{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.938271604938271,
  "eval_steps": 500,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3292181069958848,
      "grad_norm": 0.5561477541923523,
      "learning_rate": 8.333333333333333e-07,
      "logits/chosen": 1.7485754489898682,
      "logits/rejected": 1.8832639455795288,
      "logps/chosen": -70.18267059326172,
      "logps/rejected": -77.9986343383789,
      "loss": 0.6938,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": 0.005453853867948055,
      "rewards/margins": 0.013218576088547707,
      "rewards/rejected": -0.007764720823615789,
      "step": 10
    },
    {
      "epoch": 0.6584362139917695,
      "grad_norm": 0.48141908645629883,
      "learning_rate": 1.6666666666666667e-06,
      "logits/chosen": 1.9016907215118408,
      "logits/rejected": 1.9251121282577515,
      "logps/chosen": -96.5027847290039,
      "logps/rejected": -87.00735473632812,
      "loss": 0.6937,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.005797500256448984,
      "rewards/margins": -0.0004409264656715095,
      "rewards/rejected": -0.005356573965400457,
      "step": 20
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 0.4639015197753906,
      "learning_rate": 2.5e-06,
      "logits/chosen": 1.7938541173934937,
      "logits/rejected": 1.6962993144989014,
      "logps/chosen": -71.47590637207031,
      "logps/rejected": -66.45989227294922,
      "loss": 0.6942,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": -0.004830303601920605,
      "rewards/margins": -0.00935445912182331,
      "rewards/rejected": 0.004524155054241419,
      "step": 30
    },
    {
      "epoch": 1.316872427983539,
      "grad_norm": 0.44931092858314514,
      "learning_rate": 3.3333333333333333e-06,
      "logits/chosen": 1.8256213665008545,
      "logits/rejected": 1.8677200078964233,
      "logps/chosen": -75.90711975097656,
      "logps/rejected": -76.26548767089844,
      "loss": 0.6935,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.004859285429120064,
      "rewards/margins": -0.007447429001331329,
      "rewards/rejected": 0.0025881435722112656,
      "step": 40
    },
    {
      "epoch": 1.646090534979424,
      "grad_norm": 0.512350857257843,
      "learning_rate": 4.166666666666667e-06,
      "logits/chosen": 1.7572576999664307,
      "logits/rejected": 1.7408854961395264,
      "logps/chosen": -80.90664672851562,
      "logps/rejected": -85.82096862792969,
      "loss": 0.6937,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -0.00120059447363019,
      "rewards/margins": -0.0018891148502007127,
      "rewards/rejected": 0.0006885197362862527,
      "step": 50
    },
    {
      "epoch": 1.9753086419753085,
      "grad_norm": 0.5260242819786072,
      "learning_rate": 5e-06,
      "logits/chosen": 1.827275037765503,
      "logits/rejected": 1.8168131113052368,
      "logps/chosen": -86.74467468261719,
      "logps/rejected": -79.20576477050781,
      "loss": 0.6937,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": 0.0014693590346723795,
      "rewards/margins": -0.0025326632894575596,
      "rewards/rejected": 0.00400202302262187,
      "step": 60
    },
    {
      "epoch": 2.3045267489711936,
      "grad_norm": 0.5231289267539978,
      "learning_rate": 4.995770395678171e-06,
      "logits/chosen": 1.7851394414901733,
      "logits/rejected": 1.8952877521514893,
      "logps/chosen": -81.03253173828125,
      "logps/rejected": -88.5263442993164,
      "loss": 0.6932,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -0.003017458599060774,
      "rewards/margins": -0.002731734188273549,
      "rewards/rejected": -0.0002857256622519344,
      "step": 70
    },
    {
      "epoch": 2.633744855967078,
      "grad_norm": 0.5552361011505127,
      "learning_rate": 4.983095894354858e-06,
      "logits/chosen": 1.8227930068969727,
      "logits/rejected": 1.7752052545547485,
      "logps/chosen": -89.98479461669922,
      "logps/rejected": -72.01054382324219,
      "loss": 0.6928,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": 0.0013810636010020971,
      "rewards/margins": -0.0008344938978552818,
      "rewards/rejected": 0.0022155570331960917,
      "step": 80
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 0.603434681892395,
      "learning_rate": 4.962019382530521e-06,
      "logits/chosen": 1.829049825668335,
      "logits/rejected": 1.78665030002594,
      "logps/chosen": -81.95549011230469,
      "logps/rejected": -76.07003021240234,
      "loss": 0.6909,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.0062006814405322075,
      "rewards/margins": 0.003528360743075609,
      "rewards/rejected": 0.00267231953330338,
      "step": 90
    },
    {
      "epoch": 3.292181069958848,
      "grad_norm": 0.6200206279754639,
      "learning_rate": 4.93261217644956e-06,
      "logits/chosen": 1.7920262813568115,
      "logits/rejected": 1.7817165851593018,
      "logps/chosen": -84.39167022705078,
      "logps/rejected": -85.04205322265625,
      "loss": 0.6907,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.0063859038054943085,
      "rewards/margins": 0.007152262143790722,
      "rewards/rejected": -0.0007663581636734307,
      "step": 100
    },
    {
      "epoch": 3.6213991769547325,
      "grad_norm": 0.5937617421150208,
      "learning_rate": 4.894973780788722e-06,
      "logits/chosen": 1.8367125988006592,
      "logits/rejected": 1.8536920547485352,
      "logps/chosen": -67.41716003417969,
      "logps/rejected": -72.06455993652344,
      "loss": 0.6891,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.00015740413800813258,
      "rewards/margins": 0.013882984407246113,
      "rewards/rejected": -0.01404038816690445,
      "step": 110
    },
    {
      "epoch": 3.950617283950617,
      "grad_norm": 0.8695696592330933,
      "learning_rate": 4.849231551964771e-06,
      "logits/chosen": 1.8238048553466797,
      "logits/rejected": 1.7972408533096313,
      "logps/chosen": -104.34183502197266,
      "logps/rejected": -81.703369140625,
      "loss": 0.6864,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.00020323302305769175,
      "rewards/margins": 0.011230283416807652,
      "rewards/rejected": -0.011027050204575062,
      "step": 120
    },
    {
      "epoch": 4.279835390946502,
      "grad_norm": 0.7833828926086426,
      "learning_rate": 4.7955402672006855e-06,
      "logits/chosen": 1.7440074682235718,
      "logits/rejected": 1.7824723720550537,
      "logps/chosen": -92.3200912475586,
      "logps/rejected": -85.88248443603516,
      "loss": 0.6861,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.0005388978170230985,
      "rewards/margins": 0.011814715340733528,
      "rewards/rejected": -0.011275815777480602,
      "step": 130
    },
    {
      "epoch": 4.609053497942387,
      "grad_norm": 0.7381640672683716,
      "learning_rate": 4.734081600808531e-06,
      "logits/chosen": 1.6734364032745361,
      "logits/rejected": 1.7311270236968994,
      "logps/chosen": -88.11524963378906,
      "logps/rejected": -97.87281799316406,
      "loss": 0.682,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.006451706402003765,
      "rewards/margins": 0.021939774975180626,
      "rewards/rejected": -0.028391480445861816,
      "step": 140
    },
    {
      "epoch": 4.938271604938271,
      "grad_norm": 0.7793841361999512,
      "learning_rate": 4.665063509461098e-06,
      "logits/chosen": 1.8139768838882446,
      "logits/rejected": 1.766761064529419,
      "logps/chosen": -77.28334045410156,
      "logps/rejected": -78.78504943847656,
      "loss": 0.6777,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.011237703263759613,
      "rewards/margins": 0.03930521756410599,
      "rewards/rejected": -0.050542913377285004,
      "step": 150
    }
  ],
  "logging_steps": 10,
  "max_steps": 600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 5.0929507662351565e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}