state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.2901353965183752,
  "eval_steps": 50,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09671179883945841,
      "grad_norm": 0.035103704780340195,
      "learning_rate": 0.00016025641025641028,
      "loss": 1.0483,
      "step": 50
    },
    {
      "epoch": 0.09671179883945841,
      "eval_loss": 0.9260319471359253,
      "eval_runtime": 60.9012,
      "eval_samples_per_second": 1.642,
      "eval_steps_per_second": 0.213,
      "step": 50
    },
    {
      "epoch": 0.19342359767891681,
      "grad_norm": 0.030684856697916985,
      "learning_rate": 0.00032051282051282057,
      "loss": 0.8577,
      "step": 100
    },
    {
      "epoch": 0.19342359767891681,
      "eval_loss": 0.820220410823822,
      "eval_runtime": 56.5241,
      "eval_samples_per_second": 1.769,
      "eval_steps_per_second": 0.23,
      "step": 100
    }
  ],
  "logging_steps": 50,
  "max_steps": 1551,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.308517765742592e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}