checkpoint-700/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.16106764841233318,
  "eval_steps": 500,
  "global_step": 700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05752416014726185,
      "grad_norm": 0.5827537775039673,
      "learning_rate": 0.00016513000460193283,
      "loss": 5.0544,
      "step": 250
    },
    {
      "epoch": 0.1150483202945237,
      "grad_norm": 0.5847251415252686,
      "learning_rate": 0.00016024045098941557,
      "loss": 4.2362,
      "step": 500
    },
    {
      "epoch": 0.1150483202945237,
      "eval_loss": 3.4322922229766846,
      "eval_runtime": 35.9031,
      "eval_samples_per_second": 109.74,
      "eval_steps_per_second": 3.231,
      "step": 500
    }
  ],
  "logging_steps": 250,
  "max_steps": 8692,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 350,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 537067939430400.0,
  "train_batch_size": 34,
  "trial_name": null,
  "trial_params": null
}