{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9467878001297858,
  "eval_steps": 25,
  "global_step": 375,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12978585334198572,
      "grad_norm": 0.0001288286439375952,
      "learning_rate": 9.985564567238237e-05,
      "logits/chosen": -2.892515182495117,
      "logits/rejected": -2.8839669227600098,
      "logps/chosen": -65.34242248535156,
      "logps/rejected": -120.23748779296875,
      "loss": 0.2047,
      "rewards/accuracies": 0.7599999904632568,
      "rewards/chosen": -2.3054304122924805,
      "rewards/margins": 7.21456241607666,
      "rewards/rejected": -9.519991874694824,
      "step": 25
    },
    {
      "epoch": 0.12978585334198572,
      "eval_logits/chosen": -3.1952435970306396,
      "eval_logits/rejected": -3.1321091651916504,
      "eval_logps/chosen": -126.43878936767578,
      "eval_logps/rejected": -265.6844177246094,
      "eval_loss": 4.6241984819062054e-07,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -8.37295913696289,
      "eval_rewards/margins": 15.654376983642578,
      "eval_rewards/rejected": -24.02733612060547,
      "eval_runtime": 140.1846,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 25
    },
    {
      "epoch": 0.25957170668397145,
      "grad_norm": 1.2395206795190461e-05,
      "learning_rate": 9.795296239506012e-05,
      "logits/chosen": -2.2458553314208984,
      "logits/rejected": -2.1915533542633057,
      "logps/chosen": -99.26245880126953,
      "logps/rejected": -254.2440643310547,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -5.69963264465332,
      "rewards/margins": 17.1381778717041,
      "rewards/rejected": -22.837810516357422,
      "step": 50
    },
    {
      "epoch": 0.25957170668397145,
      "eval_logits/chosen": -1.7933237552642822,
      "eval_logits/rejected": -1.7323336601257324,
      "eval_logps/chosen": -89.24785614013672,
      "eval_logps/rejected": -249.6978302001953,
      "eval_loss": 4.8683162390261714e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.653865337371826,
      "eval_rewards/margins": 17.77481460571289,
      "eval_rewards/rejected": -22.428680419921875,
      "eval_runtime": 140.0068,
      "eval_samples_per_second": 0.586,
      "eval_steps_per_second": 0.586,
      "step": 50
    },
    {
      "epoch": 0.3893575600259572,
      "grad_norm": 3.164160079904832e-05,
      "learning_rate": 9.392069159975199e-05,
      "logits/chosen": -1.7290905714035034,
      "logits/rejected": -1.6691365242004395,
      "logps/chosen": -87.64579010009766,
      "logps/rejected": -252.55418395996094,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.562257766723633,
      "rewards/margins": 18.078048706054688,
      "rewards/rejected": -22.640310287475586,
      "step": 75
    },
    {
      "epoch": 0.3893575600259572,
      "eval_logits/chosen": -1.7083513736724854,
      "eval_logits/rejected": -1.6462993621826172,
      "eval_logps/chosen": -86.74995422363281,
      "eval_logps/rejected": -249.47621154785156,
      "eval_loss": 3.741008924862399e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.404075622558594,
      "eval_rewards/margins": 18.002437591552734,
      "eval_rewards/rejected": -22.406513214111328,
      "eval_runtime": 140.046,
      "eval_samples_per_second": 0.586,
      "eval_steps_per_second": 0.586,
      "step": 75
    },
    {
      "epoch": 0.5191434133679429,
      "grad_norm": 9.557259545545094e-06,
      "learning_rate": 8.793790613463955e-05,
      "logits/chosen": -1.706992506980896,
      "logits/rejected": -1.6317412853240967,
      "logps/chosen": -85.46414184570312,
      "logps/rejected": -248.94418334960938,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.233969211578369,
      "rewards/margins": 18.118690490722656,
      "rewards/rejected": -22.3526611328125,
      "step": 100
    },
    {
      "epoch": 0.5191434133679429,
      "eval_logits/chosen": -1.701735019683838,
      "eval_logits/rejected": -1.6400282382965088,
      "eval_logps/chosen": -86.22808837890625,
      "eval_logps/rejected": -250.09986877441406,
      "eval_loss": 3.298516659810957e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.3518877029418945,
      "eval_rewards/margins": 18.11699104309082,
      "eval_rewards/rejected": -22.468881607055664,
      "eval_runtime": 140.2981,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 100
    },
    {
      "epoch": 0.6489292667099286,
      "grad_norm": 6.237313300516689e-06,
      "learning_rate": 8.027030106031836e-05,
      "logits/chosen": -1.6746500730514526,
      "logits/rejected": -1.6049107313156128,
      "logps/chosen": -85.8425521850586,
      "logps/rejected": -251.0710906982422,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.243345260620117,
      "rewards/margins": 18.34173011779785,
      "rewards/rejected": -22.585073471069336,
      "step": 125
    },
    {
      "epoch": 0.6489292667099286,
      "eval_logits/chosen": -1.7027819156646729,
      "eval_logits/rejected": -1.64132559299469,
      "eval_logps/chosen": -85.93473052978516,
      "eval_logps/rejected": -250.78919982910156,
      "eval_loss": 2.9771682008572498e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.32255220413208,
      "eval_rewards/margins": 18.21526336669922,
      "eval_rewards/rejected": -22.537818908691406,
      "eval_runtime": 140.288,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 125
    },
    {
      "epoch": 0.7787151200519143,
      "grad_norm": 1.6801877791294828e-05,
      "learning_rate": 7.125839415178204e-05,
      "logits/chosen": -1.7362980842590332,
      "logits/rejected": -1.6591914892196655,
      "logps/chosen": -86.65206146240234,
      "logps/rejected": -249.38253784179688,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.383853912353516,
      "rewards/margins": 18.103958129882812,
      "rewards/rejected": -22.487812042236328,
      "step": 150
    },
    {
      "epoch": 0.7787151200519143,
      "eval_logits/chosen": -1.7042571306228638,
      "eval_logits/rejected": -1.6435108184814453,
      "eval_logps/chosen": -85.64505004882812,
      "eval_logps/rejected": -251.6405029296875,
      "eval_loss": 2.621119676859962e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.293583393096924,
      "eval_rewards/margins": 18.329362869262695,
      "eval_rewards/rejected": -22.622947692871094,
      "eval_runtime": 140.1349,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 150
    },
    {
      "epoch": 0.9085009733939,
      "grad_norm": 9.597943972039502e-06,
      "learning_rate": 6.130240352918675e-05,
      "logits/chosen": -1.6720119714736938,
      "logits/rejected": -1.6064784526824951,
      "logps/chosen": -85.91622161865234,
      "logps/rejected": -251.5516815185547,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.345089435577393,
      "rewards/margins": 18.316930770874023,
      "rewards/rejected": -22.662019729614258,
      "step": 175
    },
    {
      "epoch": 0.9085009733939,
      "eval_logits/chosen": -1.7073944807052612,
      "eval_logits/rejected": -1.6472467184066772,
      "eval_logps/chosen": -85.54946899414062,
      "eval_logps/rejected": -252.8058319091797,
      "eval_loss": 2.286074263224691e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.284026145935059,
      "eval_rewards/margins": 18.455448150634766,
      "eval_rewards/rejected": -22.739477157592773,
      "eval_runtime": 140.0268,
      "eval_samples_per_second": 0.586,
      "eval_steps_per_second": 0.586,
      "step": 175
    },
    {
      "epoch": 1.0382868267358858,
      "grad_norm": 6.195426067279186e-06,
      "learning_rate": 5.084447400069655e-05,
      "logits/chosen": -1.6958116292953491,
      "logits/rejected": -1.639534592628479,
      "logps/chosen": -84.86266326904297,
      "logps/rejected": -252.62850952148438,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.275487899780273,
      "rewards/margins": 18.430957794189453,
      "rewards/rejected": -22.706445693969727,
      "step": 200
    },
    {
      "epoch": 1.0382868267358858,
      "eval_logits/chosen": -1.7099387645721436,
      "eval_logits/rejected": -1.6504337787628174,
      "eval_logps/chosen": -85.46736907958984,
      "eval_logps/rejected": -253.481689453125,
      "eval_loss": 2.1164542118867757e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.275816440582275,
      "eval_rewards/margins": 18.531248092651367,
      "eval_rewards/rejected": -22.807064056396484,
      "eval_runtime": 140.2558,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 200
    },
    {
      "epoch": 1.1680726800778716,
      "grad_norm": 5.08538914800738e-06,
      "learning_rate": 4.034904144421135e-05,
      "logits/chosen": -1.7136883735656738,
      "logits/rejected": -1.6375319957733154,
      "logps/chosen": -85.2287826538086,
      "logps/rejected": -254.05406188964844,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.193078517913818,
      "rewards/margins": 18.700963973999023,
      "rewards/rejected": -22.894041061401367,
      "step": 225
    },
    {
      "epoch": 1.1680726800778716,
      "eval_logits/chosen": -1.7104138135910034,
      "eval_logits/rejected": -1.6512484550476074,
      "eval_logps/chosen": -85.35152435302734,
      "eval_logps/rejected": -253.81373596191406,
      "eval_loss": 2.0240927156578437e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.264232158660889,
      "eval_rewards/margins": 18.57603645324707,
      "eval_rewards/rejected": -22.840267181396484,
      "eval_runtime": 140.3068,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 225
    },
    {
      "epoch": 1.2978585334198574,
      "grad_norm": 2.7552045139600523e-05,
      "learning_rate": 3.0282207244334082e-05,
      "logits/chosen": -1.7763639688491821,
      "logits/rejected": -1.7230497598648071,
      "logps/chosen": -85.6513900756836,
      "logps/rejected": -256.676513671875,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.309081554412842,
      "rewards/margins": 18.76219940185547,
      "rewards/rejected": -23.071279525756836,
      "step": 250
    },
    {
      "epoch": 1.2978585334198574,
      "eval_logits/chosen": -1.7111458778381348,
      "eval_logits/rejected": -1.6523288488388062,
      "eval_logps/chosen": -85.29007720947266,
      "eval_logps/rejected": -254.06683349609375,
      "eval_loss": 1.954951045490816e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.258088111877441,
      "eval_rewards/margins": 18.607492446899414,
      "eval_rewards/rejected": -22.865581512451172,
      "eval_runtime": 140.2669,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 250
    },
    {
      "epoch": 1.427644386761843,
      "grad_norm": 3.1899812711344566e-06,
      "learning_rate": 2.109103876430864e-05,
      "logits/chosen": -1.7086889743804932,
      "logits/rejected": -1.6506984233856201,
      "logps/chosen": -86.3566665649414,
      "logps/rejected": -254.89923095703125,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.3109564781188965,
      "rewards/margins": 18.709728240966797,
      "rewards/rejected": -23.02068328857422,
      "step": 275
    },
    {
      "epoch": 1.427644386761843,
      "eval_logits/chosen": -1.711836814880371,
      "eval_logits/rejected": -1.6533582210540771,
      "eval_logps/chosen": -85.26038360595703,
      "eval_logps/rejected": -254.31178283691406,
      "eval_loss": 1.9007002194371125e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.255118370056152,
      "eval_rewards/margins": 18.634958267211914,
      "eval_rewards/rejected": -22.89007568359375,
      "eval_runtime": 140.4683,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 275
    },
    {
      "epoch": 1.5574302401038287,
      "grad_norm": 1.0732557711889967e-05,
      "learning_rate": 1.3183715117440142e-05,
      "logits/chosen": -1.6755353212356567,
      "logits/rejected": -1.6350597143173218,
      "logps/chosen": -84.75931549072266,
      "logps/rejected": -254.6855926513672,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.289437770843506,
      "rewards/margins": 18.664514541625977,
      "rewards/rejected": -22.95395278930664,
      "step": 300
    },
    {
      "epoch": 1.5574302401038287,
      "eval_logits/chosen": -1.7125476598739624,
      "eval_logits/rejected": -1.6541738510131836,
      "eval_logps/chosen": -85.2017822265625,
      "eval_logps/rejected": -254.44964599609375,
      "eval_loss": 1.8585970096296478e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.249257564544678,
      "eval_rewards/margins": 18.65460205078125,
      "eval_rewards/rejected": -22.903860092163086,
      "eval_runtime": 140.3494,
      "eval_samples_per_second": 0.584,
      "eval_steps_per_second": 0.584,
      "step": 300
    },
    {
      "epoch": 1.6872160934458145,
      "grad_norm": 2.5711274247441906e-06,
      "learning_rate": 6.911399962822518e-06,
      "logits/chosen": -1.7367602586746216,
      "logits/rejected": -1.6708805561065674,
      "logps/chosen": -84.29496765136719,
      "logps/rejected": -253.51669311523438,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.2055463790893555,
      "rewards/margins": 18.59516143798828,
      "rewards/rejected": -22.800708770751953,
      "step": 325
    },
    {
      "epoch": 1.6872160934458145,
      "eval_logits/chosen": -1.7123589515686035,
      "eval_logits/rejected": -1.6541532278060913,
      "eval_logps/chosen": -85.1590347290039,
      "eval_logps/rejected": -254.51121520996094,
      "eval_loss": 1.8418553793253523e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.244982719421387,
      "eval_rewards/margins": 18.665035247802734,
      "eval_rewards/rejected": -22.910018920898438,
      "eval_runtime": 139.9542,
      "eval_samples_per_second": 0.586,
      "eval_steps_per_second": 0.586,
      "step": 325
    },
    {
      "epoch": 1.8170019467878,
      "grad_norm": 2.154295998479938e-06,
      "learning_rate": 2.5526463531765465e-06,
      "logits/chosen": -1.6854662895202637,
      "logits/rejected": -1.594865322113037,
      "logps/chosen": -83.5513687133789,
      "logps/rejected": -253.59136962890625,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -3.994952917098999,
      "rewards/margins": 18.832168579101562,
      "rewards/rejected": -22.827119827270508,
      "step": 350
    },
    {
      "epoch": 1.8170019467878,
      "eval_logits/chosen": -1.7126699686050415,
      "eval_logits/rejected": -1.6544029712677002,
      "eval_logps/chosen": -85.1473617553711,
      "eval_logps/rejected": -254.5518798828125,
      "eval_loss": 1.8251670397262387e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.243814945220947,
      "eval_rewards/margins": 18.670270919799805,
      "eval_rewards/rejected": -22.914085388183594,
      "eval_runtime": 140.166,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 350
    },
    {
      "epoch": 1.9467878001297858,
      "grad_norm": 4.992621597921243e-06,
      "learning_rate": 3.0102621435333934e-07,
      "logits/chosen": -1.7460432052612305,
      "logits/rejected": -1.6790217161178589,
      "logps/chosen": -85.45972442626953,
      "logps/rejected": -255.81414794921875,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.300573825836182,
      "rewards/margins": 18.717607498168945,
      "rewards/rejected": -23.01818084716797,
      "step": 375
    },
    {
      "epoch": 1.9467878001297858,
      "eval_logits/chosen": -1.712624192237854,
      "eval_logits/rejected": -1.6543128490447998,
      "eval_logps/chosen": -85.14376831054688,
      "eval_logps/rejected": -254.5596160888672,
      "eval_loss": 1.829120499508008e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -4.243457317352295,
      "eval_rewards/margins": 18.671401977539062,
      "eval_rewards/rejected": -22.914859771728516,
      "eval_runtime": 140.1214,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.585,
      "step": 375
    }
  ],
  "logging_steps": 25,
  "max_steps": 384,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 25,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}