{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9874476987447699,
  "eval_steps": 500,
  "global_step": 59,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 6.666666666666665e-08,
      "logits/chosen": -2.828038215637207,
      "logits/rejected": -2.728235960006714,
      "logps/chosen": -198.4669189453125,
      "logps/pi_response": -116.54592895507812,
      "logps/ref_response": -116.54592895507812,
      "logps/rejected": -205.3916015625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.9440458281608213e-07,
      "logits/chosen": -2.744715690612793,
      "logits/rejected": -2.706997871398926,
      "logps/chosen": -219.86297607421875,
      "logps/pi_response": -125.73471069335938,
      "logps/ref_response": -125.32001495361328,
      "logps/rejected": -269.349853515625,
      "loss": 0.688,
      "rewards/accuracies": 0.5590277910232544,
      "rewards/chosen": -0.017961828038096428,
      "rewards/margins": 0.015605524182319641,
      "rewards/rejected": -0.03356735408306122,
      "step": 10
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.349966003036421e-07,
      "logits/chosen": -2.7669131755828857,
      "logits/rejected": -2.725191593170166,
      "logps/chosen": -252.75650024414062,
      "logps/pi_response": -119.78802490234375,
      "logps/ref_response": -122.2673110961914,
      "logps/rejected": -298.3902282714844,
      "loss": 0.6394,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.16843266785144806,
      "rewards/margins": 0.18965782225131989,
      "rewards/rejected": -0.35809049010276794,
      "step": 20
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.2952931280049625e-07,
      "logits/chosen": -2.7630202770233154,
      "logits/rejected": -2.725600481033325,
      "logps/chosen": -275.5626525878906,
      "logps/pi_response": -133.33511352539062,
      "logps/ref_response": -124.40946197509766,
      "logps/rejected": -339.12408447265625,
      "loss": 0.5989,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.38713958859443665,
      "rewards/margins": 0.39169779419898987,
      "rewards/rejected": -0.7788373827934265,
      "step": 30
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.1398695954469597e-07,
      "logits/chosen": -2.6956827640533447,
      "logits/rejected": -2.6605703830718994,
      "logps/chosen": -278.9873962402344,
      "logps/pi_response": -139.6653289794922,
      "logps/ref_response": -118.28387451171875,
      "logps/rejected": -349.42584228515625,
      "loss": 0.5801,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.5085138082504272,
      "rewards/margins": 0.4421776831150055,
      "rewards/rejected": -0.9506915211677551,
      "step": 40
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.779127764652889e-08,
      "logits/chosen": -2.6794543266296387,
      "logits/rejected": -2.631016969680786,
      "logps/chosen": -276.74078369140625,
      "logps/pi_response": -147.046875,
      "logps/ref_response": -121.82298278808594,
      "logps/rejected": -342.2689514160156,
      "loss": 0.5784,
      "rewards/accuracies": 0.7093750238418579,
      "rewards/chosen": -0.4734733998775482,
      "rewards/margins": 0.4400869905948639,
      "rewards/rejected": -0.9135602712631226,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 59,
      "total_flos": 0.0,
      "train_loss": 0.6098380088806152,
      "train_runtime": 3606.0648,
      "train_samples_per_second": 4.238,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 10,
  "max_steps": 59,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}