{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.971563981042654,
  "eval_steps": 100,
  "global_step": 104,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018957345971563982,
      "grad_norm": 13.317678186469232,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": -25.944135665893555,
      "logits/rejected": -27.06336784362793,
      "logps/chosen": -347.4261169433594,
      "logps/rejected": -454.129150390625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 66.11905089623708,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": -27.457122802734375,
      "logits/rejected": -27.471702575683594,
      "logps/chosen": -419.2057189941406,
      "logps/rejected": -449.7756652832031,
      "loss": 0.6937,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.00854549091309309,
      "rewards/margins": 0.0070810383185744286,
      "rewards/rejected": 0.0014644513139501214,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 43.91594432419604,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": -26.719680786132812,
      "logits/rejected": -27.048105239868164,
      "logps/chosen": -394.787353515625,
      "logps/rejected": -429.10382080078125,
      "loss": 0.6928,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.004536194261163473,
      "rewards/margins": -0.0009250607108697295,
      "rewards/rejected": -0.0036111343652009964,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 15.290145570441174,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": -28.016571044921875,
      "logits/rejected": -27.224132537841797,
      "logps/chosen": -421.117431640625,
      "logps/rejected": -425.31207275390625,
      "loss": 0.6843,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": 0.0752100795507431,
      "rewards/margins": 0.018335092812776566,
      "rewards/rejected": 0.05687499791383743,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 13.150827948521568,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": -27.810794830322266,
      "logits/rejected": -26.933706283569336,
      "logps/chosen": -377.9217834472656,
      "logps/rejected": -367.1111755371094,
      "loss": 0.6799,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.012064946815371513,
      "rewards/margins": 0.044392455369234085,
      "rewards/rejected": -0.056457407772541046,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 18.18319545383219,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": -28.838796615600586,
      "logits/rejected": -28.732458114624023,
      "logps/chosen": -442.86083984375,
      "logps/rejected": -489.5055236816406,
      "loss": 0.6663,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.039321351796388626,
      "rewards/margins": 0.044025082141160965,
      "rewards/rejected": -0.08334643393754959,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 13.351734828868349,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": -28.462390899658203,
      "logits/rejected": -28.101354598999023,
      "logps/chosen": -431.38885498046875,
      "logps/rejected": -449.32867431640625,
      "loss": 0.6529,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": 0.012722885236144066,
      "rewards/margins": 0.12798075377941132,
      "rewards/rejected": -0.1152578741312027,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 13.766301197572265,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": -27.991195678710938,
      "logits/rejected": -27.877147674560547,
      "logps/chosen": -396.7119140625,
      "logps/rejected": -426.5439453125,
      "loss": 0.6493,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.02727600559592247,
      "rewards/margins": 0.09371577203273773,
      "rewards/rejected": -0.1209917664527893,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 12.470159243647668,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": -27.605504989624023,
      "logits/rejected": -28.168981552124023,
      "logps/chosen": -374.8273620605469,
      "logps/rejected": -418.93878173828125,
      "loss": 0.6324,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.03951810300350189,
      "rewards/margins": 0.1497572511434555,
      "rewards/rejected": -0.1892753541469574,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 13.106571636014174,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": -27.413257598876953,
      "logits/rejected": -28.177936553955078,
      "logps/chosen": -412.2579040527344,
      "logps/rejected": -455.0975036621094,
      "loss": 0.634,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.00258343736641109,
      "rewards/margins": 0.162180095911026,
      "rewards/rejected": -0.15959665179252625,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 11.807760701507803,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": -27.838769912719727,
      "logits/rejected": -27.689228057861328,
      "logps/chosen": -404.11578369140625,
      "logps/rejected": -432.09033203125,
      "loss": 0.6264,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.01369890570640564,
      "rewards/margins": 0.14452625811100006,
      "rewards/rejected": -0.1582251489162445,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": -26.95292091369629,
      "eval_logits/rejected": -25.980209350585938,
      "eval_logps/chosen": -386.1443786621094,
      "eval_logps/rejected": -378.4776306152344,
      "eval_loss": 0.64967942237854,
      "eval_rewards/accuracies": 0.6354166865348816,
      "eval_rewards/chosen": -0.039478909224271774,
      "eval_rewards/margins": 0.09327935427427292,
      "eval_rewards/rejected": -0.1327582597732544,
      "eval_runtime": 8.4783,
      "eval_samples_per_second": 88.461,
      "eval_steps_per_second": 2.831,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.661209049133154,
      "train_runtime": 432.2704,
      "train_samples_per_second": 31.23,
      "train_steps_per_second": 0.241
    }
  ],
  "logging_steps": 10,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}