trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.6666666666666665,
  "eval_steps": 500,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.44,
      "grad_norm": 1.4642149544077814,
      "learning_rate": 2e-05,
      "loss": 1.2242,
      "step": 1
    },
    {
      "epoch": 0.89,
      "eval_loss": 1.0858471393585205,
      "eval_runtime": 30.2801,
      "eval_samples_per_second": 11.526,
      "eval_steps_per_second": 0.727,
      "step": 2
    },
    {
      "epoch": 1.78,
      "eval_loss": 1.041835069656372,
      "eval_runtime": 29.9128,
      "eval_samples_per_second": 11.667,
      "eval_steps_per_second": 0.735,
      "step": 4
    },
    {
      "epoch": 2.22,
      "grad_norm": 1.722896643097047,
      "learning_rate": 1.9098300562505266e-06,
      "loss": 1.144,
      "step": 5
    },
    {
      "epoch": 2.67,
      "eval_loss": 1.0281113386154175,
      "eval_runtime": 30.3319,
      "eval_samples_per_second": 11.506,
      "eval_steps_per_second": 0.725,
      "step": 6
    },
    {
      "epoch": 2.67,
      "step": 6,
      "total_flos": 1203933020160.0,
      "train_loss": 1.1458128690719604,
      "train_runtime": 193.3869,
      "train_samples_per_second": 2.172,
      "train_steps_per_second": 0.031
    }
  ],
  "logging_steps": 5,
  "max_steps": 6,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1203933020160.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}