trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.955223880597015,
  "eval_steps": 500,
  "global_step": 99,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29850746268656714,
      "grad_norm": 6.538838734396776,
      "learning_rate": 5e-06,
      "loss": 0.7913,
      "step": 10
    },
    {
      "epoch": 0.5970149253731343,
      "grad_norm": 1.3732512938919552,
      "learning_rate": 5e-06,
      "loss": 0.6865,
      "step": 20
    },
    {
      "epoch": 0.8955223880597015,
      "grad_norm": 1.365851577307956,
      "learning_rate": 5e-06,
      "loss": 0.648,
      "step": 30
    },
    {
      "epoch": 0.9850746268656716,
      "eval_loss": 0.6713858842849731,
      "eval_runtime": 24.2836,
      "eval_samples_per_second": 37.186,
      "eval_steps_per_second": 0.618,
      "step": 33
    },
    {
      "epoch": 1.1940298507462686,
      "grad_norm": 11.538898003006828,
      "learning_rate": 5e-06,
      "loss": 0.6545,
      "step": 40
    },
    {
      "epoch": 1.4925373134328357,
      "grad_norm": 1.62138786975469,
      "learning_rate": 5e-06,
      "loss": 0.5982,
      "step": 50
    },
    {
      "epoch": 1.7910447761194028,
      "grad_norm": 1.6675944662711073,
      "learning_rate": 5e-06,
      "loss": 0.5827,
      "step": 60
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6018304228782654,
      "eval_runtime": 23.9682,
      "eval_samples_per_second": 37.675,
      "eval_steps_per_second": 0.626,
      "step": 67
    },
    {
      "epoch": 2.08955223880597,
      "grad_norm": 1.5360982094786524,
      "learning_rate": 5e-06,
      "loss": 0.5874,
      "step": 70
    },
    {
      "epoch": 2.388059701492537,
      "grad_norm": 1.1300056952679531,
      "learning_rate": 5e-06,
      "loss": 0.5198,
      "step": 80
    },
    {
      "epoch": 2.6865671641791042,
      "grad_norm": 1.1741997970276916,
      "learning_rate": 5e-06,
      "loss": 0.5126,
      "step": 90
    },
    {
      "epoch": 2.955223880597015,
      "eval_loss": 0.5868151187896729,
      "eval_runtime": 22.6682,
      "eval_samples_per_second": 39.836,
      "eval_steps_per_second": 0.662,
      "step": 99
    },
    {
      "epoch": 2.955223880597015,
      "step": 99,
      "total_flos": 165619307642880.0,
      "train_loss": 0.6097636222839355,
      "train_runtime": 3516.5371,
      "train_samples_per_second": 14.625,
      "train_steps_per_second": 0.028
    }
  ],
  "logging_steps": 10,
  "max_steps": 99,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 165619307642880.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}