{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-07,
      "logits/chosen": -2.556900978088379,
      "logits/rejected": -2.42356276512146,
      "logps/chosen": -384.0308532714844,
      "logps/rejected": -355.0808410644531,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-06,
      "logits/chosen": -2.611865997314453,
      "logits/rejected": -2.5766751766204834,
      "logps/chosen": -304.4154052734375,
      "logps/rejected": -320.59613037109375,
      "loss": 0.6929,
      "rewards/accuracies": 0.4791666567325592,
      "rewards/chosen": 0.003964083269238472,
      "rewards/margins": 0.000560062238946557,
      "rewards/rejected": 0.003404021030291915,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": -2.6072850227355957,
      "logits/rejected": -2.522531270980835,
      "logps/chosen": -300.6770324707031,
      "logps/rejected": -291.0304870605469,
      "loss": 0.6884,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": 0.0014870109735056758,
      "rewards/margins": 0.011296196840703487,
      "rewards/rejected": -0.009809186682105064,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": -2.6037135124206543,
      "logits/rejected": -2.523104190826416,
      "logps/chosen": -316.41357421875,
      "logps/rejected": -328.901611328125,
      "loss": 0.6823,
      "rewards/accuracies": 0.6156250238418579,
      "rewards/chosen": -0.03885359317064285,
      "rewards/margins": 0.034083932638168335,
      "rewards/rejected": -0.07293752580881119,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.5474514961242676,
      "logits/rejected": -2.465183734893799,
      "logps/chosen": -314.0546569824219,
      "logps/rejected": -331.37213134765625,
      "loss": 0.6678,
      "rewards/accuracies": 0.596875011920929,
      "rewards/chosen": -0.09560319036245346,
      "rewards/margins": 0.046604275703430176,
      "rewards/rejected": -0.14220745861530304,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": -2.5767412185668945,
      "logits/rejected": -2.5177969932556152,
      "logps/chosen": -303.9281311035156,
      "logps/rejected": -316.2641296386719,
      "loss": 0.6649,
      "rewards/accuracies": 0.640625,
      "rewards/chosen": -0.10496363788843155,
      "rewards/margins": 0.07360312342643738,
      "rewards/rejected": -0.17856675386428833,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": -2.5027236938476562,
      "logits/rejected": -2.4811007976531982,
      "logps/chosen": -309.5362243652344,
      "logps/rejected": -332.8122253417969,
      "loss": 0.6648,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.12640534341335297,
      "rewards/margins": 0.08432017266750336,
      "rewards/rejected": -0.21072551608085632,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": -2.476938486099243,
      "logits/rejected": -2.454585313796997,
      "logps/chosen": -311.40582275390625,
      "logps/rejected": -342.4869384765625,
      "loss": 0.6579,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.14647440612316132,
      "rewards/margins": 0.11680541932582855,
      "rewards/rejected": -0.26327982544898987,
      "step": 70
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.5043561458587646,
      "logits/rejected": -2.4481701850891113,
      "logps/chosen": -330.54974365234375,
      "logps/rejected": -360.7945556640625,
      "loss": 0.655,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.1902867704629898,
      "rewards/margins": 0.09953726828098297,
      "rewards/rejected": -0.2898240387439728,
      "step": 80
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": -2.5335912704467773,
      "logits/rejected": -2.4210548400878906,
      "logps/chosen": -342.34173583984375,
      "logps/rejected": -353.93780517578125,
      "loss": 0.6512,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.18997912108898163,
      "rewards/margins": 0.11557333171367645,
      "rewards/rejected": -0.30555248260498047,
      "step": 90
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": -2.4827916622161865,
      "logits/rejected": -2.40724515914917,
      "logps/chosen": -337.3109436035156,
      "logps/rejected": -340.58880615234375,
      "loss": 0.6563,
      "rewards/accuracies": 0.6156250238418579,
      "rewards/chosen": -0.1919889748096466,
      "rewards/margins": 0.11718972772359848,
      "rewards/rejected": -0.3091786801815033,
      "step": 100
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.445232629776001,
      "eval_logits/rejected": -2.3392903804779053,
      "eval_logps/chosen": -329.4806213378906,
      "eval_logps/rejected": -334.0378723144531,
      "eval_loss": 0.6402544379234314,
      "eval_rewards/accuracies": 0.6159999966621399,
      "eval_rewards/chosen": -0.3107789158821106,
      "eval_rewards/margins": 0.13125000894069672,
      "eval_rewards/rejected": -0.4420289993286133,
      "eval_runtime": 384.1641,
      "eval_samples_per_second": 5.206,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": -2.487079620361328,
      "logits/rejected": -2.445570468902588,
      "logps/chosen": -350.1665344238281,
      "logps/rejected": -376.484375,
      "loss": 0.6497,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.2455425262451172,
      "rewards/margins": 0.10617774724960327,
      "rewards/rejected": -0.35172027349472046,
      "step": 110
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -2.469482183456421,
      "logits/rejected": -2.386582612991333,
      "logps/chosen": -364.2555236816406,
      "logps/rejected": -356.3113708496094,
      "loss": 0.6511,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.24673068523406982,
      "rewards/margins": 0.10269691050052643,
      "rewards/rejected": -0.34942758083343506,
      "step": 120
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": -2.4042646884918213,
      "logits/rejected": -2.3691623210906982,
      "logps/chosen": -304.0820007324219,
      "logps/rejected": -340.99212646484375,
      "loss": 0.6432,
      "rewards/accuracies": 0.653124988079071,
      "rewards/chosen": -0.22045591473579407,
      "rewards/margins": 0.12117135524749756,
      "rewards/rejected": -0.3416272699832916,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": -2.393618583679199,
      "logits/rejected": -2.3461203575134277,
      "logps/chosen": -338.6348571777344,
      "logps/rejected": -369.8921203613281,
      "loss": 0.6433,
      "rewards/accuracies": 0.659375011920929,
      "rewards/chosen": -0.230948805809021,
      "rewards/margins": 0.1425984799861908,
      "rewards/rejected": -0.3735472559928894,
      "step": 140
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": -2.4482178688049316,
      "logits/rejected": -2.404700756072998,
      "logps/chosen": -317.01458740234375,
      "logps/rejected": -332.4150695800781,
      "loss": 0.6451,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.25384414196014404,
      "rewards/margins": 0.13600240647792816,
      "rewards/rejected": -0.389846533536911,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.6604029192374303,
      "train_runtime": 7258.1064,
      "train_samples_per_second": 2.756,
      "train_steps_per_second": 0.021
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}