trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 16,
  "global_step": 192,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015625,
      "grad_norm": 0.1416465789079666,
      "learning_rate": 0.0005,
      "loss": 0.6936,
      "step": 1
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.48029133677482605,
      "learning_rate": 0.0005,
      "loss": 0.6521,
      "step": 16
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.5866118669509888,
      "eval_runtime": 211.247,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 16
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.26290011405944824,
      "learning_rate": 0.0005,
      "loss": 0.5821,
      "step": 32
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.5616697072982788,
      "eval_runtime": 211.0712,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 32
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.2908053696155548,
      "learning_rate": 0.0005,
      "loss": 0.6006,
      "step": 48
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.5537130236625671,
      "eval_runtime": 211.1601,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 48
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.2542905807495117,
      "learning_rate": 0.0005,
      "loss": 0.5708,
      "step": 64
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5481152534484863,
      "eval_runtime": 211.1765,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 64
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.44674304127693176,
      "learning_rate": 0.0005,
      "loss": 0.467,
      "step": 80
    },
    {
      "epoch": 1.25,
      "eval_loss": 0.5612103343009949,
      "eval_runtime": 211.1862,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 80
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.36239078640937805,
      "learning_rate": 0.0005,
      "loss": 0.4704,
      "step": 96
    },
    {
      "epoch": 1.5,
      "eval_loss": 0.564439594745636,
      "eval_runtime": 211.1729,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 96
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.4066231846809387,
      "learning_rate": 0.0005,
      "loss": 0.466,
      "step": 112
    },
    {
      "epoch": 1.75,
      "eval_loss": 0.5655233263969421,
      "eval_runtime": 211.1865,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 112
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3407813310623169,
      "learning_rate": 0.0005,
      "loss": 0.4697,
      "step": 128
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5586457252502441,
      "eval_runtime": 211.1864,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 128
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.3170074224472046,
      "learning_rate": 0.0005,
      "loss": 0.3078,
      "step": 144
    },
    {
      "epoch": 2.25,
      "eval_loss": 0.6427608132362366,
      "eval_runtime": 211.1241,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 144
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.4563491940498352,
      "learning_rate": 0.0005,
      "loss": 0.3243,
      "step": 160
    },
    {
      "epoch": 2.5,
      "eval_loss": 0.6296973824501038,
      "eval_runtime": 213.2049,
      "eval_samples_per_second": 0.3,
      "eval_steps_per_second": 0.3,
      "step": 160
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.4941216707229614,
      "learning_rate": 0.0005,
      "loss": 0.3473,
      "step": 176
    },
    {
      "epoch": 2.75,
      "eval_loss": 0.6032494306564331,
      "eval_runtime": 211.1238,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 176
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.3374325931072235,
      "learning_rate": 0.0005,
      "loss": 0.3338,
      "step": 192
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.608513355255127,
      "eval_runtime": 211.2654,
      "eval_samples_per_second": 0.303,
      "eval_steps_per_second": 0.303,
      "step": 192
    }
  ],
  "logging_steps": 16,
  "max_steps": 192,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 16,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.019590530565243e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}