{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.17777777777777778,
  "eval_steps": 50,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.044444444444444446,
      "grad_norm": 49.98692321777344,
      "learning_rate": 5e-07,
      "logits/chosen": -1.6455594301223755,
      "logits/rejected": -1.8609813451766968,
      "logps/chosen": -69.99601745605469,
      "logps/rejected": -28.68402099609375,
      "loss": 0.4811,
      "rewards/accuracies": 0.9200000166893005,
      "rewards/chosen": 0.43002215027809143,
      "rewards/margins": 0.5248908400535583,
      "rewards/rejected": -0.09486865252256393,
      "step": 50
    },
    {
      "epoch": 0.044444444444444446,
      "eval_logits/chosen": -1.683065414428711,
      "eval_logits/rejected": -1.8381606340408325,
      "eval_logps/chosen": -65.2982406616211,
      "eval_logps/rejected": -31.1202335357666,
      "eval_loss": 0.26570644974708557,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 0.9695585370063782,
      "eval_rewards/margins": 1.199601650238037,
      "eval_rewards/rejected": -0.23004300892353058,
      "eval_runtime": 547.5928,
      "eval_samples_per_second": 0.51,
      "eval_steps_per_second": 0.51,
      "step": 50
    },
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 18.30071449279785,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7492402791976929,
      "logits/rejected": -1.8840913772583008,
      "logps/chosen": -62.02247619628906,
      "logps/rejected": -31.541664123535156,
      "loss": 0.1648,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.368556022644043,
      "rewards/margins": 1.8034323453903198,
      "rewards/rejected": -0.4348761737346649,
      "step": 100
    },
    {
      "epoch": 0.08888888888888889,
      "eval_logits/chosen": -1.717586636543274,
      "eval_logits/rejected": -1.840760588645935,
      "eval_logps/chosen": -57.23891067504883,
      "eval_logps/rejected": -35.770687103271484,
      "eval_loss": 0.0847974419593811,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.77549147605896,
      "eval_rewards/margins": 2.4705796241760254,
      "eval_rewards/rejected": -0.6950880885124207,
      "eval_runtime": 547.9305,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.509,
      "step": 100
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 6.001527309417725,
      "learning_rate": 5e-07,
      "logits/chosen": -1.753422498703003,
      "logits/rejected": -1.856553316116333,
      "logps/chosen": -55.98763656616211,
      "logps/rejected": -39.17717361450195,
      "loss": 0.0522,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.045351266860962,
      "rewards/margins": 3.051741361618042,
      "rewards/rejected": -1.00639009475708,
      "step": 150
    },
    {
      "epoch": 0.13333333333333333,
      "eval_logits/chosen": -1.7338056564331055,
      "eval_logits/rejected": -1.838354229927063,
      "eval_logps/chosen": -51.332515716552734,
      "eval_logps/rejected": -42.16679382324219,
      "eval_loss": 0.026723211631178856,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.366131067276001,
      "eval_rewards/margins": 3.7008297443389893,
      "eval_rewards/rejected": -1.3346984386444092,
      "eval_runtime": 548.2166,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.509,
      "step": 150
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 3.3345212936401367,
      "learning_rate": 5e-07,
      "logits/chosen": -1.7408527135849,
      "logits/rejected": -1.8468482494354248,
      "logps/chosen": -51.4506721496582,
      "logps/rejected": -46.23123550415039,
      "loss": 0.0157,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.529080867767334,
      "rewards/margins": 4.344725131988525,
      "rewards/rejected": -1.8156436681747437,
      "step": 200
    },
    {
      "epoch": 0.17777777777777778,
      "eval_logits/chosen": -1.744242787361145,
      "eval_logits/rejected": -1.8309202194213867,
      "eval_logps/chosen": -47.80553436279297,
      "eval_logps/rejected": -54.1657600402832,
      "eval_loss": 0.006161259487271309,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 2.7188286781311035,
      "eval_rewards/margins": 5.253425121307373,
      "eval_rewards/rejected": -2.5345959663391113,
      "eval_runtime": 548.15,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.509,
      "step": 200
    }
  ],
  "logging_steps": 50,
  "max_steps": 1125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}