{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9965156794425087,
  "eval_steps": 50,
  "global_step": 143,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06968641114982578,
      "grad_norm": 8.879917985004713,
      "learning_rate": 3.333333333333333e-07,
      "logits/chosen": -2.5109400749206543,
      "logits/rejected": -2.4602229595184326,
      "logps/chosen": -224.4826202392578,
      "logps/rejected": -205.3321075439453,
      "loss": 0.6925,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": 0.0009716759668663144,
      "rewards/margins": 0.001607558922842145,
      "rewards/rejected": -0.0006358829559758306,
      "step": 10
    },
    {
      "epoch": 0.13937282229965156,
      "grad_norm": 8.119644042984005,
      "learning_rate": 4.981198836496775e-07,
      "logits/chosen": -2.538778305053711,
      "logits/rejected": -2.402923583984375,
      "logps/chosen": -231.76394653320312,
      "logps/rejected": -203.54684448242188,
      "loss": 0.6753,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": 0.008117455057799816,
      "rewards/margins": 0.029356980696320534,
      "rewards/rejected": -0.02123952843248844,
      "step": 20
    },
    {
      "epoch": 0.20905923344947736,
      "grad_norm": 10.589687412392658,
      "learning_rate": 4.832481997086846e-07,
      "logits/chosen": -2.5253872871398926,
      "logits/rejected": -2.430739402770996,
      "logps/chosen": -225.355712890625,
      "logps/rejected": -253.0909881591797,
      "loss": 0.6114,
      "rewards/accuracies": 0.856249988079071,
      "rewards/chosen": 0.030127260833978653,
      "rewards/margins": 0.18957777321338654,
      "rewards/rejected": -0.15945051610469818,
      "step": 30
    },
    {
      "epoch": 0.2787456445993031,
      "grad_norm": 19.230572297481732,
      "learning_rate": 4.543962032878959e-07,
      "logits/chosen": -2.4045207500457764,
      "logits/rejected": -2.3013105392456055,
      "logps/chosen": -241.72854614257812,
      "logps/rejected": -256.7030944824219,
      "loss": 0.5447,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": -0.04323701187968254,
      "rewards/margins": 0.447670042514801,
      "rewards/rejected": -0.4909070134162903,
      "step": 40
    },
    {
      "epoch": 0.34843205574912894,
      "grad_norm": 23.777968472263748,
      "learning_rate": 4.1329321073844413e-07,
      "logits/chosen": -2.2776694297790527,
      "logits/rejected": -2.1192574501037598,
      "logps/chosen": -255.7515411376953,
      "logps/rejected": -307.02410888671875,
      "loss": 0.4719,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -0.31196385622024536,
      "rewards/margins": 0.6551374197006226,
      "rewards/rejected": -0.9671012163162231,
      "step": 50
    },
    {
      "epoch": 0.34843205574912894,
      "eval_logits/chosen": -1.996337890625,
      "eval_logits/rejected": -1.9396870136260986,
      "eval_logps/chosen": -301.5677795410156,
      "eval_logps/rejected": -352.47076416015625,
      "eval_loss": 0.5898596048355103,
      "eval_rewards/accuracies": 0.70703125,
      "eval_rewards/chosen": -0.3894880414009094,
      "eval_rewards/margins": 0.5086125135421753,
      "eval_rewards/rejected": -0.8981005549430847,
      "eval_runtime": 102.5575,
      "eval_samples_per_second": 19.501,
      "eval_steps_per_second": 0.312,
      "step": 50
    },
    {
      "epoch": 0.4181184668989547,
      "grad_norm": 33.164641300897365,
      "learning_rate": 3.624028324136517e-07,
      "logits/chosen": -1.2023751735687256,
      "logits/rejected": -0.8880468606948853,
      "logps/chosen": -297.7267150878906,
      "logps/rejected": -362.5870056152344,
      "loss": 0.4024,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": -0.5735751986503601,
      "rewards/margins": 0.9970341920852661,
      "rewards/rejected": -1.570609211921692,
      "step": 60
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 26.00066903060501,
      "learning_rate": 3.047753100392174e-07,
      "logits/chosen": -0.6797115802764893,
      "logits/rejected": -0.30923840403556824,
      "logps/chosen": -267.3857116699219,
      "logps/rejected": -395.60101318359375,
      "loss": 0.362,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": -0.5475583076477051,
      "rewards/margins": 1.2465879917144775,
      "rewards/rejected": -1.7941462993621826,
      "step": 70
    },
    {
      "epoch": 0.5574912891986062,
      "grad_norm": 36.007825212896435,
      "learning_rate": 2.4386469286927194e-07,
      "logits/chosen": -0.32210594415664673,
      "logits/rejected": 0.29763275384902954,
      "logps/chosen": -286.4994812011719,
      "logps/rejected": -484.10357666015625,
      "loss": 0.3448,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -0.6800082921981812,
      "rewards/margins": 1.5236244201660156,
      "rewards/rejected": -2.2036328315734863,
      "step": 80
    },
    {
      "epoch": 0.627177700348432,
      "grad_norm": 28.429320380976726,
      "learning_rate": 1.8332181063127542e-07,
      "logits/chosen": -0.29179516434669495,
      "logits/rejected": 0.5656725168228149,
      "logps/chosen": -315.7966613769531,
      "logps/rejected": -458.8309631347656,
      "loss": 0.3319,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -0.6928261518478394,
      "rewards/margins": 1.5519336462020874,
      "rewards/rejected": -2.244760036468506,
      "step": 90
    },
    {
      "epoch": 0.6968641114982579,
      "grad_norm": 29.97250123159769,
      "learning_rate": 1.26775451942554e-07,
      "logits/chosen": -0.13269878923892975,
      "logits/rejected": 0.6727190017700195,
      "logps/chosen": -300.4376220703125,
      "logps/rejected": -451.68609619140625,
      "loss": 0.3224,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -0.6514891982078552,
      "rewards/margins": 1.551584243774414,
      "rewards/rejected": -2.203073501586914,
      "step": 100
    },
    {
      "epoch": 0.6968641114982579,
      "eval_logits/chosen": -0.2416563630104065,
      "eval_logits/rejected": 0.21337364614009857,
      "eval_logps/chosen": -330.0835266113281,
      "eval_logps/rejected": -419.1961364746094,
      "eval_loss": 0.5598118305206299,
      "eval_rewards/accuracies": 0.75,
      "eval_rewards/chosen": -0.6746450662612915,
      "eval_rewards/margins": 0.8907086849212646,
      "eval_rewards/rejected": -1.5653537511825562,
      "eval_runtime": 100.9616,
      "eval_samples_per_second": 19.81,
      "eval_steps_per_second": 0.317,
      "step": 100
    },
    {
      "epoch": 0.7665505226480837,
      "grad_norm": 26.411152004320307,
      "learning_rate": 7.761486381573326e-08,
      "logits/chosen": 0.1635294407606125,
      "logits/rejected": 1.3954848051071167,
      "logps/chosen": -325.93487548828125,
      "logps/rejected": -446.07916259765625,
      "loss": 0.3142,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -0.7772396206855774,
      "rewards/margins": 1.6457939147949219,
      "rewards/rejected": -2.4230334758758545,
      "step": 110
    },
    {
      "epoch": 0.8362369337979094,
      "grad_norm": 32.75502406597345,
      "learning_rate": 3.878660868757322e-08,
      "logits/chosen": 0.5776845216751099,
      "logits/rejected": 1.9672679901123047,
      "logps/chosen": -338.0449523925781,
      "logps/rejected": -449.5287170410156,
      "loss": 0.3042,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -0.9358948469161987,
      "rewards/margins": 1.7070610523223877,
      "rewards/rejected": -2.642955780029297,
      "step": 120
    },
    {
      "epoch": 0.9059233449477352,
      "grad_norm": 34.08869226634673,
      "learning_rate": 1.261795485174083e-08,
      "logits/chosen": 0.4366391599178314,
      "logits/rejected": 1.6738389730453491,
      "logps/chosen": -295.5234375,
      "logps/rejected": -462.7455139160156,
      "loss": 0.3275,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -0.7988389730453491,
      "rewards/margins": 1.923356294631958,
      "rewards/rejected": -2.7221951484680176,
      "step": 130
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 33.23571195007149,
      "learning_rate": 6.773858303274482e-10,
      "logits/chosen": 0.5510319471359253,
      "logits/rejected": 1.6277908086776733,
      "logps/chosen": -300.76043701171875,
      "logps/rejected": -455.19793701171875,
      "loss": 0.3272,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -0.9057596325874329,
      "rewards/margins": 1.6654990911483765,
      "rewards/rejected": -2.571258544921875,
      "step": 140
    },
    {
      "epoch": 0.9965156794425087,
      "step": 143,
      "total_flos": 0.0,
      "train_loss": 0.42797933008287337,
      "train_runtime": 3631.7108,
      "train_samples_per_second": 5.05,
      "train_steps_per_second": 0.039
    }
  ],
  "logging_steps": 10,
  "max_steps": 143,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "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": 8,
  "trial_name": null,
  "trial_params": null
}