{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.971563981042654,
  "eval_steps": 500,
  "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": 18.54137556493684,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": -1.3836915493011475,
      "logits/rejected": -1.3647675514221191,
      "logps/chosen": -369.8086242675781,
      "logps/rejected": -456.2351989746094,
      "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": 18.357440019150488,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": -1.3435577154159546,
      "logits/rejected": -1.3215419054031372,
      "logps/chosen": -312.6362609863281,
      "logps/rejected": -380.2178039550781,
      "loss": 0.6925,
      "rewards/accuracies": 0.4722222089767456,
      "rewards/chosen": 0.0017008990980684757,
      "rewards/margins": 0.003187247784808278,
      "rewards/rejected": -0.0014863491524010897,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 16.59445996406656,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": -1.3366063833236694,
      "logits/rejected": -1.3214499950408936,
      "logps/chosen": -345.36834716796875,
      "logps/rejected": -392.05963134765625,
      "loss": 0.6751,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 0.056390948593616486,
      "rewards/margins": 0.037608057260513306,
      "rewards/rejected": 0.01878289319574833,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 17.22094713854238,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": -1.3375577926635742,
      "logits/rejected": -1.3127400875091553,
      "logps/chosen": -305.53546142578125,
      "logps/rejected": -315.6367492675781,
      "loss": 0.6308,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 0.1503826230764389,
      "rewards/margins": 0.11851520836353302,
      "rewards/rejected": 0.03186742216348648,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 14.166324855532773,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": -1.3806421756744385,
      "logits/rejected": -1.3631458282470703,
      "logps/chosen": -326.44732666015625,
      "logps/rejected": -320.19818115234375,
      "loss": 0.599,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": 0.34801143407821655,
      "rewards/margins": 0.3300670087337494,
      "rewards/rejected": 0.01794440858066082,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 15.139763404001993,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": -1.362728476524353,
      "logits/rejected": -1.3629658222198486,
      "logps/chosen": -323.0518493652344,
      "logps/rejected": -405.2288513183594,
      "loss": 0.5532,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": 0.43860048055648804,
      "rewards/margins": 0.3733087182044983,
      "rewards/rejected": 0.06529171764850616,
      "step": 50
    },
    {
      "epoch": 0.985781990521327,
      "eval_logits/chosen": -1.3316012620925903,
      "eval_logits/rejected": -1.3205280303955078,
      "eval_logps/chosen": -276.7043151855469,
      "eval_logps/rejected": -322.6759948730469,
      "eval_loss": 0.5591163039207458,
      "eval_rewards/accuracies": 0.7604166865348816,
      "eval_rewards/chosen": 0.4980466663837433,
      "eval_rewards/margins": 0.42884430289268494,
      "eval_rewards/rejected": 0.06920242309570312,
      "eval_runtime": 102.7929,
      "eval_samples_per_second": 7.296,
      "eval_steps_per_second": 0.117,
      "step": 52
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 14.364642887348316,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": -1.401728630065918,
      "logits/rejected": -1.378286361694336,
      "logps/chosen": -322.6551208496094,
      "logps/rejected": -355.2884826660156,
      "loss": 0.5136,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": 0.5107188820838928,
      "rewards/margins": 0.5089418292045593,
      "rewards/rejected": 0.0017769678961485624,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 16.536036390235253,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": -1.3928285837173462,
      "logits/rejected": -1.377036690711975,
      "logps/chosen": -287.08209228515625,
      "logps/rejected": -335.96124267578125,
      "loss": 0.5006,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": 0.5614706873893738,
      "rewards/margins": 0.6524538397789001,
      "rewards/rejected": -0.09098317474126816,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 13.006389392455016,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": -1.3728402853012085,
      "logits/rejected": -1.3621151447296143,
      "logps/chosen": -333.00079345703125,
      "logps/rejected": -385.9339599609375,
      "loss": 0.4643,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": 0.5941373705863953,
      "rewards/margins": 0.8581901788711548,
      "rewards/rejected": -0.26405268907546997,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 14.37088192786904,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": -1.39938223361969,
      "logits/rejected": -1.3864850997924805,
      "logps/chosen": -325.26055908203125,
      "logps/rejected": -378.0798645019531,
      "loss": 0.4576,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": 0.4705162048339844,
      "rewards/margins": 0.7767011523246765,
      "rewards/rejected": -0.30618494749069214,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 14.30291539625333,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": -1.3900349140167236,
      "logits/rejected": -1.3774442672729492,
      "logps/chosen": -304.0644836425781,
      "logps/rejected": -361.84539794921875,
      "loss": 0.4562,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": 0.5220262408256531,
      "rewards/margins": 0.7719391584396362,
      "rewards/rejected": -0.24991300702095032,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "eval_logits/chosen": -1.3361939191818237,
      "eval_logits/rejected": -1.3248072862625122,
      "eval_logps/chosen": -277.1591491699219,
      "eval_logps/rejected": -326.42169189453125,
      "eval_loss": 0.5286681056022644,
      "eval_rewards/accuracies": 0.7708333134651184,
      "eval_rewards/chosen": 0.47530409693717957,
      "eval_rewards/margins": 0.5933875441551208,
      "eval_rewards/rejected": -0.11808355897665024,
      "eval_runtime": 94.773,
      "eval_samples_per_second": 7.914,
      "eval_steps_per_second": 0.127,
      "step": 104
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.5530139460013463,
      "train_runtime": 7803.0424,
      "train_samples_per_second": 1.73,
      "train_steps_per_second": 0.013
    }
  ],
  "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": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}