{
  "best_metric": 0.6840614080429077,
  "best_model_checkpoint": "./output/checkpoints/2024-05-27_09-01-43/checkpoint-100",
  "epoch": 0.07867820613690008,
  "eval_steps": 100,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003933910306845004,
      "grad_norm": 33.997314453125,
      "learning_rate": 1.25e-06,
      "logits/chosen": -0.2329253852367401,
      "logits/rejected": -0.7133080363273621,
      "logps/chosen": -206.9918670654297,
      "logps/rejected": -177.71676635742188,
      "loss": 0.6965,
      "rewards/accuracies": 0.16249999403953552,
      "rewards/chosen": -0.0024444584269076586,
      "rewards/margins": -0.0065179443918168545,
      "rewards/rejected": 0.004073486663401127,
      "step": 5
    },
    {
      "epoch": 0.007867820613690008,
      "grad_norm": 33.46182632446289,
      "learning_rate": 2.8125e-06,
      "logits/chosen": -0.39770540595054626,
      "logits/rejected": -0.7366135120391846,
      "logps/chosen": -200.96145629882812,
      "logps/rejected": -177.04241943359375,
      "loss": 0.6857,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": 0.028690528124570847,
      "rewards/margins": 0.01567094773054123,
      "rewards/rejected": 0.013019581325352192,
      "step": 10
    },
    {
      "epoch": 0.011801730920535013,
      "grad_norm": 36.08251953125,
      "learning_rate": 4.3750000000000005e-06,
      "logits/chosen": -0.35487601161003113,
      "logits/rejected": -0.6588561534881592,
      "logps/chosen": -217.158203125,
      "logps/rejected": -193.87928771972656,
      "loss": 0.6847,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.1420166790485382,
      "rewards/margins": 0.021750029176473618,
      "rewards/rejected": 0.12026665359735489,
      "step": 15
    },
    {
      "epoch": 0.015735641227380016,
      "grad_norm": 31.359071731567383,
      "learning_rate": 5.9375e-06,
      "logits/chosen": -0.38902169466018677,
      "logits/rejected": -0.7328646779060364,
      "logps/chosen": -208.9059600830078,
      "logps/rejected": -179.52752685546875,
      "loss": 0.6636,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.23876996338367462,
      "rewards/margins": 0.07814005762338638,
      "rewards/rejected": 0.16062989830970764,
      "step": 20
    },
    {
      "epoch": 0.01966955153422502,
      "grad_norm": 37.41947555541992,
      "learning_rate": 7.500000000000001e-06,
      "logits/chosen": -0.24571442604064941,
      "logits/rejected": -0.7017894983291626,
      "logps/chosen": -195.68539428710938,
      "logps/rejected": -178.2770233154297,
      "loss": 0.6703,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.38804444670677185,
      "rewards/margins": 0.09177973121404648,
      "rewards/rejected": 0.2962647080421448,
      "step": 25
    },
    {
      "epoch": 0.023603461841070025,
      "grad_norm": 36.168155670166016,
      "learning_rate": 9.0625e-06,
      "logits/chosen": -0.3279009461402893,
      "logits/rejected": -0.62000572681427,
      "logps/chosen": -207.6711883544922,
      "logps/rejected": -183.57412719726562,
      "loss": 0.6945,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.5336718559265137,
      "rewards/margins": 0.08061937242746353,
      "rewards/rejected": 0.45305246114730835,
      "step": 30
    },
    {
      "epoch": 0.02753737214791503,
      "grad_norm": 33.33737564086914,
      "learning_rate": 1.0625e-05,
      "logits/chosen": -0.4809085428714752,
      "logits/rejected": -0.758043646812439,
      "logps/chosen": -220.7525634765625,
      "logps/rejected": -198.58531188964844,
      "loss": 0.6564,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.8875558972358704,
      "rewards/margins": 0.20188398659229279,
      "rewards/rejected": 0.6856719255447388,
      "step": 35
    },
    {
      "epoch": 0.03147128245476003,
      "grad_norm": 32.628173828125,
      "learning_rate": 1.2187500000000001e-05,
      "logits/chosen": -0.34948527812957764,
      "logits/rejected": -0.7154799699783325,
      "logps/chosen": -195.3294677734375,
      "logps/rejected": -178.76821899414062,
      "loss": 0.6995,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.8487609624862671,
      "rewards/margins": 0.11841963231563568,
      "rewards/rejected": 0.7303413152694702,
      "step": 40
    },
    {
      "epoch": 0.03540519276160504,
      "grad_norm": 34.517208099365234,
      "learning_rate": 1.3125e-05,
      "logits/chosen": -0.6799963712692261,
      "logits/rejected": -1.0354492664337158,
      "logps/chosen": -196.0514373779297,
      "logps/rejected": -165.4368438720703,
      "loss": 0.6874,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": 0.6915052533149719,
      "rewards/margins": 0.18386030197143555,
      "rewards/rejected": 0.5076450109481812,
      "step": 45
    },
    {
      "epoch": 0.03933910306845004,
      "grad_norm": 39.48639678955078,
      "learning_rate": 1.4687500000000001e-05,
      "logits/chosen": -0.22616450488567352,
      "logits/rejected": -0.3378121256828308,
      "logps/chosen": -210.38534545898438,
      "logps/rejected": -205.64596557617188,
      "loss": 0.6612,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.6935254335403442,
      "rewards/margins": 0.17534935474395752,
      "rewards/rejected": 0.5181760191917419,
      "step": 50
    },
    {
      "epoch": 0.043273013375295044,
      "grad_norm": 54.90278244018555,
      "learning_rate": 1.6250000000000002e-05,
      "logits/chosen": -0.5758259892463684,
      "logits/rejected": -0.7866605520248413,
      "logps/chosen": -194.25332641601562,
      "logps/rejected": -179.38743591308594,
      "loss": 0.7132,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.7601491212844849,
      "rewards/margins": 0.13087329268455505,
      "rewards/rejected": 0.629275918006897,
      "step": 55
    },
    {
      "epoch": 0.04720692368214005,
      "grad_norm": 31.500804901123047,
      "learning_rate": 1.7500000000000002e-05,
      "logits/chosen": -0.3127598762512207,
      "logits/rejected": -0.7857316136360168,
      "logps/chosen": -214.54507446289062,
      "logps/rejected": -167.3846435546875,
      "loss": 0.659,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.8511220216751099,
      "rewards/margins": 0.3019554316997528,
      "rewards/rejected": 0.5491665601730347,
      "step": 60
    },
    {
      "epoch": 0.05114083398898505,
      "grad_norm": 39.18369674682617,
      "learning_rate": 1.9062500000000003e-05,
      "logits/chosen": -0.4257170557975769,
      "logits/rejected": -0.7206593751907349,
      "logps/chosen": -212.1534881591797,
      "logps/rejected": -198.5198516845703,
      "loss": 0.6603,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 1.0609362125396729,
      "rewards/margins": 0.4154728055000305,
      "rewards/rejected": 0.6454635858535767,
      "step": 65
    },
    {
      "epoch": 0.05507474429583006,
      "grad_norm": 23.921350479125977,
      "learning_rate": 2.0625000000000003e-05,
      "logits/chosen": -0.4161883294582367,
      "logits/rejected": -0.7778038382530212,
      "logps/chosen": -211.8742218017578,
      "logps/rejected": -173.7216033935547,
      "loss": 0.6306,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 1.1546392440795898,
      "rewards/margins": 0.4130212366580963,
      "rewards/rejected": 0.7416179180145264,
      "step": 70
    },
    {
      "epoch": 0.059008654602675056,
      "grad_norm": 33.804962158203125,
      "learning_rate": 2.21875e-05,
      "logits/chosen": -0.11207835376262665,
      "logits/rejected": -0.5466287136077881,
      "logps/chosen": -206.02218627929688,
      "logps/rejected": -185.80941772460938,
      "loss": 0.719,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": 1.1464534997940063,
      "rewards/margins": 0.2297494113445282,
      "rewards/rejected": 0.9167040586471558,
      "step": 75
    },
    {
      "epoch": 0.06294256490952006,
      "grad_norm": 42.39493942260742,
      "learning_rate": 2.375e-05,
      "logits/chosen": -0.2344416081905365,
      "logits/rejected": -0.49077630043029785,
      "logps/chosen": -209.4600067138672,
      "logps/rejected": -198.47911071777344,
      "loss": 0.663,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": 0.7387064099311829,
      "rewards/margins": 0.3134633004665375,
      "rewards/rejected": 0.4252430498600006,
      "step": 80
    },
    {
      "epoch": 0.06687647521636507,
      "grad_norm": 35.38875961303711,
      "learning_rate": 2.5312500000000002e-05,
      "logits/chosen": -0.26731210947036743,
      "logits/rejected": -0.7659153342247009,
      "logps/chosen": -206.7689666748047,
      "logps/rejected": -178.71986389160156,
      "loss": 0.7082,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.22556963562965393,
      "rewards/margins": 0.251675546169281,
      "rewards/rejected": -0.026105916127562523,
      "step": 85
    },
    {
      "epoch": 0.07081038552321008,
      "grad_norm": 30.606552124023438,
      "learning_rate": 2.6875000000000003e-05,
      "logits/chosen": -0.34513598680496216,
      "logits/rejected": -0.5674473643302917,
      "logps/chosen": -208.49002075195312,
      "logps/rejected": -184.6549072265625,
      "loss": 0.7702,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.2557719945907593,
      "rewards/margins": 0.22060665488243103,
      "rewards/rejected": 0.03516533225774765,
      "step": 90
    },
    {
      "epoch": 0.07474429583005507,
      "grad_norm": 25.326486587524414,
      "learning_rate": 2.8437500000000003e-05,
      "logits/chosen": -0.21977896988391876,
      "logits/rejected": -0.6439090371131897,
      "logps/chosen": -201.21401977539062,
      "logps/rejected": -183.35049438476562,
      "loss": 0.713,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": 0.6074048280715942,
      "rewards/margins": 0.23941688239574432,
      "rewards/rejected": 0.3679879307746887,
      "step": 95
    },
    {
      "epoch": 0.07867820613690008,
      "grad_norm": 34.922325134277344,
      "learning_rate": 3.0000000000000004e-05,
      "logits/chosen": -0.4304371774196625,
      "logits/rejected": -0.7793359756469727,
      "logps/chosen": -207.411376953125,
      "logps/rejected": -178.08729553222656,
      "loss": 0.7514,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.6943355798721313,
      "rewards/margins": 0.1873859167098999,
      "rewards/rejected": 0.5069497227668762,
      "step": 100
    },
    {
      "epoch": 0.07867820613690008,
      "eval_logits/chosen": 1.2959624528884888,
      "eval_logits/rejected": 1.0611025094985962,
      "eval_logps/chosen": -205.19613647460938,
      "eval_logps/rejected": -178.5214385986328,
      "eval_loss": 0.6840614080429077,
      "eval_rewards/accuracies": 0.628125011920929,
      "eval_rewards/chosen": 1.1096659898757935,
      "eval_rewards/margins": 0.36510738730430603,
      "eval_rewards/rejected": 0.7445584535598755,
      "eval_runtime": 248.7873,
      "eval_samples_per_second": 2.572,
      "eval_steps_per_second": 0.161,
      "step": 100
    }
  ],
  "logging_steps": 5,
  "max_steps": 1271,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}