{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.4946695095948828,
  "eval_steps": 500,
  "global_step": 87,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 691.5893173217773,
      "epoch": 0.017057569296375266,
      "grad_norm": 0.6600080728530884,
      "kl": 0.0,
      "learning_rate": 3.333333333333333e-07,
      "loss": 0.0895,
      "reward": 0.21837703324854374,
      "reward_std": 0.457524124532938,
      "rewards/accuracy_reward": 0.4107143059372902,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.19233726896345615,
      "step": 1
    },
    {
      "completion_length": 711.3395977020264,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.521936297416687,
      "kl": 0.00035202503204345703,
      "learning_rate": 1.6666666666666669e-06,
      "loss": 0.0873,
      "reward": 0.19321720697917044,
      "reward_std": 0.4614652991294861,
      "rewards/accuracy_reward": 0.38699778635054827,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.1937805782072246,
      "step": 5
    },
    {
      "completion_length": 684.9652038574219,
      "epoch": 0.17057569296375266,
      "grad_norm": 0.5586162805557251,
      "kl": 0.01878194808959961,
      "learning_rate": 2.9987834972573546e-06,
      "loss": 0.0742,
      "reward": 0.2827843511477113,
      "reward_std": 0.44751987904310225,
      "rewards/accuracy_reward": 0.4611607357859612,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.17837638631463051,
      "step": 10
    },
    {
      "completion_length": 641.0350692749023,
      "epoch": 0.255863539445629,
      "grad_norm": 1.2480870485305786,
      "kl": 0.5136314392089844,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.1195,
      "reward": 0.49980842545628545,
      "reward_std": 0.3398652091622353,
      "rewards/accuracy_reward": 0.6600446701049805,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.16023624502122402,
      "step": 15
    },
    {
      "completion_length": 629.3375289916992,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.5801119208335876,
      "kl": 0.020611572265625,
      "learning_rate": 2.8551756519155732e-06,
      "loss": 0.1008,
      "reward": 0.5805917799472808,
      "reward_std": 0.27404730543494227,
      "rewards/accuracy_reward": 0.7352678969502449,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.15467611663043498,
      "step": 20
    },
    {
      "completion_length": 632.6993621826172,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.5435677766799927,
      "kl": 0.01521759033203125,
      "learning_rate": 2.699164145105252e-06,
      "loss": 0.0827,
      "reward": 0.5681484438478946,
      "reward_std": 0.28131211549043655,
      "rewards/accuracy_reward": 0.710714316368103,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.14256587382405997,
      "step": 25
    },
    {
      "completion_length": 631.0065017700196,
      "epoch": 0.511727078891258,
      "grad_norm": 0.40434423089027405,
      "kl": 0.087030029296875,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0906,
      "reward": 0.5263580977916718,
      "reward_std": 0.3138962801545858,
      "rewards/accuracy_reward": 0.6600446760654449,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.13368657547980548,
      "step": 30
    },
    {
      "completion_length": 617.7948944091797,
      "epoch": 0.5970149253731343,
      "grad_norm": 1.0160555839538574,
      "kl": 0.117022705078125,
      "learning_rate": 2.25e-06,
      "loss": 0.0962,
      "reward": 0.5008847367018461,
      "reward_std": 0.3264832843095064,
      "rewards/accuracy_reward": 0.622321455180645,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12143671978265047,
      "step": 35
    },
    {
      "completion_length": 607.287973022461,
      "epoch": 0.6823027718550106,
      "grad_norm": 1.8936026096343994,
      "kl": 0.2015625,
      "learning_rate": 1.975001990702209e-06,
      "loss": 0.1194,
      "reward": 0.5203140497207641,
      "reward_std": 0.3361724577844143,
      "rewards/accuracy_reward": 0.6433036029338837,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12298955339938403,
      "step": 40
    },
    {
      "completion_length": 604.0877517700195,
      "epoch": 0.767590618336887,
      "grad_norm": 4.448797225952148,
      "kl": 5.82470703125,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.5125,
      "reward": 0.5560206107795238,
      "reward_std": 0.29633302725851535,
      "rewards/accuracy_reward": 0.6729911029338836,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.11697048898786307,
      "step": 45
    },
    {
      "completion_length": 604.8629692077636,
      "epoch": 0.8528784648187633,
      "grad_norm": 2.490539789199829,
      "kl": 0.377392578125,
      "learning_rate": 1.3793001469249112e-06,
      "loss": 0.1237,
      "reward": 0.5681209519505501,
      "reward_std": 0.28387211747467517,
      "rewards/accuracy_reward": 0.6881696730852127,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12004871871322394,
      "step": 50
    },
    {
      "completion_length": 579.4413185119629,
      "epoch": 0.9381663113006397,
      "grad_norm": 4.152231216430664,
      "kl": 0.2768310546875,
      "learning_rate": 1.0826738041253211e-06,
      "loss": 0.0852,
      "reward": 0.6077298022806644,
      "reward_std": 0.27112445086240766,
      "rewards/accuracy_reward": 0.7265625357627868,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.11883272156119347,
      "step": 55
    },
    {
      "completion_length": 566.9164165150036,
      "epoch": 1.0341151385927505,
      "grad_norm": 3.0358667373657227,
      "kl": 713.1028053977273,
      "learning_rate": 8.029152419343472e-07,
      "loss": 82.7243,
      "reward": 0.6099555316296491,
      "reward_std": 0.25437648966908455,
      "rewards/accuracy_reward": 0.7307224382053722,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12076690539040348,
      "step": 60
    },
    {
      "completion_length": 548.3241325378418,
      "epoch": 1.1194029850746268,
      "grad_norm": 2.281466245651245,
      "kl": 0.169915771484375,
      "learning_rate": 5.513319366069343e-07,
      "loss": 0.0759,
      "reward": 0.6292034849524498,
      "reward_std": 0.24837480448186397,
      "rewards/accuracy_reward": 0.7457589656114578,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.11655548624694348,
      "step": 65
    },
    {
      "completion_length": 544.6705657958985,
      "epoch": 1.2046908315565032,
      "grad_norm": 4.551150798797607,
      "kl": 0.168426513671875,
      "learning_rate": 3.380925572585183e-07,
      "loss": 0.081,
      "reward": 0.6241521313786507,
      "reward_std": 0.23203972559422253,
      "rewards/accuracy_reward": 0.7475446790456772,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12339254282414913,
      "step": 70
    },
    {
      "completion_length": 560.6620811462402,
      "epoch": 1.2899786780383795,
      "grad_norm": 1.7139493227005005,
      "kl": 0.295257568359375,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0725,
      "reward": 0.6123189993202687,
      "reward_std": 0.25482493862509725,
      "rewards/accuracy_reward": 0.7330357432365417,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12071674298495054,
      "step": 75
    },
    {
      "completion_length": 538.8129737854003,
      "epoch": 1.375266524520256,
      "grad_norm": 1.8946577310562134,
      "kl": 0.28873291015625,
      "learning_rate": 5.922283255294164e-08,
      "loss": 0.0819,
      "reward": 0.6251922458410263,
      "reward_std": 0.23853217102587224,
      "rewards/accuracy_reward": 0.7475446805357933,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12235244438052177,
      "step": 80
    },
    {
      "completion_length": 552.9801628112793,
      "epoch": 1.4605543710021323,
      "grad_norm": 1.40664541721344,
      "kl": 711.1384460449219,
      "learning_rate": 4.864037798685106e-09,
      "loss": 47.7953,
      "reward": 0.618848904967308,
      "reward_std": 0.24578485451638699,
      "rewards/accuracy_reward": 0.7415178909897804,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12266898676753044,
      "step": 85
    },
    {
      "completion_length": 589.5374145507812,
      "epoch": 1.4946695095948828,
      "kl": 0.252349853515625,
      "reward": 0.5610090717673302,
      "reward_std": 0.24640783108770847,
      "rewards/accuracy_reward": 0.6824777089059353,
      "rewards/format_reward": 0.0,
      "rewards/repetition_penalty_reward": -0.12146863108500838,
      "step": 87,
      "total_flos": 0.0,
      "train_loss": 7.607245717240476,
      "train_runtime": 13150.4833,
      "train_samples_per_second": 0.855,
      "train_steps_per_second": 0.007
    }
  ],
  "logging_steps": 5,
  "max_steps": 87,
  "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
}