trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 16,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 5.564905166625977,
      "learning_rate": 0.0002,
      "loss": 38.3533,
      "step": 1
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 6.65146541595459,
      "learning_rate": 0.0001875,
      "loss": 36.8913,
      "step": 2
    },
    {
      "epoch": 0.4,
      "grad_norm": 8.893917083740234,
      "learning_rate": 0.000175,
      "loss": 35.9052,
      "step": 3
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 8.743362426757812,
      "learning_rate": 0.00016250000000000002,
      "loss": 35.0392,
      "step": 4
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": NaN,
      "learning_rate": 0.00015000000000000001,
      "loss": 33.1986,
      "step": 5
    },
    {
      "epoch": 0.8,
      "grad_norm": 8.269606590270996,
      "learning_rate": 0.00015000000000000001,
      "loss": 33.7741,
      "step": 6
    },
    {
      "epoch": 0.9333333333333333,
      "grad_norm": 8.749699592590332,
      "learning_rate": 0.0001375,
      "loss": 32.6172,
      "step": 7
    },
    {
      "epoch": 1.0,
      "grad_norm": 5.006312847137451,
      "learning_rate": 0.000125,
      "loss": 16.351,
      "step": 8
    },
    {
      "epoch": 1.1333333333333333,
      "grad_norm": 10.173412322998047,
      "learning_rate": 0.00011250000000000001,
      "loss": 30.0547,
      "step": 9
    },
    {
      "epoch": 1.2666666666666666,
      "grad_norm": 8.951446533203125,
      "learning_rate": 0.0001,
      "loss": 30.6191,
      "step": 10
    },
    {
      "epoch": 1.4,
      "grad_norm": 9.113348960876465,
      "learning_rate": 8.75e-05,
      "loss": 29.3415,
      "step": 11
    },
    {
      "epoch": 1.5333333333333332,
      "grad_norm": 8.937090873718262,
      "learning_rate": 7.500000000000001e-05,
      "loss": 29.2385,
      "step": 12
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 8.111980438232422,
      "learning_rate": 6.25e-05,
      "loss": 28.6866,
      "step": 13
    },
    {
      "epoch": 1.8,
      "grad_norm": 8.194887161254883,
      "learning_rate": 5e-05,
      "loss": 28.2205,
      "step": 14
    },
    {
      "epoch": 1.9333333333333333,
      "grad_norm": 10.803622245788574,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 29.6655,
      "step": 15
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.887396335601807,
      "learning_rate": 2.5e-05,
      "loss": 14.1792,
      "step": 16
    },
    {
      "epoch": 2.0,
      "step": 16,
      "total_flos": 29853631825608.0,
      "train_loss": 30.133461475372314,
      "train_runtime": 397.1796,
      "train_samples_per_second": 0.599,
      "train_steps_per_second": 0.04
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 16,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 29853631825608.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}