{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9820224719101125,
  "eval_steps": 500,
  "global_step": 110,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0898876404494382,
      "grad_norm": 250.8028214771956,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": -1.31005859375,
      "logits/rejected": -1.302734375,
      "logps/chosen": -219.6999969482422,
      "logps/rejected": -277.82501220703125,
      "loss": 0.3525,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": -2.008740186691284,
      "rewards/margins": 2.612988233566284,
      "rewards/rejected": -4.619116306304932,
      "step": 5
    },
    {
      "epoch": 0.1797752808988764,
      "grad_norm": 170.2256638749618,
      "learning_rate": 9.09090909090909e-08,
      "logits/chosen": -1.200585961341858,
      "logits/rejected": -1.272363305091858,
      "logps/chosen": -205.9250030517578,
      "logps/rejected": -230.5124969482422,
      "loss": 0.3659,
      "rewards/accuracies": 0.824999988079071,
      "rewards/chosen": -1.0198485851287842,
      "rewards/margins": 2.195556640625,
      "rewards/rejected": -3.2144532203674316,
      "step": 10
    },
    {
      "epoch": 0.2696629213483146,
      "grad_norm": 295.1115992445591,
      "learning_rate": 1.3636363636363635e-07,
      "logits/chosen": -1.2916991710662842,
      "logits/rejected": -1.23388671875,
      "logps/chosen": -180.6999969482422,
      "logps/rejected": -212.14999389648438,
      "loss": 0.437,
      "rewards/accuracies": 0.8374999761581421,
      "rewards/chosen": -2.502392530441284,
      "rewards/margins": 2.3900389671325684,
      "rewards/rejected": -4.890429496765137,
      "step": 15
    },
    {
      "epoch": 0.3595505617977528,
      "grad_norm": 277.7225526041515,
      "learning_rate": 1.818181818181818e-07,
      "logits/chosen": -1.259765625,
      "logits/rejected": -1.2590820789337158,
      "logps/chosen": -197.78750610351562,
      "logps/rejected": -218.91250610351562,
      "loss": 0.3354,
      "rewards/accuracies": 0.8374999761581421,
      "rewards/chosen": -2.3748536109924316,
      "rewards/margins": 2.6907958984375,
      "rewards/rejected": -5.071179389953613,
      "step": 20
    },
    {
      "epoch": 0.449438202247191,
      "grad_norm": 115.01279948814654,
      "learning_rate": 2.2727272727272726e-07,
      "logits/chosen": -1.180078148841858,
      "logits/rejected": -1.196386694908142,
      "logps/chosen": -206.75,
      "logps/rejected": -224.52499389648438,
      "loss": 0.4124,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": -2.122851610183716,
      "rewards/margins": 1.923669457435608,
      "rewards/rejected": -4.044787406921387,
      "step": 25
    },
    {
      "epoch": 0.5393258426966292,
      "grad_norm": 72.00062004716999,
      "learning_rate": 2.727272727272727e-07,
      "logits/chosen": -1.245996117591858,
      "logits/rejected": -1.2501952648162842,
      "logps/chosen": -208.91250610351562,
      "logps/rejected": -217.8125,
      "loss": 0.3735,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -1.261322021484375,
      "rewards/margins": 2.326733350753784,
      "rewards/rejected": -3.585009813308716,
      "step": 30
    },
    {
      "epoch": 0.6292134831460674,
      "grad_norm": 159.67880662723658,
      "learning_rate": 3.1818181818181815e-07,
      "logits/chosen": -1.2498047351837158,
      "logits/rejected": -1.212011694908142,
      "logps/chosen": -200.75,
      "logps/rejected": -203.1750030517578,
      "loss": 0.3959,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": -2.0471434593200684,
      "rewards/margins": 2.1449217796325684,
      "rewards/rejected": -4.191888332366943,
      "step": 35
    },
    {
      "epoch": 0.7191011235955056,
      "grad_norm": 159.3228331743839,
      "learning_rate": 3.636363636363636e-07,
      "logits/chosen": -1.304296851158142,
      "logits/rejected": -1.2903320789337158,
      "logps/chosen": -211.1750030517578,
      "logps/rejected": -218.71249389648438,
      "loss": 0.3872,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": -1.924902319908142,
      "rewards/margins": 2.205249071121216,
      "rewards/rejected": -4.130883693695068,
      "step": 40
    },
    {
      "epoch": 0.8089887640449438,
      "grad_norm": 209.54928027137586,
      "learning_rate": 4.090909090909091e-07,
      "logits/chosen": -1.2551758289337158,
      "logits/rejected": -1.2458007335662842,
      "logps/chosen": -222.3249969482422,
      "logps/rejected": -237.64999389648438,
      "loss": 0.4076,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": -1.8654601573944092,
      "rewards/margins": 2.3793578147888184,
      "rewards/rejected": -4.247656345367432,
      "step": 45
    },
    {
      "epoch": 0.898876404494382,
      "grad_norm": 134.6125923552392,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": -1.2243163585662842,
      "logits/rejected": -1.208593726158142,
      "logps/chosen": -203.58749389648438,
      "logps/rejected": -223.9875030517578,
      "loss": 0.4013,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -2.0265870094299316,
      "rewards/margins": 2.5132813453674316,
      "rewards/rejected": -4.545800685882568,
      "step": 50
    },
    {
      "epoch": 0.9887640449438202,
      "grad_norm": 189.31983158924976,
      "learning_rate": 5e-07,
      "logits/chosen": -1.2102539539337158,
      "logits/rejected": -1.205078125,
      "logps/chosen": -204.6750030517578,
      "logps/rejected": -235.71249389648438,
      "loss": 0.4268,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.649023413658142,
      "rewards/margins": 2.216552734375,
      "rewards/rejected": -3.8661131858825684,
      "step": 55
    },
    {
      "epoch": 1.0831460674157303,
      "grad_norm": 125.21413207415151,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": -1.2853515148162842,
      "logits/rejected": -1.318261742591858,
      "logps/chosen": -202.3000030517578,
      "logps/rejected": -216.0124969482422,
      "loss": 0.3312,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -1.6000487804412842,
      "rewards/margins": 2.2996582984924316,
      "rewards/rejected": -3.900146484375,
      "step": 60
    },
    {
      "epoch": 1.1730337078651685,
      "grad_norm": 164.73999131891642,
      "learning_rate": 4.090909090909091e-07,
      "logits/chosen": -1.240136742591858,
      "logits/rejected": -1.2451171875,
      "logps/chosen": -229.66250610351562,
      "logps/rejected": -257.3999938964844,
      "loss": 0.2984,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": -1.916986107826233,
      "rewards/margins": 2.175122022628784,
      "rewards/rejected": -4.092577934265137,
      "step": 65
    },
    {
      "epoch": 1.2629213483146067,
      "grad_norm": 99.20074290357307,
      "learning_rate": 3.636363636363636e-07,
      "logits/chosen": -1.254492163658142,
      "logits/rejected": -1.2688477039337158,
      "logps/chosen": -222.0625,
      "logps/rejected": -236.8625030517578,
      "loss": 0.3024,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": -1.540551781654358,
      "rewards/margins": 2.271484375,
      "rewards/rejected": -3.810473680496216,
      "step": 70
    },
    {
      "epoch": 1.3528089887640449,
      "grad_norm": 126.64899063848813,
      "learning_rate": 3.1818181818181815e-07,
      "logits/chosen": -1.2423827648162842,
      "logits/rejected": -1.212158203125,
      "logps/chosen": -208.27499389648438,
      "logps/rejected": -230.75,
      "loss": 0.2873,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -1.522576928138733,
      "rewards/margins": 2.6622557640075684,
      "rewards/rejected": -4.184668064117432,
      "step": 75
    },
    {
      "epoch": 1.442696629213483,
      "grad_norm": 117.37315757338318,
      "learning_rate": 2.727272727272727e-07,
      "logits/chosen": -1.244140625,
      "logits/rejected": -1.20654296875,
      "logps/chosen": -203.5812530517578,
      "logps/rejected": -249.3249969482422,
      "loss": 0.3876,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -1.797338843345642,
      "rewards/margins": 2.134106397628784,
      "rewards/rejected": -3.930224657058716,
      "step": 80
    },
    {
      "epoch": 1.5325842696629213,
      "grad_norm": 133.10246641657412,
      "learning_rate": 2.2727272727272726e-07,
      "logits/chosen": -1.252050757408142,
      "logits/rejected": -1.2130858898162842,
      "logps/chosen": -202.4499969482422,
      "logps/rejected": -219.47500610351562,
      "loss": 0.3168,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": -1.588891625404358,
      "rewards/margins": 2.5704102516174316,
      "rewards/rejected": -4.15966796875,
      "step": 85
    },
    {
      "epoch": 1.6224719101123597,
      "grad_norm": 129.44212708304207,
      "learning_rate": 1.818181818181818e-07,
      "logits/chosen": -1.2580077648162842,
      "logits/rejected": -1.262597680091858,
      "logps/chosen": -165.0124969482422,
      "logps/rejected": -193.21249389648438,
      "loss": 0.2857,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -1.4154541492462158,
      "rewards/margins": 2.55810546875,
      "rewards/rejected": -3.975329637527466,
      "step": 90
    },
    {
      "epoch": 1.7123595505617977,
      "grad_norm": 133.23298622428365,
      "learning_rate": 1.3636363636363635e-07,
      "logits/chosen": -1.26171875,
      "logits/rejected": -1.2370116710662842,
      "logps/chosen": -191.8874969482422,
      "logps/rejected": -232.5124969482422,
      "loss": 0.2636,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": -1.913476586341858,
      "rewards/margins": 3.128124952316284,
      "rewards/rejected": -5.037988185882568,
      "step": 95
    },
    {
      "epoch": 1.802247191011236,
      "grad_norm": 121.7166467503501,
      "learning_rate": 9.09090909090909e-08,
      "logits/chosen": -1.2141602039337158,
      "logits/rejected": -1.228906273841858,
      "logps/chosen": -199.4375,
      "logps/rejected": -229.64999389648438,
      "loss": 0.2476,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -1.579687476158142,
      "rewards/margins": 2.887890577316284,
      "rewards/rejected": -4.467577934265137,
      "step": 100
    },
    {
      "epoch": 1.892134831460674,
      "grad_norm": 124.25297719784706,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": -1.291894555091858,
      "logits/rejected": -1.346093773841858,
      "logps/chosen": -214.0625,
      "logps/rejected": -233.0749969482422,
      "loss": 0.2455,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -2.6200804710388184,
      "rewards/margins": 2.843066453933716,
      "rewards/rejected": -5.462133884429932,
      "step": 105
    },
    {
      "epoch": 1.9820224719101125,
      "grad_norm": 113.18883753786767,
      "learning_rate": 0.0,
      "logits/chosen": -1.2438476085662842,
      "logits/rejected": -1.2707030773162842,
      "logps/chosen": -215.46249389648438,
      "logps/rejected": -218.7375030517578,
      "loss": 0.2322,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -2.182504177093506,
      "rewards/margins": 2.7933106422424316,
      "rewards/rejected": -4.975976467132568,
      "step": 110
    },
    {
      "epoch": 1.9820224719101125,
      "step": 110,
      "total_flos": 0.0,
      "train_loss": 0.340618884563446,
      "train_runtime": 1955.8801,
      "train_samples_per_second": 1.819,
      "train_steps_per_second": 0.056
    }
  ],
  "logging_steps": 5,
  "max_steps": 110,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}