.438,
      "step": 27
    },
    {
      "epoch": 4.0,
      "eval_embedding_loss": 0.15345074236392975,
      "eval_embedding_runtime": 0.1356,
      "eval_embedding_samples_per_second": 442.32,
      "eval_embedding_steps_per_second": 29.488,
      "step": 36
    }
  ],
  "logging_steps": 50,
  "max_steps": 36,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}