trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 274,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18248175182481752,
      "grad_norm": 0.0007704155868850648,
      "learning_rate": 9.803954791481238e-07,
      "logits/chosen": -3.126544237136841,
      "logits/rejected": -3.302138566970825,
      "logps/chosen": -39.869632720947266,
      "logps/rejected": -157.12303161621094,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.736296653747559,
      "rewards/margins": 19.26922607421875,
      "rewards/rejected": -10.532929420471191,
      "step": 50
    },
    {
      "epoch": 0.36496350364963503,
      "grad_norm": 0.00025232642656192183,
      "learning_rate": 8.03112705483319e-07,
      "logits/chosen": -3.0920073986053467,
      "logits/rejected": -3.2855396270751953,
      "logps/chosen": -37.723506927490234,
      "logps/rejected": -163.46827697753906,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.903458595275879,
      "rewards/margins": 20.080211639404297,
      "rewards/rejected": -11.176753044128418,
      "step": 100
    },
    {
      "epoch": 0.5474452554744526,
      "grad_norm": 1.50539008245687e-05,
      "learning_rate": 5.063851773579869e-07,
      "logits/chosen": -3.077618360519409,
      "logits/rejected": -3.280364751815796,
      "logps/chosen": -40.20799255371094,
      "logps/rejected": -170.47584533691406,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.536523818969727,
      "rewards/margins": 20.434810638427734,
      "rewards/rejected": -11.898284912109375,
      "step": 150
    },
    {
      "epoch": 0.7299270072992701,
      "grad_norm": 1.4836322407063562e-05,
      "learning_rate": 2.071415028359026e-07,
      "logits/chosen": -3.073031425476074,
      "logits/rejected": -3.256861925125122,
      "logps/chosen": -41.553260803222656,
      "logps/rejected": -174.45729064941406,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.274081230163574,
      "rewards/margins": 20.350595474243164,
      "rewards/rejected": -12.076513290405273,
      "step": 200
    },
    {
      "epoch": 0.9124087591240876,
      "grad_norm": 2.432690564546647e-07,
      "learning_rate": 2.3301803972534728e-08,
      "logits/chosen": -3.053009033203125,
      "logits/rejected": -3.253528356552124,
      "logps/chosen": -38.02189636230469,
      "logps/rejected": -169.2959747314453,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.201343536376953,
      "rewards/margins": 20.2254581451416,
      "rewards/rejected": -12.024112701416016,
      "step": 250
    }
  ],
  "logging_steps": 50,
  "max_steps": 274,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}