{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.010022222222222222,
  "eval_steps": 500,
  "global_step": 451,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 342.55,
      "epoch": 0.00022222222222222223,
      "grad_norm": 0.21378593146800995,
      "kl": 0.0005054072127677501,
      "learning_rate": 4.99391012564956e-07,
      "loss": 0.0,
      "reward": 0.25,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.25,
      "step": 10
    },
    {
      "completion_length": 337.3,
      "epoch": 0.00044444444444444447,
      "grad_norm": 1.5290135706891306e-05,
      "kl": 0.0005718494998291135,
      "learning_rate": 4.975670171853925e-07,
      "loss": 0.0,
      "reward": 0.65,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.65,
      "step": 20
    },
    {
      "completion_length": 388.8,
      "epoch": 0.0006666666666666666,
      "grad_norm": 0.31574714183807373,
      "kl": 0.0005932338128332049,
      "learning_rate": 4.945369001834514e-07,
      "loss": 0.0,
      "reward": 0.35,
      "reward_std": 0.49497473835945127,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.35,
      "step": 30
    },
    {
      "completion_length": 390.3,
      "epoch": 0.0008888888888888889,
      "grad_norm": 1.8352753613726236e-05,
      "kl": 0.0006036393286194652,
      "learning_rate": 4.903154239845797e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.1414213538169861,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 40
    },
    {
      "completion_length": 466.7,
      "epoch": 0.0011111111111111111,
      "grad_norm": 1.8186246961704455e-05,
      "kl": 0.0005532185896299779,
      "learning_rate": 4.849231551964771e-07,
      "loss": 0.0,
      "reward": 0.5,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.5,
      "step": 50
    },
    {
      "completion_length": 446.8,
      "epoch": 0.0013333333333333333,
      "grad_norm": 3.2005907996790484e-05,
      "kl": 0.0006301910732872784,
      "learning_rate": 4.783863644106502e-07,
      "loss": 0.0,
      "reward": 0.35,
      "reward_std": 0.49497473835945127,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.35,
      "step": 60
    },
    {
      "completion_length": 371.15,
      "epoch": 0.0015555555555555555,
      "grad_norm": 1.834248541854322e-05,
      "kl": 0.0005721298075513914,
      "learning_rate": 4.707368982147317e-07,
      "loss": 0.0,
      "reward": 0.3,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.3,
      "step": 70
    },
    {
      "completion_length": 380.9,
      "epoch": 0.0017777777777777779,
      "grad_norm": 0.24635720252990723,
      "kl": 0.0005963630188489333,
      "learning_rate": 4.6201202403910643e-07,
      "loss": 0.0,
      "reward": 0.45,
      "reward_std": 0.6363960921764373,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.45,
      "step": 80
    },
    {
      "completion_length": 371.1,
      "epoch": 0.002,
      "grad_norm": 0.20243793725967407,
      "kl": 0.0006782339944038541,
      "learning_rate": 4.5225424859373684e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 90
    },
    {
      "completion_length": 319.75,
      "epoch": 0.0022222222222222222,
      "grad_norm": 0.2666068971157074,
      "kl": 0.0006597590865567327,
      "learning_rate": 4.415111107797445e-07,
      "loss": 0.0,
      "reward": 0.5,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.5,
      "step": 100
    },
    {
      "completion_length": 464.1,
      "epoch": 0.0024444444444444444,
      "grad_norm": 0.19521790742874146,
      "kl": 0.0005753735254984349,
      "learning_rate": 4.2983495008466273e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 110
    },
    {
      "completion_length": 413.25,
      "epoch": 0.0026666666666666666,
      "grad_norm": 0.18933118879795074,
      "kl": 0.0005773572018370032,
      "learning_rate": 4.172826515897145e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 120
    },
    {
      "completion_length": 422.75,
      "epoch": 0.0028888888888888888,
      "grad_norm": 0.2125905156135559,
      "kl": 0.0006250702834222465,
      "learning_rate": 4.039153688314145e-07,
      "loss": 0.0,
      "reward": 0.45,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.45,
      "step": 130
    },
    {
      "completion_length": 403.05,
      "epoch": 0.003111111111111111,
      "grad_norm": 1.8821670892066322e-05,
      "kl": 0.0005768304574303329,
      "learning_rate": 3.8979822586768666e-07,
      "loss": 0.0,
      "reward": 0.3,
      "reward_std": 0.1414213538169861,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.3,
      "step": 140
    },
    {
      "completion_length": 445.75,
      "epoch": 0.0033333333333333335,
      "grad_norm": 4.957288547302596e-05,
      "kl": 0.000591657921904698,
      "learning_rate": 3.75e-07,
      "loss": 0.0,
      "reward": 0.35,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.35,
      "step": 150
    },
    {
      "completion_length": 409.8,
      "epoch": 0.0035555555555555557,
      "grad_norm": 0.20501621067523956,
      "kl": 0.0005684028874384239,
      "learning_rate": 3.595927866972693e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 160
    },
    {
      "completion_length": 364.95,
      "epoch": 0.003777777777777778,
      "grad_norm": 0.3308781385421753,
      "kl": 0.0005750875250669196,
      "learning_rate": 3.43651648353978e-07,
      "loss": 0.0,
      "reward": 0.6,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.6,
      "step": 170
    },
    {
      "completion_length": 310.55,
      "epoch": 0.004,
      "grad_norm": 9.957168913388159e-06,
      "kl": 0.0005386626056861132,
      "learning_rate": 3.272542485937368e-07,
      "loss": 0.0,
      "reward": 0.5,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.5,
      "step": 180
    },
    {
      "completion_length": 378.6,
      "epoch": 0.004222222222222222,
      "grad_norm": 0.2291877716779709,
      "kl": 0.0006608614959986881,
      "learning_rate": 3.104804738999169e-07,
      "loss": 0.0,
      "reward": 0.55,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.55,
      "step": 190
    },
    {
      "completion_length": 340.6,
      "epoch": 0.0044444444444444444,
      "grad_norm": 0.18360783159732819,
      "kl": 0.0005953012092504651,
      "learning_rate": 2.934120444167326e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.5656854152679444,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 200
    },
    {
      "completion_length": 396.55,
      "epoch": 0.004666666666666667,
      "grad_norm": 2.1740559532190673e-05,
      "kl": 0.0005484546738443897,
      "learning_rate": 2.761321158169134e-07,
      "loss": 0.0,
      "reward": 0.55,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.55,
      "step": 210
    },
    {
      "completion_length": 330.25,
      "epoch": 0.004888888888888889,
      "grad_norm": 6.363703869283199e-05,
      "kl": 0.0006559955596458167,
      "learning_rate": 2.5872487417562527e-07,
      "loss": 0.0,
      "reward": 0.35,
      "reward_std": 0.49497473835945127,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.35,
      "step": 220
    },
    {
      "completion_length": 419.25,
      "epoch": 0.005111111111111111,
      "grad_norm": 0.1977580189704895,
      "kl": 0.0006136057461844757,
      "learning_rate": 2.412751258243748e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 230
    },
    {
      "completion_length": 399.65,
      "epoch": 0.005333333333333333,
      "grad_norm": 2.009019226534292e-05,
      "kl": 0.000605167931644246,
      "learning_rate": 2.2386788418308665e-07,
      "loss": 0.0,
      "reward": 0.5,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.5,
      "step": 240
    },
    {
      "completion_length": 365.25,
      "epoch": 0.005555555555555556,
      "grad_norm": 9.316376963397488e-05,
      "kl": 0.0005963615287328139,
      "learning_rate": 2.065879555832674e-07,
      "loss": 0.0,
      "reward": 0.6,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.6,
      "step": 250
    },
    {
      "completion_length": 422.1,
      "epoch": 0.0057777777777777775,
      "grad_norm": 0.16928204894065857,
      "kl": 0.0006461634649895132,
      "learning_rate": 1.895195261000831e-07,
      "loss": 0.0,
      "reward": 0.7,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.7,
      "step": 260
    },
    {
      "completion_length": 445.45,
      "epoch": 0.006,
      "grad_norm": 0.17555122077465057,
      "kl": 0.0005888437968678773,
      "learning_rate": 1.7274575140626315e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 270
    },
    {
      "completion_length": 409.55,
      "epoch": 0.006222222222222222,
      "grad_norm": 2.910105285991449e-05,
      "kl": 0.0005833886214531958,
      "learning_rate": 1.5634835164602196e-07,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 280
    },
    {
      "completion_length": 355.3,
      "epoch": 0.0064444444444444445,
      "grad_norm": 0.2147456258535385,
      "kl": 0.000609121096204035,
      "learning_rate": 1.404072133027306e-07,
      "loss": 0.0,
      "reward": 0.45,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.45,
      "step": 290
    },
    {
      "completion_length": 380.9,
      "epoch": 0.006666666666666667,
      "grad_norm": 2.5209470550180413e-05,
      "kl": 0.0006293962011113763,
      "learning_rate": 1.2500000000000005e-07,
      "loss": 0.0,
      "reward": 0.6,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.6,
      "step": 300
    },
    {
      "completion_length": 385.7,
      "epoch": 0.006888888888888889,
      "grad_norm": 7.603858830407262e-05,
      "kl": 0.0006043465342372656,
      "learning_rate": 1.1020177413231332e-07,
      "loss": 0.0,
      "reward": 0.35,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.35,
      "step": 310
    },
    {
      "completion_length": 369.15,
      "epoch": 0.0071111111111111115,
      "grad_norm": 1.2851673091063276e-05,
      "kl": 0.0005439497355837375,
      "learning_rate": 9.608463116858542e-08,
      "loss": 0.0,
      "reward": 0.3,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.3,
      "step": 320
    },
    {
      "completion_length": 316.85,
      "epoch": 0.007333333333333333,
      "grad_norm": 0.20440705120563507,
      "kl": 0.0005756207800004631,
      "learning_rate": 8.271734841028552e-08,
      "loss": 0.0,
      "reward": 0.75,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.75,
      "step": 330
    },
    {
      "completion_length": 429.55,
      "epoch": 0.007555555555555556,
      "grad_norm": 0.2662304639816284,
      "kl": 0.0005803581763757393,
      "learning_rate": 7.016504991533726e-08,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.5656854152679444,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 340
    },
    {
      "completion_length": 467.8,
      "epoch": 0.0077777777777777776,
      "grad_norm": 1.7572709111846052e-05,
      "kl": 0.0005397438129875809,
      "learning_rate": 5.848888922025552e-08,
      "loss": 0.0,
      "reward": 0.35,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.35,
      "step": 350
    },
    {
      "completion_length": 430.1,
      "epoch": 0.008,
      "grad_norm": 1.761410203471314e-05,
      "kl": 0.0005661552131641656,
      "learning_rate": 4.774575140626316e-08,
      "loss": 0.0,
      "reward": 0.5,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.5,
      "step": 360
    },
    {
      "completion_length": 386.6,
      "epoch": 0.008222222222222223,
      "grad_norm": 0.18963927030563354,
      "kl": 0.0005982606788165867,
      "learning_rate": 3.798797596089351e-08,
      "loss": 0.0,
      "reward": 0.45,
      "reward_std": 0.49497473835945127,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.45,
      "step": 370
    },
    {
      "completion_length": 378.0,
      "epoch": 0.008444444444444444,
      "grad_norm": 2.9212462322902866e-05,
      "kl": 0.0006032192788552493,
      "learning_rate": 2.9263101785268252e-08,
      "loss": 0.0,
      "reward": 0.2,
      "reward_std": 0.1414213538169861,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.2,
      "step": 380
    },
    {
      "completion_length": 375.95,
      "epoch": 0.008666666666666666,
      "grad_norm": 1.570535641803872e-05,
      "kl": 0.0006192138127516956,
      "learning_rate": 2.1613635589349756e-08,
      "loss": 0.0,
      "reward": 0.55,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.55,
      "step": 390
    },
    {
      "completion_length": 399.75,
      "epoch": 0.008888888888888889,
      "grad_norm": 0.1566227674484253,
      "kl": 0.0005766067682998255,
      "learning_rate": 1.507684480352292e-08,
      "loss": 0.0,
      "reward": 0.4,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4,
      "step": 400
    },
    {
      "completion_length": 317.9,
      "epoch": 0.009111111111111111,
      "grad_norm": 0.42514702677726746,
      "kl": 0.000586391575052403,
      "learning_rate": 9.684576015420275e-09,
      "loss": 0.0,
      "reward": 0.5,
      "reward_std": 0.42426406145095824,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.5,
      "step": 410
    },
    {
      "completion_length": 440.05,
      "epoch": 0.009333333333333334,
      "grad_norm": 1.8029251805273816e-05,
      "kl": 0.0005952615960268304,
      "learning_rate": 5.463099816548577e-09,
      "loss": 0.0,
      "reward": 0.35,
      "reward_std": 0.49497473835945127,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.35,
      "step": 420
    },
    {
      "completion_length": 391.65,
      "epoch": 0.009555555555555555,
      "grad_norm": 0.23369018733501434,
      "kl": 0.0005777814978500829,
      "learning_rate": 2.4329828146074096e-09,
      "loss": 0.0,
      "reward": 0.45,
      "reward_std": 0.3535533845424652,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.45,
      "step": 430
    },
    {
      "completion_length": 368.75,
      "epoch": 0.009777777777777778,
      "grad_norm": 0.27213847637176514,
      "kl": 0.0006278897577431053,
      "learning_rate": 6.089874350439505e-10,
      "loss": 0.0,
      "reward": 0.45,
      "reward_std": 0.49497473835945127,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.45,
      "step": 440
    },
    {
      "completion_length": 445.5,
      "epoch": 0.01,
      "grad_norm": 2.1452080545714125e-05,
      "kl": 0.000612707485561259,
      "learning_rate": 0.0,
      "loss": 0.0,
      "reward": 0.2,
      "reward_std": 0.2828427076339722,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.2,
      "step": 450
    },
    {
      "completion_length": 379.875,
      "epoch": 0.010022222222222222,
      "kl": 0.0005991907673887908,
      "reward": 0.375,
      "reward_std": 0.5175491571426392,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.375,
      "step": 451,
      "total_flos": 0.0,
      "train_loss": 1.3050906145387637e-09,
      "train_runtime": 4363.1554,
      "train_samples_per_second": 0.103,
      "train_steps_per_second": 0.103
    }
  ],
  "logging_steps": 10,
  "max_steps": 450,
  "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": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}