last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.045151953690303906,
  "eval_steps": 10,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0011577424023154848,
      "eval_loss": 0.5778371095657349,
      "eval_runtime": 23.1824,
      "eval_samples_per_second": 15.702,
      "eval_steps_per_second": 7.851,
      "step": 1
    },
    {
      "epoch": 0.005788712011577424,
      "grad_norm": 0.7782766222953796,
      "learning_rate": 5e-05,
      "loss": 0.6121,
      "step": 5
    },
    {
      "epoch": 0.011577424023154847,
      "grad_norm": 0.7705448269844055,
      "learning_rate": 0.0001,
      "loss": 0.5238,
      "step": 10
    },
    {
      "epoch": 0.011577424023154847,
      "eval_loss": 0.4204440414905548,
      "eval_runtime": 22.911,
      "eval_samples_per_second": 15.888,
      "eval_steps_per_second": 7.944,
      "step": 10
    },
    {
      "epoch": 0.017366136034732273,
      "grad_norm": 1.1096508502960205,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.3631,
      "step": 15
    },
    {
      "epoch": 0.023154848046309694,
      "grad_norm": 1.1899404525756836,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.2529,
      "step": 20
    },
    {
      "epoch": 0.023154848046309694,
      "eval_loss": 0.21950246393680573,
      "eval_runtime": 22.9687,
      "eval_samples_per_second": 15.848,
      "eval_steps_per_second": 7.924,
      "step": 20
    },
    {
      "epoch": 0.02894356005788712,
      "grad_norm": 0.7090124487876892,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.1765,
      "step": 25
    },
    {
      "epoch": 0.03473227206946455,
      "grad_norm": 0.759050726890564,
      "learning_rate": 5e-05,
      "loss": 0.1274,
      "step": 30
    },
    {
      "epoch": 0.03473227206946455,
      "eval_loss": 0.14774048328399658,
      "eval_runtime": 23.1699,
      "eval_samples_per_second": 15.71,
      "eval_steps_per_second": 7.855,
      "step": 30
    },
    {
      "epoch": 0.04052098408104197,
      "grad_norm": 1.229648470878601,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.1307,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6839846480904192.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}