trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 7700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.4935064935064934,
      "grad_norm": 29.42378044128418,
      "learning_rate": 4.675324675324675e-05,
      "loss": 1.5516,
      "step": 500
    },
    {
      "epoch": 12.987012987012987,
      "grad_norm": 14.185086250305176,
      "learning_rate": 4.3506493506493503e-05,
      "loss": 0.9416,
      "step": 1000
    },
    {
      "epoch": 19.48051948051948,
      "grad_norm": 3.7115426063537598,
      "learning_rate": 4.025974025974026e-05,
      "loss": 0.5566,
      "step": 1500
    },
    {
      "epoch": 25.974025974025974,
      "grad_norm": 8.48068618774414,
      "learning_rate": 3.701298701298702e-05,
      "loss": 0.3817,
      "step": 2000
    },
    {
      "epoch": 32.467532467532465,
      "grad_norm": 3.017573118209839,
      "learning_rate": 3.376623376623377e-05,
      "loss": 0.2771,
      "step": 2500
    },
    {
      "epoch": 38.96103896103896,
      "grad_norm": 0.5344293117523193,
      "learning_rate": 3.051948051948052e-05,
      "loss": 0.2127,
      "step": 3000
    },
    {
      "epoch": 45.45454545454545,
      "grad_norm": 1.3606529235839844,
      "learning_rate": 2.7272727272727273e-05,
      "loss": 0.19,
      "step": 3500
    },
    {
      "epoch": 51.94805194805195,
      "grad_norm": 2.6488523483276367,
      "learning_rate": 2.4025974025974027e-05,
      "loss": 0.1353,
      "step": 4000
    },
    {
      "epoch": 58.44155844155844,
      "grad_norm": 0.12463207542896271,
      "learning_rate": 2.077922077922078e-05,
      "loss": 0.1148,
      "step": 4500
    },
    {
      "epoch": 64.93506493506493,
      "grad_norm": 0.8130776286125183,
      "learning_rate": 1.7532467532467535e-05,
      "loss": 0.0934,
      "step": 5000
    },
    {
      "epoch": 71.42857142857143,
      "grad_norm": 0.03147471696138382,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.0735,
      "step": 5500
    },
    {
      "epoch": 77.92207792207792,
      "grad_norm": 6.990360260009766,
      "learning_rate": 1.103896103896104e-05,
      "loss": 0.0427,
      "step": 6000
    },
    {
      "epoch": 84.41558441558442,
      "grad_norm": 2.2498698234558105,
      "learning_rate": 7.792207792207792e-06,
      "loss": 0.0373,
      "step": 6500
    },
    {
      "epoch": 90.9090909090909,
      "grad_norm": 0.016162734478712082,
      "learning_rate": 4.5454545454545455e-06,
      "loss": 0.028,
      "step": 7000
    },
    {
      "epoch": 97.40259740259741,
      "grad_norm": 0.5883862972259521,
      "learning_rate": 1.2987012987012988e-06,
      "loss": 0.0206,
      "step": 7500
    },
    {
      "epoch": 100.0,
      "step": 7700,
      "total_flos": 1.1008605548556e+16,
      "train_loss": 0.30270214981846993,
      "train_runtime": 8124.9371,
      "train_samples_per_second": 7.532,
      "train_steps_per_second": 0.948
    }
  ],
  "logging_steps": 500,
  "max_steps": 7700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1008605548556e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}