trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 79,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.128,
      "grad_norm": 2.646609306335449,
      "learning_rate": 1.9990212265199738e-05,
      "loss": 0.5211,
      "step": 10
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.7899451851844788,
      "learning_rate": 1.883869132745561e-05,
      "loss": 0.1484,
      "step": 20
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.5581858158111572,
      "learning_rate": 1.5984723141740578e-05,
      "loss": 0.0976,
      "step": 30
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.5447527170181274,
      "learning_rate": 1.1978019209855174e-05,
      "loss": 0.0784,
      "step": 40
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3571467101573944,
      "learning_rate": 7.590322975433857e-06,
      "loss": 0.0673,
      "step": 50
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.2840781807899475,
      "learning_rate": 3.6667619695195287e-06,
      "loss": 0.0615,
      "step": 60
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.2745964825153351,
      "learning_rate": 9.630652236279626e-07,
      "loss": 0.0588,
      "step": 70
    },
    {
      "epoch": 1.0,
      "step": 79,
      "total_flos": 1.6947342446323302e+17,
      "train_loss": 0.13819692557371116,
      "train_runtime": 1992.1207,
      "train_samples_per_second": 10.04,
      "train_steps_per_second": 0.04
    }
  ],
  "logging_steps": 10,
  "max_steps": 79,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.6947342446323302e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}