{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 1000,
  "global_step": 125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 5.49035456239289,
      "learning_rate": 1.5384615384615387e-05,
      "logits/chosen": -1.73323655128479,
      "logits/rejected": -1.963712453842163,
      "logps/chosen": -64.71795654296875,
      "logps/rejected": -92.56527709960938,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.4,
      "grad_norm": 28.445585026780677,
      "learning_rate": 0.00015384615384615385,
      "logits/chosen": -1.9578092098236084,
      "logits/rejected": -2.1086513996124268,
      "logps/chosen": -241.50640869140625,
      "logps/rejected": -268.8847351074219,
      "loss": 1.1389,
      "rewards/accuracies": 0.4722222089767456,
      "rewards/chosen": -1.7811455726623535,
      "rewards/margins": 0.029786672443151474,
      "rewards/rejected": -1.8109323978424072,
      "step": 10
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.507863124304794,
      "learning_rate": 0.00019807852804032305,
      "logits/chosen": -2.51621675491333,
      "logits/rejected": -2.6432676315307617,
      "logps/chosen": -570.351806640625,
      "logps/rejected": -496.91827392578125,
      "loss": 1.8146,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": -5.092972755432129,
      "rewards/margins": -1.0333130359649658,
      "rewards/rejected": -4.059659957885742,
      "step": 20
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.9111283957421439,
      "learning_rate": 0.00018884456359788724,
      "logits/chosen": -3.369929790496826,
      "logits/rejected": -3.283642292022705,
      "logps/chosen": -507.5963439941406,
      "logps/rejected": -572.895263671875,
      "loss": 0.7716,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -4.45947265625,
      "rewards/margins": 0.3659425377845764,
      "rewards/rejected": -4.825415134429932,
      "step": 30
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.795612553020093,
      "learning_rate": 0.0001726660322034027,
      "logits/chosen": -4.420574188232422,
      "logits/rejected": -4.248013496398926,
      "logps/chosen": -548.0731811523438,
      "logps/rejected": -683.3683471679688,
      "loss": 0.639,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -4.906264305114746,
      "rewards/margins": 0.9860424995422363,
      "rewards/rejected": -5.892307758331299,
      "step": 40
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8538474858298817,
      "learning_rate": 0.00015080753452465296,
      "logits/chosen": -5.002673149108887,
      "logits/rejected": -4.903468132019043,
      "logps/chosen": -598.4771728515625,
      "logps/rejected": -708.3795166015625,
      "loss": 0.6112,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -5.362724781036377,
      "rewards/margins": 0.840530514717102,
      "rewards/rejected": -6.203255653381348,
      "step": 50
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.6914654231679778,
      "learning_rate": 0.0001249776478167227,
      "logits/chosen": -5.280072212219238,
      "logits/rejected": -5.135374069213867,
      "logps/chosen": -556.451904296875,
      "logps/rejected": -724.9661254882812,
      "loss": 0.494,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -4.981103420257568,
      "rewards/margins": 1.3216989040374756,
      "rewards/rejected": -6.302803039550781,
      "step": 60
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.8698084153477929,
      "learning_rate": 9.719537437241312e-05,
      "logits/chosen": -5.114866733551025,
      "logits/rejected": -5.01845121383667,
      "logps/chosen": -614.8002319335938,
      "logps/rejected": -744.7828369140625,
      "loss": 0.5243,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -5.515006065368652,
      "rewards/margins": 1.0367931127548218,
      "rewards/rejected": -6.5517988204956055,
      "step": 70
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.7549954414499753,
      "learning_rate": 6.963232548903853e-05,
      "logits/chosen": -4.472952365875244,
      "logits/rejected": -4.432765007019043,
      "logps/chosen": -761.5465087890625,
      "logps/rejected": -860.75537109375,
      "loss": 1.6723,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -7.0114946365356445,
      "rewards/margins": 0.6902900338172913,
      "rewards/rejected": -7.701784610748291,
      "step": 80
    },
    {
      "epoch": 3.6,
      "grad_norm": 1.1425715631983737,
      "learning_rate": 4.444297669803981e-05,
      "logits/chosen": -5.620628356933594,
      "logits/rejected": -5.535118103027344,
      "logps/chosen": -655.6566772460938,
      "logps/rejected": -807.3014526367188,
      "loss": 0.4738,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -5.936182498931885,
      "rewards/margins": 1.246588945388794,
      "rewards/rejected": -7.1827712059021,
      "step": 90
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9528369738814884,
      "learning_rate": 2.3596262417839255e-05,
      "logits/chosen": -5.550174236297607,
      "logits/rejected": -5.511833667755127,
      "logps/chosen": -628.8092041015625,
      "logps/rejected": -796.935302734375,
      "loss": 0.449,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -5.668156623840332,
      "rewards/margins": 1.3851256370544434,
      "rewards/rejected": -7.053282260894775,
      "step": 100
    },
    {
      "epoch": 4.4,
      "grad_norm": 2.890782481250775,
      "learning_rate": 8.72167349386811e-06,
      "logits/chosen": -5.907111167907715,
      "logits/rejected": -5.945315361022949,
      "logps/chosen": -695.7470092773438,
      "logps/rejected": -886.4703369140625,
      "loss": 0.3633,
      "rewards/accuracies": 0.856249988079071,
      "rewards/chosen": -6.3334550857543945,
      "rewards/margins": 1.6471582651138306,
      "rewards/rejected": -7.980613708496094,
      "step": 110
    },
    {
      "epoch": 4.8,
      "grad_norm": 0.7286712015593821,
      "learning_rate": 9.818874663554357e-07,
      "logits/chosen": -5.903909683227539,
      "logits/rejected": -5.952976226806641,
      "logps/chosen": -664.2296752929688,
      "logps/rejected": -862.1025390625,
      "loss": 0.3324,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -6.032934665679932,
      "rewards/margins": 1.685362458229065,
      "rewards/rejected": -7.718297004699707,
      "step": 120
    },
    {
      "epoch": 5.0,
      "step": 125,
      "total_flos": 0.0,
      "train_loss": 0.7484294853210449,
      "train_runtime": 12665.918,
      "train_samples_per_second": 0.632,
      "train_steps_per_second": 0.01
    }
  ],
  "logging_steps": 10,
  "max_steps": 125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}