{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-07,
      "logits/chosen": -2.6029441356658936,
      "logits/rejected": -2.5599632263183594,
      "logps/chosen": -189.41246032714844,
      "logps/rejected": -190.13034057617188,
      "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": 3.125e-06,
      "logits/chosen": -2.6171505451202393,
      "logits/rejected": -2.537480592727661,
      "logps/chosen": -160.8114776611328,
      "logps/rejected": -149.8831024169922,
      "loss": 0.6913,
      "rewards/accuracies": 0.4618055522441864,
      "rewards/chosen": -0.009910209104418755,
      "rewards/margins": 0.004514796659350395,
      "rewards/rejected": -0.01442500576376915,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": -2.623683452606201,
      "logits/rejected": -2.5250072479248047,
      "logps/chosen": -167.83267211914062,
      "logps/rejected": -158.3927764892578,
      "loss": 0.676,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.0703786239027977,
      "rewards/margins": 0.037703804671764374,
      "rewards/rejected": -0.10808243602514267,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": -2.6501214504241943,
      "logits/rejected": -2.4790871143341064,
      "logps/chosen": -172.5028839111328,
      "logps/rejected": -160.69882202148438,
      "loss": 0.6547,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.05348671227693558,
      "rewards/margins": 0.1112779751420021,
      "rewards/rejected": -0.1647646725177765,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.573338031768799,
      "logits/rejected": -2.4198756217956543,
      "logps/chosen": -172.42422485351562,
      "logps/rejected": -173.40711975097656,
      "loss": 0.6345,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.13879454135894775,
      "rewards/margins": 0.16979987919330597,
      "rewards/rejected": -0.30859440565109253,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": -2.5714492797851562,
      "logits/rejected": -2.4478890895843506,
      "logps/chosen": -178.8824005126953,
      "logps/rejected": -183.39822387695312,
      "loss": 0.6304,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.18915040791034698,
      "rewards/margins": 0.17477598786354065,
      "rewards/rejected": -0.36392638087272644,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": -2.4946584701538086,
      "logits/rejected": -2.3706579208374023,
      "logps/chosen": -199.9522247314453,
      "logps/rejected": -195.98255920410156,
      "loss": 0.631,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.360156387090683,
      "rewards/margins": 0.17627720534801483,
      "rewards/rejected": -0.5364335775375366,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": -2.4831032752990723,
      "logits/rejected": -2.368504285812378,
      "logps/chosen": -189.5545654296875,
      "logps/rejected": -190.16159057617188,
      "loss": 0.6327,
      "rewards/accuracies": 0.621874988079071,
      "rewards/chosen": -0.35114872455596924,
      "rewards/margins": 0.14523856341838837,
      "rewards/rejected": -0.4963873028755188,
      "step": 70
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.4916341304779053,
      "logits/rejected": -2.3981051445007324,
      "logps/chosen": -186.53948974609375,
      "logps/rejected": -197.5457305908203,
      "loss": 0.6266,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.2931613028049469,
      "rewards/margins": 0.19320784509181976,
      "rewards/rejected": -0.48636919260025024,
      "step": 80
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": -2.549828290939331,
      "logits/rejected": -2.365307092666626,
      "logps/chosen": -222.1770477294922,
      "logps/rejected": -225.6692657470703,
      "loss": 0.6123,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.42118874192237854,
      "rewards/margins": 0.27340102195739746,
      "rewards/rejected": -0.6945897340774536,
      "step": 90
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": -2.5293145179748535,
      "logits/rejected": -2.3946516513824463,
      "logps/chosen": -191.87879943847656,
      "logps/rejected": -201.5653839111328,
      "loss": 0.6268,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.3195632994174957,
      "rewards/margins": 0.2095298022031784,
      "rewards/rejected": -0.5290930271148682,
      "step": 100
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.445504665374756,
      "eval_logits/rejected": -2.33659029006958,
      "eval_logps/chosen": -360.80682373046875,
      "eval_logps/rejected": -362.4261169433594,
      "eval_loss": 0.6731494069099426,
      "eval_rewards/accuracies": 0.5699999928474426,
      "eval_rewards/chosen": -0.6240409016609192,
      "eval_rewards/margins": 0.10187112540006638,
      "eval_rewards/rejected": -0.7259120941162109,
      "eval_runtime": 381.8816,
      "eval_samples_per_second": 5.237,
      "eval_steps_per_second": 0.655,
      "step": 100
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": -2.5149567127227783,
      "logits/rejected": -2.3778505325317383,
      "logps/chosen": -199.1842803955078,
      "logps/rejected": -212.69827270507812,
      "loss": 0.6132,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.2891606390476227,
      "rewards/margins": 0.2867799699306488,
      "rewards/rejected": -0.5759405493736267,
      "step": 110
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -2.4683775901794434,
      "logits/rejected": -2.3485958576202393,
      "logps/chosen": -196.60177612304688,
      "logps/rejected": -211.6150360107422,
      "loss": 0.6128,
      "rewards/accuracies": 0.7093750238418579,
      "rewards/chosen": -0.29994842410087585,
      "rewards/margins": 0.27045875787734985,
      "rewards/rejected": -0.5704071521759033,
      "step": 120
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": -2.4233736991882324,
      "logits/rejected": -2.3031530380249023,
      "logps/chosen": -190.49278259277344,
      "logps/rejected": -205.83090209960938,
      "loss": 0.6139,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.3687855005264282,
      "rewards/margins": 0.23157742619514465,
      "rewards/rejected": -0.6003628969192505,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": -2.4012184143066406,
      "logits/rejected": -2.203899621963501,
      "logps/chosen": -198.17774963378906,
      "logps/rejected": -214.18106079101562,
      "loss": 0.604,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.3292800188064575,
      "rewards/margins": 0.3036113381385803,
      "rewards/rejected": -0.6328912973403931,
      "step": 140
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": -2.4568684101104736,
      "logits/rejected": -2.2847976684570312,
      "logps/chosen": -184.83816528320312,
      "logps/rejected": -199.9399871826172,
      "loss": 0.6024,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.33008551597595215,
      "rewards/margins": 0.30454346537590027,
      "rewards/rejected": -0.63462895154953,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.6301614909599988,
      "train_runtime": 6946.965,
      "train_samples_per_second": 2.879,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}