{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.995910949568378,
  "eval_steps": 400,
  "global_step": 137,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03634711494775102,
      "grad_norm": 262.34647903554514,
      "learning_rate": 2.857142857142857e-07,
      "logits/chosen": -9.14239501953125,
      "logits/rejected": -9.232463836669922,
      "logps/chosen": -1.7616302967071533,
      "logps/rejected": -1.7519614696502686,
      "loss": 7.6689,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -17.61629867553711,
      "rewards/margins": -0.09668443351984024,
      "rewards/rejected": -17.519615173339844,
      "step": 5
    },
    {
      "epoch": 0.07269422989550205,
      "grad_norm": 160.26775090527158,
      "learning_rate": 5.714285714285714e-07,
      "logits/chosen": -9.558967590332031,
      "logits/rejected": -9.606477737426758,
      "logps/chosen": -1.4764775037765503,
      "logps/rejected": -1.439325213432312,
      "loss": 6.6349,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -14.764776229858398,
      "rewards/margins": -0.3715229332447052,
      "rewards/rejected": -14.3932523727417,
      "step": 10
    },
    {
      "epoch": 0.10904134484325306,
      "grad_norm": 120.66971185794887,
      "learning_rate": 7.998695344323425e-07,
      "logits/chosen": -8.532218933105469,
      "logits/rejected": -8.379935264587402,
      "logps/chosen": -1.2477877140045166,
      "logps/rejected": -1.3596165180206299,
      "loss": 6.3247,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -12.477876663208008,
      "rewards/margins": 1.1182864904403687,
      "rewards/rejected": -13.596163749694824,
      "step": 15
    },
    {
      "epoch": 0.1453884597910041,
      "grad_norm": 78.51354956499227,
      "learning_rate": 7.953121695121394e-07,
      "logits/chosen": -7.22064208984375,
      "logits/rejected": -6.8809967041015625,
      "logps/chosen": -0.9391318559646606,
      "logps/rejected": -1.0672681331634521,
      "loss": 4.9513,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -9.391318321228027,
      "rewards/margins": 1.2813628911972046,
      "rewards/rejected": -10.67268180847168,
      "step": 20
    },
    {
      "epoch": 0.18173557473875512,
      "grad_norm": 73.47123034470297,
      "learning_rate": 7.843163833184991e-07,
      "logits/chosen": -6.873021125793457,
      "logits/rejected": -6.335709095001221,
      "logps/chosen": -0.7527015209197998,
      "logps/rejected": -0.8773846626281738,
      "loss": 4.5988,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -7.527015686035156,
      "rewards/margins": 1.246830701828003,
      "rewards/rejected": -8.773846626281738,
      "step": 25
    },
    {
      "epoch": 0.21808268968650613,
      "grad_norm": 77.81006578282152,
      "learning_rate": 7.670612634414511e-07,
      "logits/chosen": -6.2256245613098145,
      "logits/rejected": -6.002863883972168,
      "logps/chosen": -0.7618826031684875,
      "logps/rejected": -0.9232344627380371,
      "loss": 4.4284,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -7.618826389312744,
      "rewards/margins": 1.613519310951233,
      "rewards/rejected": -9.232345581054688,
      "step": 30
    },
    {
      "epoch": 0.25442980463425713,
      "grad_norm": 61.664034648588775,
      "learning_rate": 7.438278427948805e-07,
      "logits/chosen": -6.400006294250488,
      "logits/rejected": -6.255277633666992,
      "logps/chosen": -0.7186794877052307,
      "logps/rejected": -0.9298878908157349,
      "loss": 4.3183,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -7.186795234680176,
      "rewards/margins": 2.1120834350585938,
      "rewards/rejected": -9.29887866973877,
      "step": 35
    },
    {
      "epoch": 0.2907769195820082,
      "grad_norm": 68.56556472157719,
      "learning_rate": 7.149945224533862e-07,
      "logits/chosen": -5.755168914794922,
      "logits/rejected": -5.6258063316345215,
      "logps/chosen": -0.8867130279541016,
      "logps/rejected": -1.0443212985992432,
      "loss": 4.4094,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -8.8671293258667,
      "rewards/margins": 1.5760822296142578,
      "rewards/rejected": -10.443211555480957,
      "step": 40
    },
    {
      "epoch": 0.3271240345297592,
      "grad_norm": 75.32781896911939,
      "learning_rate": 6.810309086608129e-07,
      "logits/chosen": -6.590113639831543,
      "logits/rejected": -6.312002658843994,
      "logps/chosen": -0.8065680265426636,
      "logps/rejected": -1.0438225269317627,
      "loss": 4.1676,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -8.065679550170898,
      "rewards/margins": 2.3725459575653076,
      "rewards/rejected": -10.438226699829102,
      "step": 45
    },
    {
      "epoch": 0.36347114947751025,
      "grad_norm": 74.83324411670193,
      "learning_rate": 6.424901643866552e-07,
      "logits/chosen": -6.4003729820251465,
      "logits/rejected": -6.2513837814331055,
      "logps/chosen": -0.7952122688293457,
      "logps/rejected": -1.0395910739898682,
      "loss": 4.1941,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -7.952122688293457,
      "rewards/margins": 2.443786144256592,
      "rewards/rejected": -10.39590835571289,
      "step": 50
    },
    {
      "epoch": 0.39981826442526125,
      "grad_norm": 59.64575454875869,
      "learning_rate": 6e-07,
      "logits/chosen": -6.756447792053223,
      "logits/rejected": -6.111125946044922,
      "logps/chosen": -0.8663609623908997,
      "logps/rejected": -1.0826358795166016,
      "loss": 3.9943,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -8.663609504699707,
      "rewards/margins": 2.162750005722046,
      "rewards/rejected": -10.826359748840332,
      "step": 55
    },
    {
      "epoch": 0.43616537937301225,
      "grad_norm": 67.48524030832144,
      "learning_rate": 5.542524497952543e-07,
      "logits/chosen": -6.575152397155762,
      "logits/rejected": -6.2960124015808105,
      "logps/chosen": -0.8099439740180969,
      "logps/rejected": -1.0211912393569946,
      "loss": 3.7201,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -8.09943962097168,
      "rewards/margins": 2.1124720573425293,
      "rewards/rejected": -10.211912155151367,
      "step": 60
    },
    {
      "epoch": 0.4725124943207633,
      "grad_norm": 82.16104654213201,
      "learning_rate": 5.059926008786647e-07,
      "logits/chosen": -6.274717330932617,
      "logits/rejected": -6.062811851501465,
      "logps/chosen": -0.9238764643669128,
      "logps/rejected": -1.0822433233261108,
      "loss": 4.027,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -9.238764762878418,
      "rewards/margins": 1.5836679935455322,
      "rewards/rejected": -10.822432518005371,
      "step": 65
    },
    {
      "epoch": 0.5088596092685143,
      "grad_norm": 69.13096378464603,
      "learning_rate": 4.5600645798745166e-07,
      "logits/chosen": -6.3393049240112305,
      "logits/rejected": -6.236004829406738,
      "logps/chosen": -0.7679723501205444,
      "logps/rejected": -0.968636155128479,
      "loss": 3.7213,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -7.679723262786865,
      "rewards/margins": 2.006639003753662,
      "rewards/rejected": -9.686361312866211,
      "step": 70
    },
    {
      "epoch": 0.5452067242162654,
      "grad_norm": 82.34301802283626,
      "learning_rate": 4.051081418863895e-07,
      "logits/chosen": -6.352741241455078,
      "logits/rejected": -6.1597676277160645,
      "logps/chosen": -0.7970303297042847,
      "logps/rejected": -1.043393850326538,
      "loss": 3.7425,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -7.970303535461426,
      "rewards/margins": 2.4636356830596924,
      "rewards/rejected": -10.433938980102539,
      "step": 75
    },
    {
      "epoch": 0.5815538391640164,
      "grad_norm": 72.13584102425693,
      "learning_rate": 3.541266298406398e-07,
      "logits/chosen": -6.617664337158203,
      "logits/rejected": -6.41934871673584,
      "logps/chosen": -0.8421775698661804,
      "logps/rejected": -1.0213863849639893,
      "loss": 3.7518,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -8.42177677154541,
      "rewards/margins": 1.7920877933502197,
      "rewards/rejected": -10.213863372802734,
      "step": 80
    },
    {
      "epoch": 0.6179009541117674,
      "grad_norm": 92.05622250652328,
      "learning_rate": 3.0389225412181565e-07,
      "logits/chosen": -6.101731300354004,
      "logits/rejected": -5.849137306213379,
      "logps/chosen": -0.7452789545059204,
      "logps/rejected": -1.045109748840332,
      "loss": 3.5802,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -7.452790260314941,
      "rewards/margins": 2.998307466506958,
      "rewards/rejected": -10.45109748840332,
      "step": 85
    },
    {
      "epoch": 0.6542480690595184,
      "grad_norm": 74.65829731544908,
      "learning_rate": 2.5522317844515273e-07,
      "logits/chosen": -6.542179107666016,
      "logits/rejected": -6.144691467285156,
      "logps/chosen": -0.8069820404052734,
      "logps/rejected": -1.0035241842269897,
      "loss": 3.705,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -8.069819450378418,
      "rewards/margins": 1.9654232263565063,
      "rewards/rejected": -10.035242080688477,
      "step": 90
    },
    {
      "epoch": 0.6905951840072694,
      "grad_norm": 75.40130974842236,
      "learning_rate": 2.0891207259509476e-07,
      "logits/chosen": -6.643575191497803,
      "logits/rejected": -6.608299255371094,
      "logps/chosen": -0.8557837605476379,
      "logps/rejected": -1.0176724195480347,
      "loss": 3.7581,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -8.557836532592773,
      "rewards/margins": 1.6188881397247314,
      "rewards/rejected": -10.176724433898926,
      "step": 95
    },
    {
      "epoch": 0.7269422989550205,
      "grad_norm": 83.85915732184031,
      "learning_rate": 1.6571320226872206e-07,
      "logits/chosen": -6.411711692810059,
      "logits/rejected": -6.091231346130371,
      "logps/chosen": -0.8256447911262512,
      "logps/rejected": -1.1177527904510498,
      "loss": 3.7165,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -8.256447792053223,
      "rewards/margins": 2.9210803508758545,
      "rewards/rejected": -11.177528381347656,
      "step": 100
    },
    {
      "epoch": 0.7632894139027715,
      "grad_norm": 102.74344975744411,
      "learning_rate": 1.2633014440382787e-07,
      "logits/chosen": -7.004655361175537,
      "logits/rejected": -6.623693943023682,
      "logps/chosen": -0.8427413105964661,
      "logps/rejected": -1.0215742588043213,
      "loss": 3.8455,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -8.427413940429688,
      "rewards/margins": 1.7883294820785522,
      "rewards/rejected": -10.215742111206055,
      "step": 105
    },
    {
      "epoch": 0.7996365288505225,
      "grad_norm": 84.46889768857108,
      "learning_rate": 9.14043280712228e-08,
      "logits/chosen": -6.523610591888428,
      "logits/rejected": -6.330638408660889,
      "logps/chosen": -0.9024786949157715,
      "logps/rejected": -1.0619735717773438,
      "loss": 3.9125,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -9.024787902832031,
      "rewards/margins": 1.5949490070343018,
      "rewards/rejected": -10.619735717773438,
      "step": 110
    },
    {
      "epoch": 0.8359836437982735,
      "grad_norm": 80.56642671916686,
      "learning_rate": 6.150458756494239e-08,
      "logits/chosen": -7.126093864440918,
      "logits/rejected": -6.942577362060547,
      "logps/chosen": -0.8365545272827148,
      "logps/rejected": -1.044828176498413,
      "loss": 3.6672,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -8.365545272827148,
      "rewards/margins": 2.082737445831299,
      "rewards/rejected": -10.448283195495605,
      "step": 115
    },
    {
      "epoch": 0.8723307587460245,
      "grad_norm": 89.06959408393006,
      "learning_rate": 3.711789783843522e-08,
      "logits/chosen": -6.565405368804932,
      "logits/rejected": -6.4343461990356445,
      "logps/chosen": -0.8695400953292847,
      "logps/rejected": -1.0754826068878174,
      "loss": 3.881,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -8.695401191711426,
      "rewards/margins": 2.0594258308410645,
      "rewards/rejected": -10.754826545715332,
      "step": 120
    },
    {
      "epoch": 0.9086778736937755,
      "grad_norm": 74.5528985533384,
      "learning_rate": 1.8641443178027784e-08,
      "logits/chosen": -7.178065299987793,
      "logits/rejected": -6.637632846832275,
      "logps/chosen": -0.8037236332893372,
      "logps/rejected": -0.9828687906265259,
      "loss": 3.7674,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -8.037236213684082,
      "rewards/margins": 1.791452169418335,
      "rewards/rejected": -9.828689575195312,
      "step": 125
    },
    {
      "epoch": 0.9450249886415266,
      "grad_norm": 70.70049345302832,
      "learning_rate": 6.376148290617145e-09,
      "logits/chosen": -7.0094404220581055,
      "logits/rejected": -6.915109157562256,
      "logps/chosen": -0.957663357257843,
      "logps/rejected": -1.1495531797409058,
      "loss": 3.7884,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -9.57663345336914,
      "rewards/margins": 1.9188988208770752,
      "rewards/rejected": -11.495532989501953,
      "step": 130
    },
    {
      "epoch": 0.9813721035892776,
      "grad_norm": 66.91451872853837,
      "learning_rate": 5.217771643080127e-10,
      "logits/chosen": -6.611492156982422,
      "logits/rejected": -6.406693458557129,
      "logps/chosen": -0.8434920310974121,
      "logps/rejected": -1.1371880769729614,
      "loss": 3.5998,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -8.434919357299805,
      "rewards/margins": 2.936959743499756,
      "rewards/rejected": -11.371880531311035,
      "step": 135
    },
    {
      "epoch": 0.995910949568378,
      "step": 137,
      "total_flos": 0.0,
      "train_loss": 4.287031420826041,
      "train_runtime": 1653.4486,
      "train_samples_per_second": 10.647,
      "train_steps_per_second": 0.083
    }
  ],
  "logging_steps": 5,
  "max_steps": 137,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}