{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9968652037617555,
  "eval_steps": 500,
  "global_step": 159,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-08,
      "logits/chosen": -2.191464900970459,
      "logits/rejected": -2.192023754119873,
      "logps/chosen": -337.12921142578125,
      "logps/rejected": -401.0337829589844,
      "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": 3.1249999999999997e-07,
      "logits/chosen": -2.265860080718994,
      "logits/rejected": -2.208857536315918,
      "logps/chosen": -230.86334228515625,
      "logps/rejected": -311.0320739746094,
      "loss": 0.6885,
      "rewards/accuracies": 0.5208333134651184,
      "rewards/chosen": -0.012912267819046974,
      "rewards/margins": 0.01672884076833725,
      "rewards/rejected": -0.029641108587384224,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.990353313429303e-07,
      "logits/chosen": -2.1798629760742188,
      "logits/rejected": -2.165292263031006,
      "logps/chosen": -277.3706359863281,
      "logps/rejected": -361.240234375,
      "loss": 0.679,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.4758126735687256,
      "rewards/margins": 0.16070665419101715,
      "rewards/rejected": -0.6365193128585815,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.882681251368548e-07,
      "logits/chosen": -2.065920829772949,
      "logits/rejected": -2.056377649307251,
      "logps/chosen": -284.94183349609375,
      "logps/rejected": -382.9114685058594,
      "loss": 0.6709,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.44684070348739624,
      "rewards/margins": 0.24182553589344025,
      "rewards/rejected": -0.6886661648750305,
      "step": 30
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.6604720940421207e-07,
      "logits/chosen": -2.139819383621216,
      "logits/rejected": -2.0983104705810547,
      "logps/chosen": -261.02471923828125,
      "logps/rejected": -352.98162841796875,
      "loss": 0.6523,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.22146382927894592,
      "rewards/margins": 0.14547555148601532,
      "rewards/rejected": -0.36693936586380005,
      "step": 40
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3344075855595097e-07,
      "logits/chosen": -2.1886887550354004,
      "logits/rejected": -2.159538984298706,
      "logps/chosen": -278.70196533203125,
      "logps/rejected": -405.36700439453125,
      "loss": 0.6368,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.3322455883026123,
      "rewards/margins": 0.2992514371871948,
      "rewards/rejected": -0.6314969658851624,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.920161866827889e-07,
      "logits/chosen": -2.08825421333313,
      "logits/rejected": -2.0380659103393555,
      "logps/chosen": -288.3702392578125,
      "logps/rejected": -361.09112548828125,
      "loss": 0.644,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.3619961142539978,
      "rewards/margins": 0.1461850106716156,
      "rewards/rejected": -0.5081811547279358,
      "step": 60
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4376480090239047e-07,
      "logits/chosen": -2.074207305908203,
      "logits/rejected": -2.0528454780578613,
      "logps/chosen": -277.1039123535156,
      "logps/rejected": -329.77154541015625,
      "loss": 0.6446,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.2841487526893616,
      "rewards/margins": 0.1815183013677597,
      "rewards/rejected": -0.4656670093536377,
      "step": 70
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.910060778827554e-07,
      "logits/chosen": -2.169900894165039,
      "logits/rejected": -2.136660099029541,
      "logps/chosen": -323.7901916503906,
      "logps/rejected": -358.14471435546875,
      "loss": 0.6368,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.4767797887325287,
      "rewards/margins": 0.1753493845462799,
      "rewards/rejected": -0.6521291732788086,
      "step": 80
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.3627616503391812e-07,
      "logits/chosen": -2.2116239070892334,
      "logits/rejected": -2.107689380645752,
      "logps/chosen": -287.3326721191406,
      "logps/rejected": -396.2935791015625,
      "loss": 0.6241,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.35074305534362793,
      "rewards/margins": 0.3240629732608795,
      "rewards/rejected": -0.6748059988021851,
      "step": 90
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8220596619089573e-07,
      "logits/chosen": -2.1734859943389893,
      "logits/rejected": -2.1512904167175293,
      "logps/chosen": -280.8328552246094,
      "logps/rejected": -403.6314697265625,
      "loss": 0.6331,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.32056325674057007,
      "rewards/margins": 0.2918151617050171,
      "rewards/rejected": -0.6123784184455872,
      "step": 100
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3139467229135998e-07,
      "logits/chosen": -2.1452653408050537,
      "logits/rejected": -2.156574249267578,
      "logps/chosen": -303.7337646484375,
      "logps/rejected": -390.61248779296875,
      "loss": 0.6426,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.33146318793296814,
      "rewards/margins": 0.27743712067604065,
      "rewards/rejected": -0.6089003086090088,
      "step": 110
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.628481651367875e-08,
      "logits/chosen": -2.1824703216552734,
      "logits/rejected": -2.148677110671997,
      "logps/chosen": -270.36383056640625,
      "logps/rejected": -371.7115478515625,
      "loss": 0.6145,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.35747700929641724,
      "rewards/margins": 0.259312242269516,
      "rewards/rejected": -0.6167892217636108,
      "step": 120
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.904486005914027e-08,
      "logits/chosen": -2.155510187149048,
      "logits/rejected": -2.1049563884735107,
      "logps/chosen": -283.0008850097656,
      "logps/rejected": -388.17022705078125,
      "loss": 0.6513,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.3318966329097748,
      "rewards/margins": 0.3010888993740082,
      "rewards/rejected": -0.632985532283783,
      "step": 130
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1464952759020856e-08,
      "logits/chosen": -2.147397756576538,
      "logits/rejected": -2.157912492752075,
      "logps/chosen": -267.1644592285156,
      "logps/rejected": -378.49908447265625,
      "loss": 0.6245,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.32303911447525024,
      "rewards/margins": 0.2776426374912262,
      "rewards/rejected": -0.6006818413734436,
      "step": 140
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.8708793644441086e-09,
      "logits/chosen": -2.1643731594085693,
      "logits/rejected": -2.187424659729004,
      "logps/chosen": -267.67608642578125,
      "logps/rejected": -405.1327209472656,
      "loss": 0.6384,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.3011611998081207,
      "rewards/margins": 0.31678175926208496,
      "rewards/rejected": -0.6179429888725281,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 159,
      "total_flos": 0.0,
      "train_loss": 0.6447140285803837,
      "train_runtime": 2654.6606,
      "train_samples_per_second": 7.676,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 10,
  "max_steps": 159,
  "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
}