{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9893390191897654,
  "eval_steps": 500,
  "global_step": 58,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.3348236083984,
      "epoch": 0.017057569296375266,
      "grad_norm": 295.6246337890625,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0007,
      "reward": 1.3604911267757416,
      "reward_std": 0.43993261456489563,
      "rewards/accuracy_reward": 0.4196428768336773,
      "rewards/format_reward": 0.9408482611179352,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.4154644012451,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.2173105925321579,
      "kl": 0.0015696585178375244,
      "learning_rate": 2.5e-06,
      "loss": 0.0003,
      "reward": 1.36104916036129,
      "reward_std": 0.4599178032949567,
      "rewards/accuracy_reward": 0.4065290354192257,
      "rewards/format_reward": 0.9545201268047094,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.4203186035156,
      "epoch": 0.17057569296375266,
      "grad_norm": 0.3614032566547394,
      "kl": 80.7047048330307,
      "learning_rate": 2.956412726139078e-06,
      "loss": 3.2442,
      "reward": 1.4700893461704254,
      "reward_std": 0.42430560290813446,
      "rewards/accuracy_reward": 0.5145089522004127,
      "rewards/format_reward": 0.9555803969502449,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.5118347167969,
      "epoch": 0.255863539445629,
      "grad_norm": 5230.3447265625,
      "kl": 0.45456657409667967,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0182,
      "reward": 1.5975447058677674,
      "reward_std": 0.3535365674644709,
      "rewards/accuracy_reward": 0.6444196701049805,
      "rewards/format_reward": 0.9531250447034836,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.7593780517578,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.2595553398132324,
      "kl": 3.4332809448242188,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.1369,
      "reward": 1.663169714808464,
      "reward_std": 0.30231964513659476,
      "rewards/accuracy_reward": 0.7095982447266579,
      "rewards/format_reward": 0.9535714656114578,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.8439743041993,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.1941014677286148,
      "kl": 0.010175323486328125,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0004,
      "reward": 1.632812574505806,
      "reward_std": 0.3012897618114948,
      "rewards/accuracy_reward": 0.6939732454717159,
      "rewards/format_reward": 0.9388393223285675,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.8616088867187,
      "epoch": 0.511727078891258,
      "grad_norm": 0.18191584944725037,
      "kl": 0.021715545654296876,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0008,
      "reward": 1.6042411476373672,
      "reward_std": 0.3133077774196863,
      "rewards/accuracy_reward": 0.6660714574158192,
      "rewards/format_reward": 0.9381696850061416,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.3654098510742,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.32974478602409363,
      "kl": 0.02928009033203125,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.001,
      "reward": 1.5727679282426834,
      "reward_std": 0.3634680099785328,
      "rewards/accuracy_reward": 0.6535714581608772,
      "rewards/format_reward": 0.9191964700818062,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.0618392944336,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.38136547803878784,
      "kl": 0.0487884521484375,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0015,
      "reward": 1.5879464954137803,
      "reward_std": 0.3696904189884663,
      "rewards/accuracy_reward": 0.6725446730852127,
      "rewards/format_reward": 0.9154018253087998,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1021.6725723266602,
      "epoch": 0.767590618336887,
      "grad_norm": 0.5596392750740051,
      "kl": 0.0637298583984375,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0009,
      "reward": 1.5805804252624511,
      "reward_std": 0.3978343676775694,
      "rewards/accuracy_reward": 0.6763393178582191,
      "rewards/format_reward": 0.9042411163449288,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1020.4614181518555,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.46670985221862793,
      "kl": 0.07415771484375,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0019,
      "reward": 1.5774554371833802,
      "reward_std": 0.40023380927741525,
      "rewards/accuracy_reward": 0.6667411029338837,
      "rewards/format_reward": 0.9107143297791481,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1020.0493682861328,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.28767386078834534,
      "kl": 0.143035888671875,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0027,
      "reward": 1.6111607879400254,
      "reward_std": 0.3957536414265633,
      "rewards/accuracy_reward": 0.6975446730852127,
      "rewards/format_reward": 0.9136161133646965,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1021.3299789428711,
      "epoch": 0.9893390191897654,
      "kl": 0.07426961263020833,
      "reward": 1.6130953133106232,
      "reward_std": 0.3950556789835294,
      "rewards/accuracy_reward": 0.691592293481032,
      "rewards/format_reward": 0.9215030198295912,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.294493970670938,
      "train_runtime": 20554.7381,
      "train_samples_per_second": 0.365,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 5,
  "max_steps": 58,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 16,
  "trial_name": null,
  "trial_params": null
}