{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9748953974895398,
  "eval_steps": 500,
  "global_step": 118,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016736401673640166,
      "grad_norm": 7.772315443006173,
      "learning_rate": 4.166666666666666e-08,
      "logits/chosen": -2.807276487350464,
      "logits/rejected": -2.7759768962860107,
      "logps/chosen": -315.42626953125,
      "logps/rejected": -227.5915985107422,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 7.471216567487246,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.757866382598877,
      "logits/rejected": -2.7485244274139404,
      "logps/chosen": -272.32708740234375,
      "logps/rejected": -260.8048095703125,
      "loss": 0.6928,
      "rewards/accuracies": 0.4895833432674408,
      "rewards/chosen": 0.0014287387020885944,
      "rewards/margins": 0.000867457827553153,
      "rewards/rejected": 0.0005612808163277805,
      "step": 10
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 6.635008900115668,
      "learning_rate": 4.930057285201027e-07,
      "logits/chosen": -2.796506404876709,
      "logits/rejected": -2.7773609161376953,
      "logps/chosen": -261.6402893066406,
      "logps/rejected": -249.42807006835938,
      "loss": 0.6816,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": 0.03041391633450985,
      "rewards/margins": 0.024269890040159225,
      "rewards/rejected": 0.006144027225673199,
      "step": 20
    },
    {
      "epoch": 0.502092050209205,
      "grad_norm": 7.298705785979101,
      "learning_rate": 4.652609029418388e-07,
      "logits/chosen": -2.790365219116211,
      "logits/rejected": -2.771557331085205,
      "logps/chosen": -291.3450012207031,
      "logps/rejected": -249.59716796875,
      "loss": 0.6546,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": 0.0011108840117231011,
      "rewards/margins": 0.10550198704004288,
      "rewards/rejected": -0.10439109802246094,
      "step": 30
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 8.142120480110528,
      "learning_rate": 4.187457503795526e-07,
      "logits/chosen": -2.7561163902282715,
      "logits/rejected": -2.7333474159240723,
      "logps/chosen": -268.2800598144531,
      "logps/rejected": -261.1468200683594,
      "loss": 0.6342,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.1536058485507965,
      "rewards/margins": 0.17475590109825134,
      "rewards/rejected": -0.32836171984672546,
      "step": 40
    },
    {
      "epoch": 0.8368200836820083,
      "grad_norm": 10.537722600272415,
      "learning_rate": 3.5751630056913013e-07,
      "logits/chosen": -2.7716526985168457,
      "logits/rejected": -2.751047134399414,
      "logps/chosen": -292.7337646484375,
      "logps/rejected": -306.22894287109375,
      "loss": 0.6074,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.3572509288787842,
      "rewards/margins": 0.1899140328168869,
      "rewards/rejected": -0.5471649765968323,
      "step": 50
    },
    {
      "epoch": 1.00418410041841,
      "grad_norm": 10.000348369076317,
      "learning_rate": 2.8691164100062034e-07,
      "logits/chosen": -2.7915871143341064,
      "logits/rejected": -2.757317304611206,
      "logps/chosen": -319.2095947265625,
      "logps/rejected": -317.7526550292969,
      "loss": 0.5933,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.38744381070137024,
      "rewards/margins": 0.33469024300575256,
      "rewards/rejected": -0.722133994102478,
      "step": 60
    },
    {
      "epoch": 1.1715481171548117,
      "grad_norm": 12.875178033360463,
      "learning_rate": 2.1308835899937972e-07,
      "logits/chosen": -2.8087921142578125,
      "logits/rejected": -2.763150691986084,
      "logps/chosen": -327.20989990234375,
      "logps/rejected": -348.8834228515625,
      "loss": 0.4977,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.3436049818992615,
      "rewards/margins": 0.6280093789100647,
      "rewards/rejected": -0.9716142416000366,
      "step": 70
    },
    {
      "epoch": 1.3389121338912133,
      "grad_norm": 14.484605319084237,
      "learning_rate": 1.4248369943086995e-07,
      "logits/chosen": -2.686325788497925,
      "logits/rejected": -2.6680502891540527,
      "logps/chosen": -312.7579650878906,
      "logps/rejected": -380.6637268066406,
      "loss": 0.4745,
      "rewards/accuracies": 0.840624988079071,
      "rewards/chosen": -0.3891797661781311,
      "rewards/margins": 0.8134768605232239,
      "rewards/rejected": -1.2026567459106445,
      "step": 80
    },
    {
      "epoch": 1.506276150627615,
      "grad_norm": 13.391182563638015,
      "learning_rate": 8.125424962044741e-08,
      "logits/chosen": -2.7235772609710693,
      "logits/rejected": -2.7114028930664062,
      "logps/chosen": -313.1542053222656,
      "logps/rejected": -370.19891357421875,
      "loss": 0.4657,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -0.5553286671638489,
      "rewards/margins": 0.7646803259849548,
      "rewards/rejected": -1.3200088739395142,
      "step": 90
    },
    {
      "epoch": 1.6736401673640167,
      "grad_norm": 16.395163898429864,
      "learning_rate": 3.473909705816111e-08,
      "logits/chosen": -2.637983798980713,
      "logits/rejected": -2.605607748031616,
      "logps/chosen": -308.8186340332031,
      "logps/rejected": -373.8332824707031,
      "loss": 0.4672,
      "rewards/accuracies": 0.8343750238418579,
      "rewards/chosen": -0.5417979955673218,
      "rewards/margins": 0.7996915578842163,
      "rewards/rejected": -1.341489553451538,
      "step": 100
    },
    {
      "epoch": 1.8410041841004183,
      "grad_norm": 15.975673050956397,
      "learning_rate": 6.994271479897313e-09,
      "logits/chosen": -2.689378261566162,
      "logits/rejected": -2.638415813446045,
      "logps/chosen": -333.02862548828125,
      "logps/rejected": -368.8213195800781,
      "loss": 0.4608,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": -0.6281369924545288,
      "rewards/margins": 0.7409355044364929,
      "rewards/rejected": -1.3690725564956665,
      "step": 110
    },
    {
      "epoch": 1.9748953974895398,
      "step": 118,
      "total_flos": 0.0,
      "train_loss": 0.5588717278787645,
      "train_runtime": 3312.4835,
      "train_samples_per_second": 9.228,
      "train_steps_per_second": 0.036
    }
  ],
  "logging_steps": 10,
  "max_steps": 118,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}