{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.996510067114094,
  "eval_steps": 400,
  "global_step": 116,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008590604026845637,
      "grad_norm": 26.041106753964655,
      "learning_rate": 4.166666666666666e-08,
      "logits/chosen": -1.5759165287017822,
      "logits/rejected": -1.7932627201080322,
      "logps/chosen": -402.52313232421875,
      "logps/rejected": -1099.9710693359375,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.042953020134228186,
      "grad_norm": 31.18860734724652,
      "learning_rate": 2.0833333333333333e-07,
      "logits/chosen": -1.7433890104293823,
      "logits/rejected": -2.201554775238037,
      "logps/chosen": -531.8023681640625,
      "logps/rejected": -2555.797607421875,
      "loss": 0.6926,
      "rewards/accuracies": 0.3515625,
      "rewards/chosen": 0.0006599800544790924,
      "rewards/margins": 0.000775136286392808,
      "rewards/rejected": -0.00011515617370605469,
      "step": 5
    },
    {
      "epoch": 0.08590604026845637,
      "grad_norm": 27.29698099739781,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -1.6542924642562866,
      "logits/rejected": -2.1482906341552734,
      "logps/chosen": -526.591064453125,
      "logps/rejected": -2324.423828125,
      "loss": 0.6773,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.0068870047107338905,
      "rewards/margins": 0.027588676661252975,
      "rewards/rejected": -0.03447568044066429,
      "step": 10
    },
    {
      "epoch": 0.12885906040268458,
      "grad_norm": 19.203276482338538,
      "learning_rate": 4.989741394042727e-07,
      "logits/chosen": -1.768988847732544,
      "logits/rejected": -2.3421971797943115,
      "logps/chosen": -543.0718994140625,
      "logps/rejected": -2687.59716796875,
      "loss": 0.6056,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -0.043904103338718414,
      "rewards/margins": 0.24259641766548157,
      "rewards/rejected": -0.2865005135536194,
      "step": 15
    },
    {
      "epoch": 0.17181208053691274,
      "grad_norm": 10.652934421675868,
      "learning_rate": 4.92735454356513e-07,
      "logits/chosen": -1.692357063293457,
      "logits/rejected": -2.1176271438598633,
      "logps/chosen": -575.0624389648438,
      "logps/rejected": -2352.69873046875,
      "loss": 0.4794,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": -0.19454456865787506,
      "rewards/margins": 0.9079583287239075,
      "rewards/rejected": -1.1025029420852661,
      "step": 20
    },
    {
      "epoch": 0.21476510067114093,
      "grad_norm": 6.620549530322006,
      "learning_rate": 4.809698831278217e-07,
      "logits/chosen": -1.7178752422332764,
      "logits/rejected": -2.392866611480713,
      "logps/chosen": -683.6615600585938,
      "logps/rejected": -2998.378662109375,
      "loss": 0.4252,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -0.37876173853874207,
      "rewards/margins": 2.0470335483551025,
      "rewards/rejected": -2.425795078277588,
      "step": 25
    },
    {
      "epoch": 0.25771812080536916,
      "grad_norm": 13.984907018293939,
      "learning_rate": 4.639453180753619e-07,
      "logits/chosen": -1.6685655117034912,
      "logits/rejected": -2.3044064044952393,
      "logps/chosen": -616.4549560546875,
      "logps/rejected": -2758.712646484375,
      "loss": 0.3458,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -0.731316864490509,
      "rewards/margins": 3.5061488151550293,
      "rewards/rejected": -4.237465858459473,
      "step": 30
    },
    {
      "epoch": 0.3006711409395973,
      "grad_norm": 12.563360315413236,
      "learning_rate": 4.420493945100701e-07,
      "logits/chosen": -1.7690975666046143,
      "logits/rejected": -2.2737460136413574,
      "logps/chosen": -740.88134765625,
      "logps/rejected": -3318.674560546875,
      "loss": 0.3563,
      "rewards/accuracies": 0.875,
      "rewards/chosen": -1.2466011047363281,
      "rewards/margins": 4.410061359405518,
      "rewards/rejected": -5.656662940979004,
      "step": 35
    },
    {
      "epoch": 0.3436241610738255,
      "grad_norm": 58.638493321053595,
      "learning_rate": 4.157806645601988e-07,
      "logits/chosen": -1.5280271768569946,
      "logits/rejected": -1.8888028860092163,
      "logps/chosen": -631.5662231445312,
      "logps/rejected": -3006.968017578125,
      "loss": 0.2634,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -1.5436829328536987,
      "rewards/margins": 4.522477626800537,
      "rewards/rejected": -6.066160678863525,
      "step": 40
    },
    {
      "epoch": 0.3865771812080537,
      "grad_norm": 34.148235602710315,
      "learning_rate": 3.857372455503697e-07,
      "logits/chosen": -1.418347716331482,
      "logits/rejected": -1.708924651145935,
      "logps/chosen": -827.2987060546875,
      "logps/rejected": -3175.490966796875,
      "loss": 0.2708,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -2.246793270111084,
      "rewards/margins": 5.506977081298828,
      "rewards/rejected": -7.753770351409912,
      "step": 45
    },
    {
      "epoch": 0.42953020134228187,
      "grad_norm": 75.04768803288361,
      "learning_rate": 3.5260320136318924e-07,
      "logits/chosen": -1.4919449090957642,
      "logits/rejected": -1.7019599676132202,
      "logps/chosen": -761.5653686523438,
      "logps/rejected": -3045.13427734375,
      "loss": 0.2368,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -2.335999011993408,
      "rewards/margins": 5.300050258636475,
      "rewards/rejected": -7.636049747467041,
      "step": 50
    },
    {
      "epoch": 0.47248322147651006,
      "grad_norm": 30.998467510893132,
      "learning_rate": 3.171329668685942e-07,
      "logits/chosen": -1.581739068031311,
      "logits/rejected": -1.926107406616211,
      "logps/chosen": -772.8366088867188,
      "logps/rejected": -3056.0234375,
      "loss": 0.2615,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -1.8509727716445923,
      "rewards/margins": 4.722422122955322,
      "rewards/rejected": -6.573394775390625,
      "step": 55
    },
    {
      "epoch": 0.5154362416107383,
      "grad_norm": 18.56633622085772,
      "learning_rate": 2.801341700638307e-07,
      "logits/chosen": -1.4350240230560303,
      "logits/rejected": -1.619866967201233,
      "logps/chosen": -880.1119995117188,
      "logps/rejected": -3234.364013671875,
      "loss": 0.194,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -2.119612216949463,
      "rewards/margins": 6.008899688720703,
      "rewards/rejected": -8.128512382507324,
      "step": 60
    },
    {
      "epoch": 0.5583892617449664,
      "grad_norm": 39.169508412837246,
      "learning_rate": 2.424492430497778e-07,
      "logits/chosen": -1.7028038501739502,
      "logits/rejected": -1.7408069372177124,
      "logps/chosen": -853.0597534179688,
      "logps/rejected": -3378.33642578125,
      "loss": 0.1911,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -2.2176003456115723,
      "rewards/margins": 6.097519874572754,
      "rewards/rejected": -8.315119743347168,
      "step": 65
    },
    {
      "epoch": 0.6013422818791946,
      "grad_norm": 45.21213119496434,
      "learning_rate": 2.0493624054652355e-07,
      "logits/chosen": -1.540661096572876,
      "logits/rejected": -1.7918895483016968,
      "logps/chosen": -935.87939453125,
      "logps/rejected": -3163.64208984375,
      "loss": 0.2123,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -2.6157093048095703,
      "rewards/margins": 5.561184883117676,
      "rewards/rejected": -8.176894187927246,
      "step": 70
    },
    {
      "epoch": 0.6442953020134228,
      "grad_norm": 62.20110512263006,
      "learning_rate": 1.6844930269478273e-07,
      "logits/chosen": -1.660971999168396,
      "logits/rejected": -1.7829008102416992,
      "logps/chosen": -786.4994506835938,
      "logps/rejected": -3442.186279296875,
      "loss": 0.1726,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -1.9392259120941162,
      "rewards/margins": 7.3069915771484375,
      "rewards/rejected": -9.246217727661133,
      "step": 75
    },
    {
      "epoch": 0.687248322147651,
      "grad_norm": 12.849059907790142,
      "learning_rate": 1.3381920698905784e-07,
      "logits/chosen": -1.7256042957305908,
      "logits/rejected": -1.6946690082550049,
      "logps/chosen": -784.163330078125,
      "logps/rejected": -3483.552734375,
      "loss": 0.228,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -2.541985273361206,
      "rewards/margins": 7.170411109924316,
      "rewards/rejected": -9.712396621704102,
      "step": 80
    },
    {
      "epoch": 0.7302013422818792,
      "grad_norm": 26.23757128264686,
      "learning_rate": 1.0183445215899584e-07,
      "logits/chosen": -1.7951923608779907,
      "logits/rejected": -1.84537672996521,
      "logps/chosen": -798.8233642578125,
      "logps/rejected": -3609.947265625,
      "loss": 0.1788,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -2.160109043121338,
      "rewards/margins": 5.754651069641113,
      "rewards/rejected": -7.914759635925293,
      "step": 85
    },
    {
      "epoch": 0.7731543624161074,
      "grad_norm": 10.192170734753343,
      "learning_rate": 7.322330470336313e-08,
      "logits/chosen": -1.6331850290298462,
      "logits/rejected": -1.5111037492752075,
      "logps/chosen": -793.5105590820312,
      "logps/rejected": -2994.8486328125,
      "loss": 0.1363,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -2.2809009552001953,
      "rewards/margins": 6.116978168487549,
      "rewards/rejected": -8.397878646850586,
      "step": 90
    },
    {
      "epoch": 0.8161073825503355,
      "grad_norm": 25.55626868432015,
      "learning_rate": 4.863721686226349e-08,
      "logits/chosen": -1.4511371850967407,
      "logits/rejected": -1.4803770780563354,
      "logps/chosen": -715.542236328125,
      "logps/rejected": -3007.235107421875,
      "loss": 0.2881,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -2.2814974784851074,
      "rewards/margins": 7.453909397125244,
      "rewards/rejected": -9.735406875610352,
      "step": 95
    },
    {
      "epoch": 0.8590604026845637,
      "grad_norm": 50.01533007273521,
      "learning_rate": 2.863599358669755e-08,
      "logits/chosen": -1.7344436645507812,
      "logits/rejected": -1.474273681640625,
      "logps/chosen": -685.488525390625,
      "logps/rejected": -3361.528564453125,
      "loss": 0.2129,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -2.257279872894287,
      "rewards/margins": 8.946730613708496,
      "rewards/rejected": -11.204010963439941,
      "step": 100
    },
    {
      "epoch": 0.9020134228187919,
      "grad_norm": 31.116313265685257,
      "learning_rate": 1.3675046241339916e-08,
      "logits/chosen": -1.6095584630966187,
      "logits/rejected": -1.4253101348876953,
      "logps/chosen": -858.84130859375,
      "logps/rejected": -3048.67578125,
      "loss": 0.2266,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -2.7480015754699707,
      "rewards/margins": 6.334467887878418,
      "rewards/rejected": -9.082470893859863,
      "step": 105
    },
    {
      "epoch": 0.9449664429530201,
      "grad_norm": 24.56992314384152,
      "learning_rate": 4.0950232632141205e-09,
      "logits/chosen": -1.5636495351791382,
      "logits/rejected": -1.5311402082443237,
      "logps/chosen": -720.26708984375,
      "logps/rejected": -3027.19873046875,
      "loss": 0.2279,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -2.2057013511657715,
      "rewards/margins": 7.12615442276001,
      "rewards/rejected": -9.331854820251465,
      "step": 110
    },
    {
      "epoch": 0.9879194630872483,
      "grad_norm": 24.93170608381977,
      "learning_rate": 1.1405387761664887e-10,
      "logits/chosen": -1.6431703567504883,
      "logits/rejected": -1.5806033611297607,
      "logps/chosen": -681.1459350585938,
      "logps/rejected": -3343.632080078125,
      "loss": 0.1556,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -1.9711568355560303,
      "rewards/margins": 8.062986373901367,
      "rewards/rejected": -10.034143447875977,
      "step": 115
    },
    {
      "epoch": 0.996510067114094,
      "step": 116,
      "total_flos": 0.0,
      "train_loss": 0.3046778245218869,
      "train_runtime": 5634.2561,
      "train_samples_per_second": 2.644,
      "train_steps_per_second": 0.021
    }
  ],
  "logging_steps": 5,
  "max_steps": 116,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000000,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}