{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.3333333333333333,
  "eval_steps": 100,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 0.5337432026863098,
      "learning_rate": 0.0004999965616101344,
      "logits/chosen": 0.23934367299079895,
      "logits/rejected": 0.42762428522109985,
      "logps/chosen": -113.45499420166016,
      "logps/rejected": -75.61299896240234,
      "loss": 0.271,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": 0.9617708325386047,
      "rewards/margins": 3.3194146156311035,
      "rewards/rejected": -2.3576436042785645,
      "step": 10
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.0004492771113291383,
      "learning_rate": 0.0004999690550586881,
      "logits/chosen": 0.39065027236938477,
      "logits/rejected": 0.9774702191352844,
      "logps/chosen": -103.68326568603516,
      "logps/rejected": -159.09165954589844,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5827335119247437,
      "rewards/margins": 12.054422378540039,
      "rewards/rejected": -10.471688270568848,
      "step": 20
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.009628095664083958,
      "learning_rate": 0.0004999140449822787,
      "logits/chosen": 0.39641499519348145,
      "logits/rejected": 1.1506410837173462,
      "logps/chosen": -112.4819564819336,
      "logps/rejected": -203.46591186523438,
      "loss": 0.0144,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": 0.7971093654632568,
      "rewards/margins": 15.904172897338867,
      "rewards/rejected": -15.107061386108398,
      "step": 30
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.16114771366119385,
      "learning_rate": 0.0004998315374335394,
      "logits/chosen": 0.42631012201309204,
      "logits/rejected": 1.436499834060669,
      "logps/chosen": -120.41705322265625,
      "logps/rejected": -188.07252502441406,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.19870629906654358,
      "rewards/margins": 13.751760482788086,
      "rewards/rejected": -13.553054809570312,
      "step": 40
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 2.7873742510564625e-05,
      "learning_rate": 0.0004997215414905875,
      "logits/chosen": 0.45363932847976685,
      "logits/rejected": 1.7247793674468994,
      "logps/chosen": -107.3564453125,
      "logps/rejected": -212.8771514892578,
      "loss": 0.0011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.316047191619873,
      "rewards/margins": 17.600120544433594,
      "rewards/rejected": -16.284074783325195,
      "step": 50
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.3321376829699147e-06,
      "learning_rate": 0.0004995840692560257,
      "logits/chosen": 0.3636583983898163,
      "logits/rejected": 1.690006971359253,
      "logps/chosen": -117.81230163574219,
      "logps/rejected": -231.50650024414062,
      "loss": 0.0005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.40089258551597595,
      "rewards/margins": 18.288105010986328,
      "rewards/rejected": -17.8872127532959,
      "step": 60
    },
    {
      "epoch": 0.9333333333333333,
      "grad_norm": 9.223950655723456e-06,
      "learning_rate": 0.0004994191358556106,
      "logits/chosen": 0.4559127390384674,
      "logits/rejected": 1.189163088798523,
      "logps/chosen": -108.7114486694336,
      "logps/rejected": -249.8244171142578,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3001081943511963,
      "rewards/margins": 20.991580963134766,
      "rewards/rejected": -19.691471099853516,
      "step": 70
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 1.353669034642735e-07,
      "learning_rate": 0.0004992267594365879,
      "logits/chosen": 0.617511510848999,
      "logits/rejected": 1.2590217590332031,
      "logps/chosen": -98.4377670288086,
      "logps/rejected": -260.7948913574219,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.062624216079712,
      "rewards/margins": 22.795236587524414,
      "rewards/rejected": -20.73261070251465,
      "step": 80
    },
    {
      "epoch": 1.2,
      "grad_norm": 7.2894854383775964e-06,
      "learning_rate": 0.0004990069611656963,
      "logits/chosen": 0.8833354115486145,
      "logits/rejected": 1.289421796798706,
      "logps/chosen": -99.6741943359375,
      "logps/rejected": -247.78750610351562,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.3024544715881348,
      "rewards/margins": 21.798070907592773,
      "rewards/rejected": -19.49561882019043,
      "step": 90
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 5.883700850972673e-06,
      "learning_rate": 0.0004987597652268379,
      "logits/chosen": 0.957192599773407,
      "logits/rejected": 1.1422635316848755,
      "logps/chosen": -107.17437744140625,
      "logps/rejected": -239.02804565429688,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3617346286773682,
      "rewards/margins": 20.076194763183594,
      "rewards/rejected": -18.714458465576172,
      "step": 100
    },
    {
      "epoch": 1.3333333333333333,
      "eval_logits/chosen": 0.9479974508285522,
      "eval_logits/rejected": 1.0917534828186035,
      "eval_logps/chosen": -110.24896240234375,
      "eval_logps/rejected": -238.44570922851562,
      "eval_loss": 8.311428246088326e-05,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.203629493713379,
      "eval_rewards/margins": 19.736553192138672,
      "eval_rewards/rejected": -18.532922744750977,
      "eval_runtime": 161.9066,
      "eval_samples_per_second": 1.235,
      "eval_steps_per_second": 1.235,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 3000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 100,
  "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": 4,
  "trial_name": null,
  "trial_params": null
}