{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.592,
  "eval_steps": 50,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 3.3149960041046143,
      "learning_rate": 5e-06,
      "logits/chosen": -0.24067819118499756,
      "logits/rejected": -0.4968351423740387,
      "logps/chosen": -433.0858459472656,
      "logps/rejected": -68.33470153808594,
      "loss": 0.1722,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.1653084754943848,
      "rewards/margins": 1.9695370197296143,
      "rewards/rejected": 0.19577142596244812,
      "step": 5
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.4687750339508057,
      "learning_rate": 1e-05,
      "logits/chosen": -0.23465164005756378,
      "logits/rejected": -0.5149508118629456,
      "logps/chosen": -433.3921813964844,
      "logps/rejected": -65.57392883300781,
      "loss": 0.1348,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 2.6356241703033447,
      "rewards/margins": 2.478445529937744,
      "rewards/rejected": 0.15717869997024536,
      "step": 10
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4518139958381653,
      "learning_rate": 1.5e-05,
      "logits/chosen": -0.22109150886535645,
      "logits/rejected": -0.5422734022140503,
      "logps/chosen": -421.8102111816406,
      "logps/rejected": -68.03514099121094,
      "loss": 0.0432,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.584801197052002,
      "rewards/margins": 3.5185985565185547,
      "rewards/rejected": 0.06620248407125473,
      "step": 15
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.12663038074970245,
      "learning_rate": 2e-05,
      "logits/chosen": -0.21289744973182678,
      "logits/rejected": -0.5218192338943481,
      "logps/chosen": -414.16058349609375,
      "logps/rejected": -72.08072662353516,
      "loss": 0.0138,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.662674903869629,
      "rewards/margins": 4.634940147399902,
      "rewards/rejected": 0.02773415483534336,
      "step": 20
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.043434809893369675,
      "learning_rate": 2.5e-05,
      "logits/chosen": -0.1724880337715149,
      "logits/rejected": -0.632266640663147,
      "logps/chosen": -391.53204345703125,
      "logps/rejected": -71.50444030761719,
      "loss": 0.0039,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.770969390869141,
      "rewards/margins": 5.899745464324951,
      "rewards/rejected": -0.12877611815929413,
      "step": 25
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.012639075517654419,
      "learning_rate": 3e-05,
      "logits/chosen": -0.11160198599100113,
      "logits/rejected": -0.46575218439102173,
      "logps/chosen": -387.0035095214844,
      "logps/rejected": -68.87992095947266,
      "loss": 0.0012,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 7.1022233963012695,
      "rewards/margins": 7.20810604095459,
      "rewards/rejected": -0.10588403791189194,
      "step": 30
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.008335842750966549,
      "learning_rate": 3.5e-05,
      "logits/chosen": -0.16194215416908264,
      "logits/rejected": -0.5091412663459778,
      "logps/chosen": -380.4056701660156,
      "logps/rejected": -70.05772399902344,
      "loss": 0.0007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 7.549788475036621,
      "rewards/margins": 7.717337608337402,
      "rewards/rejected": -0.16754867136478424,
      "step": 35
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.0034919867757707834,
      "learning_rate": 4e-05,
      "logits/chosen": -0.12451864778995514,
      "logits/rejected": -0.4964370131492615,
      "logps/chosen": -379.21673583984375,
      "logps/rejected": -69.48336029052734,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.368996620178223,
      "rewards/margins": 8.601344108581543,
      "rewards/rejected": -0.23234805464744568,
      "step": 40
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.0023931912146508694,
      "learning_rate": 4.5e-05,
      "logits/chosen": -0.11491024494171143,
      "logits/rejected": -0.5707582831382751,
      "logps/chosen": -367.6617431640625,
      "logps/rejected": -71.28264617919922,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.458444595336914,
      "rewards/margins": 8.877108573913574,
      "rewards/rejected": -0.41866397857666016,
      "step": 45
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.002425891114398837,
      "learning_rate": 5e-05,
      "logits/chosen": -0.1450928896665573,
      "logits/rejected": -0.6191288232803345,
      "logps/chosen": -371.6398620605469,
      "logps/rejected": -75.66804504394531,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.693218231201172,
      "rewards/margins": 9.168657302856445,
      "rewards/rejected": -0.4754392206668854,
      "step": 50
    },
    {
      "epoch": 0.8,
      "eval_logits/chosen": -0.09945501387119293,
      "eval_logits/rejected": -0.5488065481185913,
      "eval_logps/chosen": -366.370849609375,
      "eval_logps/rejected": -72.13478088378906,
      "eval_loss": 0.0001304554898524657,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 8.9785737991333,
      "eval_rewards/margins": 9.362648963928223,
      "eval_rewards/rejected": -0.3840752840042114,
      "eval_runtime": 31.4719,
      "eval_samples_per_second": 3.177,
      "eval_steps_per_second": 0.794,
      "step": 50
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.0031152062583714724,
      "learning_rate": 5.500000000000001e-05,
      "logits/chosen": -0.10387110710144043,
      "logits/rejected": -0.5721521377563477,
      "logps/chosen": -367.0964660644531,
      "logps/rejected": -73.10216522216797,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 8.92541217803955,
      "rewards/margins": 9.32882308959961,
      "rewards/rejected": -0.40341073274612427,
      "step": 55
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.0020932538900524378,
      "learning_rate": 6e-05,
      "logits/chosen": -0.09782540053129196,
      "logits/rejected": -0.634772002696991,
      "logps/chosen": -365.34393310546875,
      "logps/rejected": -74.85542297363281,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.201251983642578,
      "rewards/margins": 9.6889009475708,
      "rewards/rejected": -0.48764729499816895,
      "step": 60
    },
    {
      "epoch": 1.032,
      "grad_norm": 0.0024136879947036505,
      "learning_rate": 6.500000000000001e-05,
      "logits/chosen": -0.06437685340642929,
      "logits/rejected": -0.5220383405685425,
      "logps/chosen": -371.1302490234375,
      "logps/rejected": -70.0087890625,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.469672203063965,
      "rewards/margins": 9.84286880493164,
      "rewards/rejected": -0.37319669127464294,
      "step": 65
    },
    {
      "epoch": 1.112,
      "grad_norm": 0.0008256471483036876,
      "learning_rate": 7e-05,
      "logits/chosen": -0.10976336151361465,
      "logits/rejected": -0.47059911489486694,
      "logps/chosen": -359.65667724609375,
      "logps/rejected": -67.18486785888672,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.607645988464355,
      "rewards/margins": 10.04837703704834,
      "rewards/rejected": -0.4407329559326172,
      "step": 70
    },
    {
      "epoch": 1.192,
      "grad_norm": 0.003003115998581052,
      "learning_rate": 7.500000000000001e-05,
      "logits/chosen": -0.1105552464723587,
      "logits/rejected": -0.6176896095275879,
      "logps/chosen": -365.54412841796875,
      "logps/rejected": -81.55340576171875,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.260071754455566,
      "rewards/margins": 9.807526588439941,
      "rewards/rejected": -0.547455906867981,
      "step": 75
    },
    {
      "epoch": 1.272,
      "grad_norm": 0.0006993189454078674,
      "learning_rate": 8e-05,
      "logits/chosen": -0.10653767734766006,
      "logits/rejected": -0.5204580426216125,
      "logps/chosen": -364.65301513671875,
      "logps/rejected": -69.9474105834961,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.50252628326416,
      "rewards/margins": 9.944330215454102,
      "rewards/rejected": -0.441803514957428,
      "step": 80
    },
    {
      "epoch": 1.3519999999999999,
      "grad_norm": 0.0006063419277779758,
      "learning_rate": 8.5e-05,
      "logits/chosen": -0.09719086438417435,
      "logits/rejected": -0.5659133791923523,
      "logps/chosen": -360.6310119628906,
      "logps/rejected": -71.894775390625,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.610506057739258,
      "rewards/margins": 10.054449081420898,
      "rewards/rejected": -0.4439435601234436,
      "step": 85
    },
    {
      "epoch": 1.432,
      "grad_norm": 0.001517809578217566,
      "learning_rate": 9e-05,
      "logits/chosen": -0.06048806384205818,
      "logits/rejected": -0.6155588626861572,
      "logps/chosen": -361.54010009765625,
      "logps/rejected": -69.9623031616211,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.503868103027344,
      "rewards/margins": 9.950047492980957,
      "rewards/rejected": -0.4461793303489685,
      "step": 90
    },
    {
      "epoch": 1.512,
      "grad_norm": 0.0008690144750289619,
      "learning_rate": 9.5e-05,
      "logits/chosen": -0.09672050923109055,
      "logits/rejected": -0.6058477759361267,
      "logps/chosen": -362.9765625,
      "logps/rejected": -75.24437713623047,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.525094032287598,
      "rewards/margins": 9.997543334960938,
      "rewards/rejected": -0.47244787216186523,
      "step": 95
    },
    {
      "epoch": 1.592,
      "grad_norm": 0.0008889890741556883,
      "learning_rate": 0.0001,
      "logits/chosen": -0.09034673869609833,
      "logits/rejected": -0.49319934844970703,
      "logps/chosen": -358.0984802246094,
      "logps/rejected": -73.61305236816406,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 9.614057540893555,
      "rewards/margins": 10.11325740814209,
      "rewards/rejected": -0.4991990625858307,
      "step": 100
    },
    {
      "epoch": 1.592,
      "eval_logits/chosen": -0.08441560715436935,
      "eval_logits/rejected": -0.5524138808250427,
      "eval_logps/chosen": -359.42413330078125,
      "eval_logps/rejected": -73.16493225097656,
      "eval_loss": 5.7459325034869835e-05,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 9.673240661621094,
      "eval_rewards/margins": 10.160331726074219,
      "eval_rewards/rejected": -0.48709091544151306,
      "eval_runtime": 31.6934,
      "eval_samples_per_second": 3.155,
      "eval_steps_per_second": 0.789,
      "step": 100
    }
  ],
  "logging_steps": 5,
  "max_steps": 186,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}