trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.248730964467005,
  "eval_steps": 500,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.65,
      "eval_loss": 1.0991822481155396,
      "eval_runtime": 73.7864,
      "eval_samples_per_second": 2.67,
      "eval_steps_per_second": 2.67,
      "step": 1
    },
    {
      "epoch": 1.95,
      "eval_loss": 1.0894932746887207,
      "eval_runtime": 73.7905,
      "eval_samples_per_second": 2.67,
      "eval_steps_per_second": 2.67,
      "step": 3
    },
    {
      "epoch": 2.6,
      "eval_loss": 1.087380290031433,
      "eval_runtime": 73.7864,
      "eval_samples_per_second": 2.67,
      "eval_steps_per_second": 2.67,
      "step": 4
    },
    {
      "epoch": 3.25,
      "grad_norm": 0.20658676326274872,
      "learning_rate": 0.0,
      "loss": 1.092,
      "step": 5
    },
    {
      "epoch": 3.25,
      "eval_loss": 1.0868003368377686,
      "eval_runtime": 73.782,
      "eval_samples_per_second": 2.67,
      "eval_steps_per_second": 2.67,
      "step": 5
    },
    {
      "epoch": 3.25,
      "step": 5,
      "total_flos": 5.634931594100736e+16,
      "train_loss": 1.0919825553894043,
      "train_runtime": 964.1278,
      "train_samples_per_second": 1.022,
      "train_steps_per_second": 0.005
    }
  ],
  "logging_steps": 5,
  "max_steps": 5,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 5.634931594100736e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}