{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9893390191897655,
  "eval_steps": 100,
  "global_step": 116,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 1228.861671447754,
      "epoch": 0.017057569296375266,
      "grad_norm": 0.06720272451639175,
      "kl": 0.0,
      "learning_rate": 2.5e-07,
      "loss": 0.0434,
      "reward": 0.6573660969734192,
      "reward_std": 0.18419872969388962,
      "rewards/accuracy_reward": 0.6573660969734192,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1225.2754497528076,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.06428399682044983,
      "kl": 0.00012925267219543457,
      "learning_rate": 1.25e-06,
      "loss": 0.0339,
      "reward": 0.6021205559372902,
      "reward_std": 0.16342801752034575,
      "rewards/accuracy_reward": 0.6021205559372902,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1126.743356323242,
      "epoch": 0.17057569296375266,
      "grad_norm": 0.06763080507516861,
      "kl": 0.00020864009857177735,
      "learning_rate": 2.5e-06,
      "loss": 0.0277,
      "reward": 0.6676339641213417,
      "reward_std": 0.1722445989958942,
      "rewards/accuracy_reward": 0.6676339641213417,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1019.001612854004,
      "epoch": 0.255863539445629,
      "grad_norm": 0.06711237877607346,
      "kl": 0.0011625289916992188,
      "learning_rate": 2.9938448364256362e-06,
      "loss": 0.0245,
      "reward": 0.7089286029338837,
      "reward_std": 0.16926544005982577,
      "rewards/accuracy_reward": 0.7089286029338837,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 911.8040542602539,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.05857737362384796,
      "kl": 0.0029880523681640623,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.0173,
      "reward": 0.7337053924798965,
      "reward_std": 0.1726918333210051,
      "rewards/accuracy_reward": 0.7337053924798965,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 943.9683486938477,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.05815846845507622,
      "kl": 0.003909683227539063,
      "learning_rate": 2.88581929876693e-06,
      "loss": 0.0094,
      "reward": 0.7319196790456772,
      "reward_std": 0.17025341242551803,
      "rewards/accuracy_reward": 0.7319196790456772,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 975.7393280029297,
      "epoch": 0.511727078891258,
      "grad_norm": 0.06264527887105942,
      "kl": 0.006205368041992188,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0197,
      "reward": 0.7424107506871224,
      "reward_std": 0.17087422600015997,
      "rewards/accuracy_reward": 0.7424107506871224,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 959.3453567504882,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.05786522477865219,
      "kl": 0.006237030029296875,
      "learning_rate": 2.652296367060421e-06,
      "loss": 0.0126,
      "reward": 0.7372768208384514,
      "reward_std": 0.15754695991054177,
      "rewards/accuracy_reward": 0.7372768208384514,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 939.0549591064453,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.0710252970457077,
      "kl": 0.006754302978515625,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0152,
      "reward": 0.7680803894996643,
      "reward_std": 0.15489136478863658,
      "rewards/accuracy_reward": 0.7680803894996643,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 944.1554000854492,
      "epoch": 0.767590618336887,
      "grad_norm": 0.06381860375404358,
      "kl": 0.007495880126953125,
      "learning_rate": 2.314423473302218e-06,
      "loss": 0.0181,
      "reward": 0.7691964656114578,
      "reward_std": 0.16262863762676716,
      "rewards/accuracy_reward": 0.7691964656114578,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 952.2839721679687,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.0656137615442276,
      "kl": 0.00856170654296875,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0141,
      "reward": 0.7388393133878708,
      "reward_std": 0.16094484115019442,
      "rewards/accuracy_reward": 0.7388393133878708,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 949.2908905029296,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.061446413397789,
      "kl": 0.00842742919921875,
      "learning_rate": 1.9027978012115653e-06,
      "loss": 0.017,
      "reward": 0.768303607404232,
      "reward_std": 0.15686179576441645,
      "rewards/accuracy_reward": 0.768303607404232,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 936.9299240112305,
      "epoch": 1.0341151385927505,
      "grad_norm": 0.06138957291841507,
      "kl": 0.00886383056640625,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0141,
      "reward": 0.7736607477068901,
      "reward_std": 0.15758724519982933,
      "rewards/accuracy_reward": 0.7736607477068901,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 920.1882110595703,
      "epoch": 1.1194029850746268,
      "grad_norm": 0.063966765999794,
      "kl": 0.009485626220703125,
      "learning_rate": 1.454695458298667e-06,
      "loss": 0.0157,
      "reward": 0.7738839641213417,
      "reward_std": 0.1468926408328116,
      "rewards/accuracy_reward": 0.7738839641213417,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 944.9875381469726,
      "epoch": 1.2046908315565032,
      "grad_norm": 0.08325506001710892,
      "kl": 0.011992645263671876,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0216,
      "reward": 0.7598214663565159,
      "reward_std": 0.13879431886598467,
      "rewards/accuracy_reward": 0.7598214663565159,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 947.0911117553711,
      "epoch": 1.2899786780383795,
      "grad_norm": 0.0815800353884697,
      "kl": 0.01367034912109375,
      "learning_rate": 1.0106958161686963e-06,
      "loss": 0.0215,
      "reward": 0.759375037252903,
      "reward_std": 0.15064039630815387,
      "rewards/accuracy_reward": 0.759375037252903,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 918.9600845336914,
      "epoch": 1.375266524520256,
      "grad_norm": 0.08615086227655411,
      "kl": 0.019728851318359376,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0195,
      "reward": 0.7761161088943481,
      "reward_std": 0.1427250973880291,
      "rewards/accuracy_reward": 0.7761161088943481,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 926.689549255371,
      "epoch": 1.4605543710021323,
      "grad_norm": 0.08062485605478287,
      "kl": 0.01549224853515625,
      "learning_rate": 6.11006712953975e-07,
      "loss": 0.0165,
      "reward": 0.755580386519432,
      "reward_std": 0.1542196120135486,
      "rewards/accuracy_reward": 0.755580386519432,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 951.833529663086,
      "epoch": 1.5458422174840085,
      "grad_norm": 0.08391877263784409,
      "kl": 0.01885833740234375,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0217,
      "reward": 0.745089316368103,
      "reward_std": 0.15332454843446613,
      "rewards/accuracy_reward": 0.745089316368103,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 961.7170059204102,
      "epoch": 1.6311300639658848,
      "grad_norm": 0.10089623183012009,
      "kl": 0.02075958251953125,
      "learning_rate": 2.9182330117358096e-07,
      "loss": 0.026,
      "reward": 0.7386161059141159,
      "reward_std": 0.17298963768407702,
      "rewards/accuracy_reward": 0.7386161059141159,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 95
    },
    {
      "epoch": 1.716417910447761,
      "grad_norm": 0.08750782907009125,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0169,
      "step": 100
    },
    {
      "epoch": 1.716417910447761,
      "eval_clip_ratio": 0.0,
      "eval_completion_length": 908.4477716512954,
      "eval_kl": 0.01996454110922524,
      "eval_loss": 0.018831240013241768,
      "eval_reward": 0.7642058762498557,
      "eval_reward_std": 0.16209243547421293,
      "eval_rewards/accuracy_reward": 0.7642058762498557,
      "eval_rewards/format_reward": 0.0,
      "eval_rewards/len_penalty": 0.0,
      "eval_runtime": 7276.3561,
      "eval_samples_per_second": 0.687,
      "eval_steps_per_second": 0.006,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 948.437540435791,
      "epoch": 1.8017057569296375,
      "grad_norm": 0.07978012412786484,
      "kl": 0.0216644287109375,
      "learning_rate": 8.20502774480395e-08,
      "loss": 0.0227,
      "reward": 0.7496652126312255,
      "reward_std": 0.1498533170670271,
      "rewards/accuracy_reward": 0.7496652126312255,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 941.6007141113281,
      "epoch": 1.886993603411514,
      "grad_norm": 0.08210857957601547,
      "kl": 0.02252197265625,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0152,
      "reward": 0.7466518208384514,
      "reward_std": 0.1587499282322824,
      "rewards/accuracy_reward": 0.7466518208384514,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 948.6848648071289,
      "epoch": 1.9722814498933903,
      "grad_norm": 0.10769649595022202,
      "kl": 0.02285919189453125,
      "learning_rate": 6.843232656998933e-10,
      "loss": 0.0252,
      "reward": 0.750669677555561,
      "reward_std": 0.16108589163050055,
      "rewards/accuracy_reward": 0.750669677555561,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 983.2754592895508,
      "epoch": 1.9893390191897655,
      "kl": 0.02458953857421875,
      "reward": 0.7120535969734192,
      "reward_std": 0.1434549919795245,
      "rewards/accuracy_reward": 0.7120535969734192,
      "rewards/format_reward": 0.0,
      "rewards/len_penalty": 0.0,
      "step": 116,
      "total_flos": 0.0,
      "train_loss": 0.019923711259817255,
      "train_runtime": 38027.4532,
      "train_samples_per_second": 0.394,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 5,
  "max_steps": 116,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 10,
  "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": 16,
  "trial_name": null,
  "trial_params": null
}