{
  "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": 7.974906667324644,
      "learning_rate": 3.333333333333333e-07,
      "logits/chosen": -2.747501850128174,
      "logits/rejected": -2.7370193004608154,
      "logps/chosen": -291.8797302246094,
      "logps/rejected": -281.50189208984375,
      "loss": 0.693,
      "rewards/accuracies": 0.3812499940395355,
      "rewards/chosen": -0.00011452056060079485,
      "rewards/margins": -3.700270826811902e-05,
      "rewards/rejected": -7.751785597065464e-05,
      "step": 10
    },
    {
      "epoch": 0.13937282229965156,
      "grad_norm": 8.1980266894883,
      "learning_rate": 4.981198836496775e-07,
      "logits/chosen": -2.7630066871643066,
      "logits/rejected": -2.748826026916504,
      "logps/chosen": -298.94158935546875,
      "logps/rejected": -320.3427429199219,
      "loss": 0.6902,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": 0.008970660157501698,
      "rewards/margins": 0.006611389108002186,
      "rewards/rejected": 0.002359271515160799,
      "step": 20
    },
    {
      "epoch": 0.20905923344947736,
      "grad_norm": 8.031029605438988,
      "learning_rate": 4.832481997086846e-07,
      "logits/chosen": -2.7137038707733154,
      "logits/rejected": -2.7142727375030518,
      "logps/chosen": -296.2449951171875,
      "logps/rejected": -289.28814697265625,
      "loss": 0.681,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": 0.040586382150650024,
      "rewards/margins": 0.03953887149691582,
      "rewards/rejected": 0.0010475128656253219,
      "step": 30
    },
    {
      "epoch": 0.2787456445993031,
      "grad_norm": 9.014305613217058,
      "learning_rate": 4.543962032878959e-07,
      "logits/chosen": -2.679992198944092,
      "logits/rejected": -2.675081253051758,
      "logps/chosen": -298.83929443359375,
      "logps/rejected": -318.2642822265625,
      "loss": 0.6637,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.003969565033912659,
      "rewards/margins": 0.06901098042726517,
      "rewards/rejected": -0.07298054546117783,
      "step": 40
    },
    {
      "epoch": 0.34843205574912894,
      "grad_norm": 10.923357206738944,
      "learning_rate": 4.1329321073844413e-07,
      "logits/chosen": -2.547337770462036,
      "logits/rejected": -2.5558876991271973,
      "logps/chosen": -272.10382080078125,
      "logps/rejected": -269.55987548828125,
      "loss": 0.66,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.13082024455070496,
      "rewards/margins": 0.06251106411218643,
      "rewards/rejected": -0.1933313012123108,
      "step": 50
    },
    {
      "epoch": 0.34843205574912894,
      "eval_logits/chosen": -2.536876678466797,
      "eval_logits/rejected": -2.4980242252349854,
      "eval_logps/chosen": -273.9512634277344,
      "eval_logps/rejected": -286.3990783691406,
      "eval_loss": 0.6512594819068909,
      "eval_rewards/accuracies": 0.6640625,
      "eval_rewards/chosen": -0.11321306228637695,
      "eval_rewards/margins": 0.12415696680545807,
      "eval_rewards/rejected": -0.23737002909183502,
      "eval_runtime": 103.5252,
      "eval_samples_per_second": 19.319,
      "eval_steps_per_second": 0.309,
      "step": 50
    },
    {
      "epoch": 0.4181184668989547,
      "grad_norm": 11.450114313746024,
      "learning_rate": 3.624028324136517e-07,
      "logits/chosen": -2.5856661796569824,
      "logits/rejected": -2.5894103050231934,
      "logps/chosen": -315.35540771484375,
      "logps/rejected": -320.2020263671875,
      "loss": 0.6507,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.12617941200733185,
      "rewards/margins": 0.15071702003479004,
      "rewards/rejected": -0.2768964171409607,
      "step": 60
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 10.762410651532795,
      "learning_rate": 3.047753100392174e-07,
      "logits/chosen": -2.531461238861084,
      "logits/rejected": -2.5228896141052246,
      "logps/chosen": -301.96588134765625,
      "logps/rejected": -300.63177490234375,
      "loss": 0.6458,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.23531746864318848,
      "rewards/margins": 0.1557210385799408,
      "rewards/rejected": -0.3910384774208069,
      "step": 70
    },
    {
      "epoch": 0.5574912891986062,
      "grad_norm": 10.243677934319797,
      "learning_rate": 2.4386469286927194e-07,
      "logits/chosen": -2.531421184539795,
      "logits/rejected": -2.5143864154815674,
      "logps/chosen": -289.48651123046875,
      "logps/rejected": -301.8067932128906,
      "loss": 0.6511,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.187635600566864,
      "rewards/margins": 0.1388576477766037,
      "rewards/rejected": -0.3264932334423065,
      "step": 80
    },
    {
      "epoch": 0.627177700348432,
      "grad_norm": 10.605767289766804,
      "learning_rate": 1.8332181063127542e-07,
      "logits/chosen": -2.370906114578247,
      "logits/rejected": -2.396063804626465,
      "logps/chosen": -266.50262451171875,
      "logps/rejected": -298.68292236328125,
      "loss": 0.634,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.13985350728034973,
      "rewards/margins": 0.18151375651359558,
      "rewards/rejected": -0.3213672339916229,
      "step": 90
    },
    {
      "epoch": 0.6968641114982579,
      "grad_norm": 11.921015517224783,
      "learning_rate": 1.26775451942554e-07,
      "logits/chosen": -2.410521984100342,
      "logits/rejected": -2.4090819358825684,
      "logps/chosen": -325.18603515625,
      "logps/rejected": -335.88739013671875,
      "loss": 0.6376,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.2180793583393097,
      "rewards/margins": 0.21515274047851562,
      "rewards/rejected": -0.4332321286201477,
      "step": 100
    },
    {
      "epoch": 0.6968641114982579,
      "eval_logits/chosen": -2.345639705657959,
      "eval_logits/rejected": -2.2959439754486084,
      "eval_logps/chosen": -289.0152893066406,
      "eval_logps/rejected": -311.52935791015625,
      "eval_loss": 0.6328787803649902,
      "eval_rewards/accuracies": 0.6875,
      "eval_rewards/chosen": -0.26385369896888733,
      "eval_rewards/margins": 0.22481901943683624,
      "eval_rewards/rejected": -0.4886727035045624,
      "eval_runtime": 105.3794,
      "eval_samples_per_second": 18.979,
      "eval_steps_per_second": 0.304,
      "step": 100
    },
    {
      "epoch": 0.7665505226480837,
      "grad_norm": 10.917825147199611,
      "learning_rate": 7.761486381573326e-08,
      "logits/chosen": -2.425245761871338,
      "logits/rejected": -2.4225878715515137,
      "logps/chosen": -340.6777038574219,
      "logps/rejected": -376.40008544921875,
      "loss": 0.6263,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.29727238416671753,
      "rewards/margins": 0.22700247168540955,
      "rewards/rejected": -0.5242748856544495,
      "step": 110
    },
    {
      "epoch": 0.8362369337979094,
      "grad_norm": 11.561681770035124,
      "learning_rate": 3.878660868757322e-08,
      "logits/chosen": -2.329202175140381,
      "logits/rejected": -2.3225998878479004,
      "logps/chosen": -323.3353271484375,
      "logps/rejected": -341.76959228515625,
      "loss": 0.6324,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.2884915769100189,
      "rewards/margins": 0.22677147388458252,
      "rewards/rejected": -0.5152631402015686,
      "step": 120
    },
    {
      "epoch": 0.9059233449477352,
      "grad_norm": 11.514501964616796,
      "learning_rate": 1.261795485174083e-08,
      "logits/chosen": -2.403071165084839,
      "logits/rejected": -2.402184009552002,
      "logps/chosen": -327.9507141113281,
      "logps/rejected": -337.8323059082031,
      "loss": 0.6248,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.27428099513053894,
      "rewards/margins": 0.22150692343711853,
      "rewards/rejected": -0.49578791856765747,
      "step": 130
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 11.032500717539952,
      "learning_rate": 6.773858303274482e-10,
      "logits/chosen": -2.3575150966644287,
      "logits/rejected": -2.346174716949463,
      "logps/chosen": -298.3825988769531,
      "logps/rejected": -313.13848876953125,
      "loss": 0.6258,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.3439212739467621,
      "rewards/margins": 0.12675735354423523,
      "rewards/rejected": -0.4706786274909973,
      "step": 140
    },
    {
      "epoch": 0.9965156794425087,
      "step": 143,
      "total_flos": 0.0,
      "train_loss": 0.6509068312344851,
      "train_runtime": 3880.8302,
      "train_samples_per_second": 4.726,
      "train_steps_per_second": 0.037
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}