trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.84,
  "eval_steps": 25,
  "global_step": 12,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "grad_norm": 1.3747080564498901,
      "learning_rate": 0.00018333333333333334,
      "loss": 9.5405,
      "step": 1
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.4412273168563843,
      "learning_rate": 0.0001666666666666667,
      "loss": 9.4519,
      "step": 2
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.961427092552185,
      "learning_rate": 0.00015000000000000001,
      "loss": 9.1814,
      "step": 3
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.2182157039642334,
      "learning_rate": 0.00013333333333333334,
      "loss": 8.8873,
      "step": 4
    },
    {
      "epoch": 1.6,
      "grad_norm": 2.371889591217041,
      "learning_rate": 0.00011666666666666668,
      "loss": 8.6062,
      "step": 5
    },
    {
      "epoch": 1.92,
      "grad_norm": 1.932769536972046,
      "learning_rate": 0.0001,
      "loss": 8.3597,
      "step": 6
    },
    {
      "epoch": 2.24,
      "grad_norm": 1.71647047996521,
      "learning_rate": 8.333333333333334e-05,
      "loss": 8.2076,
      "step": 7
    },
    {
      "epoch": 2.56,
      "grad_norm": 1.4216786623001099,
      "learning_rate": 6.666666666666667e-05,
      "loss": 8.0357,
      "step": 8
    },
    {
      "epoch": 2.88,
      "grad_norm": 1.3276846408843994,
      "learning_rate": 5e-05,
      "loss": 7.9892,
      "step": 9
    },
    {
      "epoch": 3.2,
      "grad_norm": 1.295423150062561,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 7.9392,
      "step": 10
    },
    {
      "epoch": 3.52,
      "grad_norm": 1.1277278661727905,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 7.8902,
      "step": 11
    },
    {
      "epoch": 3.84,
      "grad_norm": 1.0911381244659424,
      "learning_rate": 0.0,
      "loss": 7.8333,
      "step": 12
    },
    {
      "epoch": 3.84,
      "step": 12,
      "total_flos": 22357590798924.0,
      "train_loss": 8.493526140848795,
      "train_runtime": 47.9357,
      "train_samples_per_second": 4.089,
      "train_steps_per_second": 0.25
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 12,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 22357590798924.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}