{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.99581589958159,
  "eval_steps": 500,
  "global_step": 119,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 8.333333333333333e-08,
      "logits/chosen": -2.6852197647094727,
      "logits/rejected": -2.6903719902038574,
      "logps/chosen": -263.7275390625,
      "logps/rejected": -230.14215087890625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.08,
      "learning_rate": 8.333333333333333e-07,
      "logits/chosen": -2.7360000610351562,
      "logits/rejected": -2.735159158706665,
      "logps/chosen": -277.9869079589844,
      "logps/rejected": -253.91012573242188,
      "loss": 0.6916,
      "rewards/accuracies": 0.4791666567325592,
      "rewards/chosen": 0.004290024284273386,
      "rewards/margins": 0.002072603441774845,
      "rewards/rejected": 0.0022174210753291845,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.862705056474794e-07,
      "logits/chosen": -2.764007568359375,
      "logits/rejected": -2.7425270080566406,
      "logps/chosen": -286.0954284667969,
      "logps/rejected": -274.6139221191406,
      "loss": 0.6689,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": 0.009724309667944908,
      "rewards/margins": 0.05151768773794174,
      "rewards/rejected": -0.041793376207351685,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.317841607379106e-07,
      "logits/chosen": -2.739412784576416,
      "logits/rejected": -2.731487512588501,
      "logps/chosen": -286.3190002441406,
      "logps/rejected": -291.80010986328125,
      "loss": 0.6266,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.25924447178840637,
      "rewards/margins": 0.24311673641204834,
      "rewards/rejected": -0.5023611783981323,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.403425107745314e-07,
      "logits/chosen": -2.7835636138916016,
      "logits/rejected": -2.7717299461364746,
      "logps/chosen": -322.2847900390625,
      "logps/rejected": -343.01287841796875,
      "loss": 0.6088,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.4827890396118164,
      "rewards/margins": 0.33479487895965576,
      "rewards/rejected": -0.8175839185714722,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 7.197718133561508e-07,
      "logits/chosen": -2.7819108963012695,
      "logits/rejected": -2.7709670066833496,
      "logps/chosen": -355.53387451171875,
      "logps/rejected": -347.73553466796875,
      "loss": 0.5815,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.5724222660064697,
      "rewards/margins": 0.429451048374176,
      "rewards/rejected": -1.001873254776001,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 5.803914069597341e-07,
      "logits/chosen": -2.65649676322937,
      "logits/rejected": -2.6217668056488037,
      "logps/chosen": -364.63458251953125,
      "logps/rejected": -355.6944580078125,
      "loss": 0.5917,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.7049036026000977,
      "rewards/margins": 0.4729110598564148,
      "rewards/rejected": -1.1778147220611572,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.3413050507959067e-07,
      "logits/chosen": -2.5201010704040527,
      "logits/rejected": -2.4627785682678223,
      "logps/chosen": -344.98284912109375,
      "logps/rejected": -325.82452392578125,
      "loss": 0.5671,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.628750741481781,
      "rewards/margins": 0.43215298652648926,
      "rewards/rejected": -1.060903787612915,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.935072052698059e-07,
      "logits/chosen": -2.376207113265991,
      "logits/rejected": -2.3495357036590576,
      "logps/chosen": -290.0788269042969,
      "logps/rejected": -357.1901550292969,
      "loss": 0.5714,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.6383817791938782,
      "rewards/margins": 0.5417054891586304,
      "rewards/rejected": -1.1800873279571533,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.7055709710194448e-07,
      "logits/chosen": -2.415030002593994,
      "logits/rejected": -2.389885187149048,
      "logps/chosen": -351.54156494140625,
      "logps/rejected": -358.383544921875,
      "loss": 0.5425,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.820050835609436,
      "rewards/margins": 0.46901077032089233,
      "rewards/rejected": -1.289061427116394,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.580316675034254e-08,
      "logits/chosen": -2.3294150829315186,
      "logits/rejected": -2.289402484893799,
      "logps/chosen": -357.5911560058594,
      "logps/rejected": -436.22357177734375,
      "loss": 0.5717,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.0840901136398315,
      "rewards/margins": 0.5104671120643616,
      "rewards/rejected": -1.5945571660995483,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.7355161444279343e-08,
      "logits/chosen": -2.406877040863037,
      "logits/rejected": -2.3612313270568848,
      "logps/chosen": -392.439697265625,
      "logps/rejected": -412.04833984375,
      "loss": 0.5647,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.9446671605110168,
      "rewards/margins": 0.5704779624938965,
      "rewards/rejected": -1.5151450634002686,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.596198278314927,
      "train_runtime": 1970.3201,
      "train_samples_per_second": 7.757,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 10,
  "max_steps": 119,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}