last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.019102196752626553,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0019102196752626551,
      "grad_norm": 1.206204891204834,
      "learning_rate": 2e-05,
      "loss": 2.9644,
      "step": 1
    },
    {
      "epoch": 0.0019102196752626551,
      "eval_loss": 3.0215418338775635,
      "eval_runtime": 13.9127,
      "eval_samples_per_second": 15.885,
      "eval_steps_per_second": 7.978,
      "step": 1
    },
    {
      "epoch": 0.0038204393505253103,
      "grad_norm": 1.066440463066101,
      "learning_rate": 4e-05,
      "loss": 3.1595,
      "step": 2
    },
    {
      "epoch": 0.0057306590257879654,
      "grad_norm": 1.1520076990127563,
      "learning_rate": 6e-05,
      "loss": 3.0597,
      "step": 3
    },
    {
      "epoch": 0.0057306590257879654,
      "eval_loss": 3.008910655975342,
      "eval_runtime": 12.8616,
      "eval_samples_per_second": 17.183,
      "eval_steps_per_second": 8.63,
      "step": 3
    },
    {
      "epoch": 0.007640878701050621,
      "grad_norm": 1.1618703603744507,
      "learning_rate": 8e-05,
      "loss": 3.0982,
      "step": 4
    },
    {
      "epoch": 0.009551098376313277,
      "grad_norm": 1.2491005659103394,
      "learning_rate": 0.0001,
      "loss": 2.8439,
      "step": 5
    },
    {
      "epoch": 0.011461318051575931,
      "grad_norm": 1.0564581155776978,
      "learning_rate": 0.00012,
      "loss": 3.1059,
      "step": 6
    },
    {
      "epoch": 0.011461318051575931,
      "eval_loss": 2.8710217475891113,
      "eval_runtime": 12.8655,
      "eval_samples_per_second": 17.178,
      "eval_steps_per_second": 8.628,
      "step": 6
    },
    {
      "epoch": 0.013371537726838587,
      "grad_norm": 0.9323328137397766,
      "learning_rate": 0.00014,
      "loss": 2.7436,
      "step": 7
    },
    {
      "epoch": 0.015281757402101241,
      "grad_norm": 0.8185105323791504,
      "learning_rate": 0.00016,
      "loss": 2.8966,
      "step": 8
    },
    {
      "epoch": 0.017191977077363897,
      "grad_norm": 0.8913972973823547,
      "learning_rate": 0.00018,
      "loss": 2.6794,
      "step": 9
    },
    {
      "epoch": 0.017191977077363897,
      "eval_loss": 2.79345703125,
      "eval_runtime": 12.8619,
      "eval_samples_per_second": 17.183,
      "eval_steps_per_second": 8.63,
      "step": 9
    },
    {
      "epoch": 0.019102196752626553,
      "grad_norm": 1.5539528131484985,
      "learning_rate": 0.0002,
      "loss": 2.8706,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3699128496291840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}