trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 1000.0,
  "global_step": 7494,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.200160128102482,
      "grad_norm": 6.359441757202148,
      "learning_rate": 4.6663997864958635e-05,
      "loss": 6.6546,
      "step": 500
    },
    {
      "epoch": 0.400320256204964,
      "grad_norm": 5.609431743621826,
      "learning_rate": 4.332799572991727e-05,
      "loss": 6.1195,
      "step": 1000
    },
    {
      "epoch": 0.600480384307446,
      "grad_norm": 8.655993461608887,
      "learning_rate": 3.99919935948759e-05,
      "loss": 5.8641,
      "step": 1500
    },
    {
      "epoch": 0.800640512409928,
      "grad_norm": 5.4739837646484375,
      "learning_rate": 3.665599145983454e-05,
      "loss": 5.7106,
      "step": 2000
    },
    {
      "epoch": 1.00080064051241,
      "grad_norm": 5.922306060791016,
      "learning_rate": 3.3319989324793174e-05,
      "loss": 5.5891,
      "step": 2500
    },
    {
      "epoch": 1.200960768614892,
      "grad_norm": 7.8872504234313965,
      "learning_rate": 2.9983987189751807e-05,
      "loss": 5.1944,
      "step": 3000
    },
    {
      "epoch": 1.4011208967173738,
      "grad_norm": 5.556982040405273,
      "learning_rate": 2.6647985054710433e-05,
      "loss": 5.1804,
      "step": 3500
    },
    {
      "epoch": 1.601281024819856,
      "grad_norm": 6.548351287841797,
      "learning_rate": 2.331198291966907e-05,
      "loss": 5.1859,
      "step": 4000
    },
    {
      "epoch": 1.801441152922338,
      "grad_norm": 7.128460884094238,
      "learning_rate": 1.9975980784627705e-05,
      "loss": 5.0938,
      "step": 4500
    },
    {
      "epoch": 2.00160128102482,
      "grad_norm": 5.632880210876465,
      "learning_rate": 1.6639978649586335e-05,
      "loss": 5.0563,
      "step": 5000
    },
    {
      "epoch": 2.2017614091273017,
      "grad_norm": 6.655033588409424,
      "learning_rate": 1.330397651454497e-05,
      "loss": 4.68,
      "step": 5500
    },
    {
      "epoch": 2.401921537229784,
      "grad_norm": 8.128520965576172,
      "learning_rate": 9.967974379503602e-06,
      "loss": 4.7047,
      "step": 6000
    },
    {
      "epoch": 2.602081665332266,
      "grad_norm": 8.819676399230957,
      "learning_rate": 6.631972244462237e-06,
      "loss": 4.6252,
      "step": 6500
    },
    {
      "epoch": 2.8022417934347477,
      "grad_norm": 6.872320175170898,
      "learning_rate": 3.29597010942087e-06,
      "loss": 4.656,
      "step": 7000
    },
    {
      "epoch": 3.0,
      "step": 7494,
      "total_flos": 3.246076355120333e+16,
      "train_loss": 5.265000151162597,
      "train_runtime": 4607.9912,
      "train_samples_per_second": 3.252,
      "train_steps_per_second": 1.626
    }
  ],
  "logging_steps": 500,
  "max_steps": 7494,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.246076355120333e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}