{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.405405405405405,
  "eval_steps": 500,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.36036036036036034,
      "grad_norm": 0.5300200581550598,
      "learning_rate": 9.259259259259259e-07,
      "logits/chosen": 1.7405741214752197,
      "logits/rejected": 1.663368582725525,
      "logps/chosen": -97.57881164550781,
      "logps/rejected": -70.59793853759766,
      "loss": 0.6938,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": 0.0018835498485714197,
      "rewards/margins": -0.0008800366776995361,
      "rewards/rejected": 0.0027635858859866858,
      "step": 10
    },
    {
      "epoch": 0.7207207207207207,
      "grad_norm": 0.4883837401866913,
      "learning_rate": 1.8518518518518519e-06,
      "logits/chosen": 1.8280715942382812,
      "logits/rejected": 1.8215343952178955,
      "logps/chosen": -90.60624694824219,
      "logps/rejected": -79.04981994628906,
      "loss": 0.6945,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": 0.0014454321935772896,
      "rewards/margins": -0.004174981266260147,
      "rewards/rejected": 0.005620413459837437,
      "step": 20
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 0.5554109215736389,
      "learning_rate": 2.7777777777777783e-06,
      "logits/chosen": 1.7980501651763916,
      "logits/rejected": 1.841104507446289,
      "logps/chosen": -80.78228759765625,
      "logps/rejected": -85.08882141113281,
      "loss": 0.693,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": 0.0042568682692945,
      "rewards/margins": -0.0038191028870642185,
      "rewards/rejected": 0.008075973019003868,
      "step": 30
    },
    {
      "epoch": 1.4414414414414414,
      "grad_norm": 0.537497341632843,
      "learning_rate": 3.7037037037037037e-06,
      "logits/chosen": 1.7320470809936523,
      "logits/rejected": 1.7411377429962158,
      "logps/chosen": -82.24813842773438,
      "logps/rejected": -80.90709686279297,
      "loss": 0.6952,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": -0.00350201572291553,
      "rewards/margins": -0.012091752141714096,
      "rewards/rejected": 0.008589735254645348,
      "step": 40
    },
    {
      "epoch": 1.8018018018018018,
      "grad_norm": 0.5023094415664673,
      "learning_rate": 4.62962962962963e-06,
      "logits/chosen": 1.8292573690414429,
      "logits/rejected": 1.8632844686508179,
      "logps/chosen": -85.98481750488281,
      "logps/rejected": -86.14008331298828,
      "loss": 0.6928,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": 0.00168000184930861,
      "rewards/margins": -0.011136507615447044,
      "rewards/rejected": 0.01281650923192501,
      "step": 50
    },
    {
      "epoch": 2.1621621621621623,
      "grad_norm": 0.4981901943683624,
      "learning_rate": 4.998119881260576e-06,
      "logits/chosen": 1.8536121845245361,
      "logits/rejected": 1.7948782444000244,
      "logps/chosen": -90.0439682006836,
      "logps/rejected": -79.8309555053711,
      "loss": 0.6914,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": 0.0025153260212391615,
      "rewards/margins": -0.0021209525875747204,
      "rewards/rejected": 0.004636278375983238,
      "step": 60
    },
    {
      "epoch": 2.5225225225225225,
      "grad_norm": 0.6208468079566956,
      "learning_rate": 4.9866405060165044e-06,
      "logits/chosen": 1.7845121622085571,
      "logits/rejected": 1.8721017837524414,
      "logps/chosen": -74.6461181640625,
      "logps/rejected": -98.91078186035156,
      "loss": 0.694,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": 0.00022823773906566203,
      "rewards/margins": 0.0024594543501734734,
      "rewards/rejected": -0.00223121652379632,
      "step": 70
    },
    {
      "epoch": 2.8828828828828827,
      "grad_norm": 0.5074446797370911,
      "learning_rate": 4.964774158361991e-06,
      "logits/chosen": 1.8522154092788696,
      "logits/rejected": 1.8161392211914062,
      "logps/chosen": -89.16864013671875,
      "logps/rejected": -81.97349548339844,
      "loss": 0.6911,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": 0.006215038243681192,
      "rewards/margins": 0.005521883722394705,
      "rewards/rejected": 0.0006931538810022175,
      "step": 80
    },
    {
      "epoch": 3.2432432432432434,
      "grad_norm": 0.6148263812065125,
      "learning_rate": 4.93261217644956e-06,
      "logits/chosen": 1.7915083169937134,
      "logits/rejected": 1.7579513788223267,
      "logps/chosen": -88.66618347167969,
      "logps/rejected": -80.2238998413086,
      "loss": 0.6918,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.005774274934083223,
      "rewards/margins": 0.016623441129922867,
      "rewards/rejected": -0.010849165730178356,
      "step": 90
    },
    {
      "epoch": 3.6036036036036037,
      "grad_norm": 0.6143240332603455,
      "learning_rate": 4.8902889044347e-06,
      "logits/chosen": 1.6886920928955078,
      "logits/rejected": 1.8121178150177002,
      "logps/chosen": -72.908203125,
      "logps/rejected": -92.44017028808594,
      "loss": 0.6911,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.0015121791511774063,
      "rewards/margins": 0.0036420777905732393,
      "rewards/rejected": -0.005154256708920002,
      "step": 100
    },
    {
      "epoch": 3.963963963963964,
      "grad_norm": 0.7140340209007263,
      "learning_rate": 4.837981131305475e-06,
      "logits/chosen": 1.7481105327606201,
      "logits/rejected": 1.723141074180603,
      "logps/chosen": -80.63452911376953,
      "logps/rejected": -72.64492797851562,
      "loss": 0.6894,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.006246576085686684,
      "rewards/margins": -0.0035218377597630024,
      "rewards/rejected": -0.0027247383259236813,
      "step": 110
    },
    {
      "epoch": 4.324324324324325,
      "grad_norm": 1.1645854711532593,
      "learning_rate": 4.775907352415367e-06,
      "logits/chosen": 1.7416290044784546,
      "logits/rejected": 1.8237574100494385,
      "logps/chosen": -89.01248931884766,
      "logps/rejected": -92.31901550292969,
      "loss": 0.6869,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.008855113759636879,
      "rewards/margins": 0.013151508755981922,
      "rewards/rejected": -0.022006623446941376,
      "step": 120
    },
    {
      "epoch": 4.684684684684685,
      "grad_norm": 0.6638664603233337,
      "learning_rate": 4.70432685680402e-06,
      "logits/chosen": 1.7124770879745483,
      "logits/rejected": 1.7777938842773438,
      "logps/chosen": -84.87271881103516,
      "logps/rejected": -92.4839096069336,
      "loss": 0.6855,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.004437069408595562,
      "rewards/margins": 0.019811339676380157,
      "rewards/rejected": -0.024248410016298294,
      "step": 130
    },
    {
      "epoch": 5.045045045045045,
      "grad_norm": 0.785169780254364,
      "learning_rate": 4.623538644118244e-06,
      "logits/chosen": 1.7838690280914307,
      "logits/rejected": 1.8052574396133423,
      "logps/chosen": -94.57842254638672,
      "logps/rejected": -80.6390609741211,
      "loss": 0.6836,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.026147600263357162,
      "rewards/margins": 0.010936126112937927,
      "rewards/rejected": -0.03708372637629509,
      "step": 140
    },
    {
      "epoch": 5.405405405405405,
      "grad_norm": 0.745952844619751,
      "learning_rate": 4.533880175657419e-06,
      "logits/chosen": 1.7925735712051392,
      "logits/rejected": 1.7987396717071533,
      "logps/chosen": -82.4854507446289,
      "logps/rejected": -77.50787353515625,
      "loss": 0.6767,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.013469865545630455,
      "rewards/margins": 0.04023212194442749,
      "rewards/rejected": -0.053701985627412796,
      "step": 150
    }
  ],
  "logging_steps": 10,
  "max_steps": 540,
  "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": 5.08406132209877e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}