{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9992429977289932,
  "eval_steps": 100,
  "global_step": 165,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.9411764705882356e-07,
      "logits/chosen": -2.5782699584960938,
      "logits/rejected": -2.4518439769744873,
      "logps/chosen": -312.26776123046875,
      "logps/rejected": -214.7339324951172,
      "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": 2.9411764705882355e-06,
      "logits/chosen": -2.643641233444214,
      "logits/rejected": -2.5239078998565674,
      "logps/chosen": -266.10455322265625,
      "logps/rejected": -183.74024963378906,
      "loss": 0.6918,
      "rewards/accuracies": 0.5763888955116272,
      "rewards/chosen": 0.004169768653810024,
      "rewards/margins": 0.003206828376278281,
      "rewards/rejected": 0.0009629399282857776,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.994932636402032e-06,
      "logits/chosen": -2.627925395965576,
      "logits/rejected": -2.516096830368042,
      "logps/chosen": -251.03671264648438,
      "logps/rejected": -191.53500366210938,
      "loss": 0.6733,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.07321886718273163,
      "rewards/margins": 0.03704284131526947,
      "rewards/rejected": 0.03617602214217186,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.905416503522124e-06,
      "logits/chosen": -2.591784715652466,
      "logits/rejected": -2.4847412109375,
      "logps/chosen": -269.05462646484375,
      "logps/rejected": -215.66268920898438,
      "loss": 0.6351,
      "rewards/accuracies": 0.746874988079071,
      "rewards/chosen": 0.00681548286229372,
      "rewards/margins": 0.12864618003368378,
      "rewards/rejected": -0.12183071672916412,
      "step": 30
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.707922373336524e-06,
      "logits/chosen": -2.4711194038391113,
      "logits/rejected": -2.359039783477783,
      "logps/chosen": -262.1103820800781,
      "logps/rejected": -221.37106323242188,
      "loss": 0.615,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.05128273367881775,
      "rewards/margins": 0.1722000390291214,
      "rewards/rejected": -0.22348275780677795,
      "step": 40
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.411315662967732e-06,
      "logits/chosen": -2.4395766258239746,
      "logits/rejected": -2.291720390319824,
      "logps/chosen": -260.3160705566406,
      "logps/rejected": -219.9590301513672,
      "loss": 0.5814,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": 0.013967704959213734,
      "rewards/margins": 0.27147507667541504,
      "rewards/rejected": -0.2575073838233948,
      "step": 50
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.028910905897229e-06,
      "logits/chosen": -2.4264254570007324,
      "logits/rejected": -2.2081563472747803,
      "logps/chosen": -284.1495666503906,
      "logps/rejected": -232.05477905273438,
      "loss": 0.5703,
      "rewards/accuracies": 0.824999988079071,
      "rewards/chosen": -0.018397916108369827,
      "rewards/margins": 0.331583172082901,
      "rewards/rejected": -0.3499810993671417,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.577874068920446e-06,
      "logits/chosen": -2.405364751815796,
      "logits/rejected": -2.20412278175354,
      "logps/chosen": -281.3593444824219,
      "logps/rejected": -230.1668701171875,
      "loss": 0.5628,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -0.03580819442868233,
      "rewards/margins": 0.31651827692985535,
      "rewards/rejected": -0.3523264527320862,
      "step": 70
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.0784519801008546e-06,
      "logits/chosen": -2.336822271347046,
      "logits/rejected": -2.152445077896118,
      "logps/chosen": -271.4667053222656,
      "logps/rejected": -227.0992889404297,
      "loss": 0.5528,
      "rewards/accuracies": 0.7593749761581421,
      "rewards/chosen": -0.05491885542869568,
      "rewards/margins": 0.3223936855792999,
      "rewards/rejected": -0.3773125112056732,
      "step": 80
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.553063458334059e-06,
      "logits/chosen": -2.3475964069366455,
      "logits/rejected": -2.181095600128174,
      "logps/chosen": -268.07257080078125,
      "logps/rejected": -234.201904296875,
      "loss": 0.5639,
      "rewards/accuracies": 0.8031250238418579,
      "rewards/chosen": -0.08924304693937302,
      "rewards/margins": 0.3508160710334778,
      "rewards/rejected": -0.440059095621109,
      "step": 90
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.025292943281429e-06,
      "logits/chosen": -2.2798125743865967,
      "logits/rejected": -2.2160840034484863,
      "logps/chosen": -270.85076904296875,
      "logps/rejected": -248.674072265625,
      "loss": 0.548,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.071611687541008,
      "rewards/margins": 0.3644859194755554,
      "rewards/rejected": -0.4360976219177246,
      "step": 100
    },
    {
      "epoch": 0.61,
      "eval_logits/chosen": -2.2954230308532715,
      "eval_logits/rejected": -2.1948401927948,
      "eval_logps/chosen": -296.5362548828125,
      "eval_logps/rejected": -282.4073791503906,
      "eval_loss": 0.6396384239196777,
      "eval_rewards/accuracies": 0.6259999871253967,
      "eval_rewards/chosen": -0.033822380006313324,
      "eval_rewards/margins": 0.1209036335349083,
      "eval_rewards/rejected": -0.1547260284423828,
      "eval_runtime": 383.8924,
      "eval_samples_per_second": 5.21,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5188318011445907e-06,
      "logits/chosen": -2.293454170227051,
      "logits/rejected": -2.161669969558716,
      "logps/chosen": -272.22222900390625,
      "logps/rejected": -258.5532531738281,
      "loss": 0.5477,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.13500520586967468,
      "rewards/margins": 0.3878163695335388,
      "rewards/rejected": -0.5228215456008911,
      "step": 110
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.0564148305586296e-06,
      "logits/chosen": -2.346884250640869,
      "logits/rejected": -2.1674342155456543,
      "logps/chosen": -285.71630859375,
      "logps/rejected": -253.7765655517578,
      "loss": 0.5415,
      "rewards/accuracies": 0.778124988079071,
      "rewards/chosen": -0.0934138298034668,
      "rewards/margins": 0.37636226415634155,
      "rewards/rejected": -0.4697761535644531,
      "step": 120
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.587997083462197e-07,
      "logits/chosen": -2.2985992431640625,
      "logits/rejected": -2.160468339920044,
      "logps/chosen": -288.93536376953125,
      "logps/rejected": -263.5223693847656,
      "loss": 0.534,
      "rewards/accuracies": 0.815625011920929,
      "rewards/chosen": -0.1033419594168663,
      "rewards/margins": 0.399946391582489,
      "rewards/rejected": -0.5032883882522583,
      "step": 130
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.438351873250492e-07,
      "logits/chosen": -2.281365156173706,
      "logits/rejected": -2.1153156757354736,
      "logps/chosen": -278.9851989746094,
      "logps/rejected": -236.8888702392578,
      "loss": 0.5358,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.09803396463394165,
      "rewards/margins": 0.41163355112075806,
      "rewards/rejected": -0.5096675157546997,
      "step": 140
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.2565987432367032e-07,
      "logits/chosen": -2.314439058303833,
      "logits/rejected": -2.1131763458251953,
      "logps/chosen": -279.57867431640625,
      "logps/rejected": -245.7858123779297,
      "loss": 0.5348,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -0.0749698355793953,
      "rewards/margins": 0.4277007579803467,
      "rewards/rejected": -0.5026706457138062,
      "step": 150
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4067554877743861e-08,
      "logits/chosen": -2.238752603530884,
      "logits/rejected": -2.113945960998535,
      "logps/chosen": -260.85333251953125,
      "logps/rejected": -233.0952606201172,
      "loss": 0.5351,
      "rewards/accuracies": 0.824999988079071,
      "rewards/chosen": -0.06594842672348022,
      "rewards/margins": 0.4099964201450348,
      "rewards/rejected": -0.4759448170661926,
      "step": 160
    },
    {
      "epoch": 1.0,
      "step": 165,
      "total_flos": 0.0,
      "train_loss": 0.5745967612122045,
      "train_runtime": 7254.1779,
      "train_samples_per_second": 2.913,
      "train_steps_per_second": 0.023
    }
  ],
  "logging_steps": 10,
  "max_steps": 165,
  "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
}