{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 554.109375,
      "epoch": 0.0008,
      "grad_norm": 1.106962696508615,
      "kl": 0.0,
      "learning_rate": 4.999999122701883e-06,
      "loss": 0.0,
      "reward": 1.375,
      "reward_std": 0.4996672570705414,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.875,
      "step": 1
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 423.8125,
      "epoch": 0.0016,
      "grad_norm": 1.314941487556923,
      "kl": 0.0002727508544921875,
      "learning_rate": 4.999996490808146e-06,
      "loss": 0.0,
      "reward": 1.53125,
      "reward_std": 0.5366887450218201,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.859375,
      "step": 2
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 472.328125,
      "epoch": 0.0024,
      "grad_norm": 1.2349312957323104,
      "kl": 0.00128936767578125,
      "learning_rate": 4.9999921043206356e-06,
      "loss": 0.0001,
      "reward": 1.46875,
      "reward_std": 0.4995177984237671,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.890625,
      "step": 3
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 504.171875,
      "epoch": 0.0032,
      "grad_norm": 0.6879182927151376,
      "kl": 0.00153350830078125,
      "learning_rate": 4.999985963242432e-06,
      "loss": 0.0001,
      "reward": 1.75,
      "reward_std": 0.20266614854335785,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 4
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 415.375,
      "epoch": 0.004,
      "grad_norm": 1.0634938924971327,
      "kl": 0.006317138671875,
      "learning_rate": 4.999978067577844e-06,
      "loss": 0.0003,
      "reward": 1.59375,
      "reward_std": 0.26409149169921875,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 5
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 421.15625,
      "epoch": 0.0048,
      "grad_norm": 1.0266108197177204,
      "kl": 0.009521484375,
      "learning_rate": 4.999968417332415e-06,
      "loss": 0.0004,
      "reward": 1.625,
      "reward_std": 0.31300368905067444,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 6
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 428.140625,
      "epoch": 0.0056,
      "grad_norm": 4.588923240071208,
      "kl": 0.0113525390625,
      "learning_rate": 4.999957012512916e-06,
      "loss": 0.0005,
      "reward": 1.71875,
      "reward_std": 0.2845909595489502,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 7
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 400.9375,
      "epoch": 0.0064,
      "grad_norm": 1.2374348718708577,
      "kl": 0.014892578125,
      "learning_rate": 4.999943853127351e-06,
      "loss": 0.0006,
      "reward": 1.3125,
      "reward_std": 0.34352827072143555,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 8
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 385.46875,
      "epoch": 0.0072,
      "grad_norm": 1.5154801241797453,
      "kl": 0.020263671875,
      "learning_rate": 4.999928939184958e-06,
      "loss": 0.0008,
      "reward": 1.703125,
      "reward_std": 0.3492845892906189,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 9
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 467.640625,
      "epoch": 0.008,
      "grad_norm": 2.8984064250207613,
      "kl": 0.0242919921875,
      "learning_rate": 4.999912270696202e-06,
      "loss": 0.001,
      "reward": 1.546875,
      "reward_std": 0.2993341088294983,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 10
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 457.203125,
      "epoch": 0.0088,
      "grad_norm": 0.997494152734877,
      "kl": 0.027099609375,
      "learning_rate": 4.999893847672783e-06,
      "loss": 0.0011,
      "reward": 1.765625,
      "reward_std": 0.17782479524612427,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 11
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 459.328125,
      "epoch": 0.0096,
      "grad_norm": 0.8699166450034578,
      "kl": 0.0281982421875,
      "learning_rate": 4.99987367012763e-06,
      "loss": 0.0011,
      "reward": 1.65625,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 12
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 425.515625,
      "epoch": 0.0104,
      "grad_norm": 1.312698419751893,
      "kl": 0.034423828125,
      "learning_rate": 4.999851738074904e-06,
      "loss": 0.0014,
      "reward": 1.5,
      "reward_std": 0.30509042739868164,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.984375,
      "step": 13
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 421.1875,
      "epoch": 0.0112,
      "grad_norm": 1.0983638696965168,
      "kl": 0.038330078125,
      "learning_rate": 4.9998280515300006e-06,
      "loss": 0.0015,
      "reward": 1.625,
      "reward_std": 0.26409146189689636,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 14
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 388.53125,
      "epoch": 0.012,
      "grad_norm": 1.8508794118515073,
      "kl": 0.046630859375,
      "learning_rate": 4.999802610509541e-06,
      "loss": 0.0019,
      "reward": 1.6875,
      "reward_std": 0.2845909595489502,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 15
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 457.9375,
      "epoch": 0.0128,
      "grad_norm": 0.8599607884439239,
      "kl": 0.046875,
      "learning_rate": 4.999775415031381e-06,
      "loss": 0.0019,
      "reward": 1.921875,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 16
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 420.953125,
      "epoch": 0.0136,
      "grad_norm": 1.2918338302283918,
      "kl": 0.055419921875,
      "learning_rate": 4.999746465114609e-06,
      "loss": 0.0022,
      "reward": 1.59375,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 17
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 442.765625,
      "epoch": 0.0144,
      "grad_norm": 0.9977107372357084,
      "kl": 0.0546875,
      "learning_rate": 4.999715760779541e-06,
      "loss": 0.0022,
      "reward": 1.578125,
      "reward_std": 0.2688094973564148,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 18
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 438.546875,
      "epoch": 0.0152,
      "grad_norm": 0.8113982563207017,
      "kl": 0.05908203125,
      "learning_rate": 4.999683302047729e-06,
      "loss": 0.0024,
      "reward": 1.46875,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 19
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 446.578125,
      "epoch": 0.016,
      "grad_norm": 1.2752550143906236,
      "kl": 0.05419921875,
      "learning_rate": 4.999649088941951e-06,
      "loss": 0.0022,
      "reward": 1.5625,
      "reward_std": 0.22201895713806152,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 20
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 433.015625,
      "epoch": 0.0168,
      "grad_norm": 1.6448308314454292,
      "kl": 0.0625,
      "learning_rate": 4.999613121486222e-06,
      "loss": 0.0025,
      "reward": 1.65625,
      "reward_std": 0.24359199404716492,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 21
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 486.875,
      "epoch": 0.0176,
      "grad_norm": 1.5079682066474072,
      "kl": 0.060546875,
      "learning_rate": 4.999575399705782e-06,
      "loss": 0.0024,
      "reward": 1.609375,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 22
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 421.21875,
      "epoch": 0.0184,
      "grad_norm": 1.1243397903925967,
      "kl": 0.06201171875,
      "learning_rate": 4.9995359236271094e-06,
      "loss": 0.0025,
      "reward": 1.53125,
      "reward_std": 0.3029785752296448,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 23
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 484.265625,
      "epoch": 0.0192,
      "grad_norm": 0.9576568461832615,
      "kl": 0.0654296875,
      "learning_rate": 4.9994946932779076e-06,
      "loss": 0.0026,
      "reward": 1.53125,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 24
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 488.375,
      "epoch": 0.02,
      "grad_norm": 0.9539507807360903,
      "kl": 0.05859375,
      "learning_rate": 4.999451708687114e-06,
      "loss": 0.0023,
      "reward": 1.6875,
      "reward_std": 0.26409146189689636,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 25
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 478.6875,
      "epoch": 0.0208,
      "grad_norm": 1.584300461540275,
      "kl": 0.064453125,
      "learning_rate": 4.999406969884897e-06,
      "loss": 0.0026,
      "reward": 1.265625,
      "reward_std": 0.2993341088294983,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 1.0,
      "step": 26
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 466.859375,
      "epoch": 0.0216,
      "grad_norm": 1.0007005937653206,
      "kl": 0.06884765625,
      "learning_rate": 4.999360476902656e-06,
      "loss": 0.0027,
      "reward": 1.5625,
      "reward_std": 0.2708277702331543,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 27
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 432.71875,
      "epoch": 0.0224,
      "grad_norm": 1.2573243271025918,
      "kl": 0.07421875,
      "learning_rate": 4.999312229773022e-06,
      "loss": 0.003,
      "reward": 1.578125,
      "reward_std": 0.3266732692718506,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 28
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 508.546875,
      "epoch": 0.0232,
      "grad_norm": 0.9794499459288435,
      "kl": 0.0751953125,
      "learning_rate": 4.999262228529855e-06,
      "loss": 0.003,
      "reward": 1.703125,
      "reward_std": 0.40822190046310425,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 29
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 455.328125,
      "epoch": 0.024,
      "grad_norm": 1.169183523278842,
      "kl": 0.08056640625,
      "learning_rate": 4.99921047320825e-06,
      "loss": 0.0032,
      "reward": 1.453125,
      "reward_std": 0.5022729635238647,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.96875,
      "step": 30
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 494.53125,
      "epoch": 0.0248,
      "grad_norm": 1.785017154453716,
      "kl": 0.0771484375,
      "learning_rate": 4.99915696384453e-06,
      "loss": 0.0031,
      "reward": 1.640625,
      "reward_std": 0.4545377492904663,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.953125,
      "step": 31
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 491.71875,
      "epoch": 0.0256,
      "grad_norm": 0.9898909426994473,
      "kl": 0.072265625,
      "learning_rate": 4.99910170047625e-06,
      "loss": 0.0029,
      "reward": 1.75,
      "reward_std": 0.26409149169921875,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 32
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 472.171875,
      "epoch": 0.0264,
      "grad_norm": 0.8838784425314528,
      "kl": 0.08203125,
      "learning_rate": 4.999044683142196e-06,
      "loss": 0.0033,
      "reward": 1.734375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 33
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 472.25,
      "epoch": 0.0272,
      "grad_norm": 1.2255748101481239,
      "kl": 0.0849609375,
      "learning_rate": 4.998985911882383e-06,
      "loss": 0.0034,
      "reward": 1.453125,
      "reward_std": 0.3403330445289612,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 34
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 457.578125,
      "epoch": 0.028,
      "grad_norm": 0.9181810045745323,
      "kl": 0.087890625,
      "learning_rate": 4.998925386738063e-06,
      "loss": 0.0035,
      "reward": 1.765625,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 35
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 493.90625,
      "epoch": 0.0288,
      "grad_norm": 0.9492186816581748,
      "kl": 0.08447265625,
      "learning_rate": 4.998863107751711e-06,
      "loss": 0.0034,
      "reward": 1.359375,
      "reward_std": 0.25726157426834106,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.984375,
      "step": 36
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 433.828125,
      "epoch": 0.0296,
      "grad_norm": 0.7312117876501095,
      "kl": 0.0869140625,
      "learning_rate": 4.99879907496704e-06,
      "loss": 0.0035,
      "reward": 1.734375,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 37
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 511.15625,
      "epoch": 0.0304,
      "grad_norm": 3.226160403793959,
      "kl": 0.09423828125,
      "learning_rate": 4.998733288428987e-06,
      "loss": 0.0038,
      "reward": 1.5,
      "reward_std": 0.5134496092796326,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.953125,
      "step": 38
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.375,
      "completion_length": 427.625,
      "epoch": 0.0312,
      "grad_norm": 1.4155213994713876,
      "kl": 0.0927734375,
      "learning_rate": 4.998665748183727e-06,
      "loss": 0.0037,
      "reward": 1.28125,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 39
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 440.984375,
      "epoch": 0.032,
      "grad_norm": 1.2465987661295874,
      "kl": 0.10302734375,
      "learning_rate": 4.998596454278661e-06,
      "loss": 0.0041,
      "reward": 1.421875,
      "reward_std": 0.30617380142211914,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.984375,
      "step": 40
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 519.984375,
      "epoch": 0.0328,
      "grad_norm": 0.9485069789727323,
      "kl": 0.08837890625,
      "learning_rate": 4.998525406762422e-06,
      "loss": 0.0035,
      "reward": 1.84375,
      "reward_std": 0.26409146189689636,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 41
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 473.8125,
      "epoch": 0.0336,
      "grad_norm": 1.1552791242133729,
      "kl": 0.10107421875,
      "learning_rate": 4.998452605684874e-06,
      "loss": 0.004,
      "reward": 1.765625,
      "reward_std": 0.3492845892906189,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 42
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 432.265625,
      "epoch": 0.0344,
      "grad_norm": 0.8775955518065937,
      "kl": 0.1005859375,
      "learning_rate": 4.998378051097111e-06,
      "loss": 0.004,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 43
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 509.265625,
      "epoch": 0.0352,
      "grad_norm": 1.9400186881457824,
      "kl": 0.099609375,
      "learning_rate": 4.998301743051459e-06,
      "loss": 0.004,
      "reward": 1.71875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 44
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 459.78125,
      "epoch": 0.036,
      "grad_norm": 1.2807091748853519,
      "kl": 0.099609375,
      "learning_rate": 4.9982236816014735e-06,
      "loss": 0.004,
      "reward": 1.609375,
      "reward_std": 0.38380008935928345,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 45
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 538.15625,
      "epoch": 0.0368,
      "grad_norm": 3.0208592385584594,
      "kl": 0.0947265625,
      "learning_rate": 4.998143866801941e-06,
      "loss": 0.0038,
      "reward": 1.359375,
      "reward_std": 0.31983357667922974,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.984375,
      "step": 46
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 498.3125,
      "epoch": 0.0376,
      "grad_norm": 0.9409790694547606,
      "kl": 0.10693359375,
      "learning_rate": 4.99806229870888e-06,
      "loss": 0.0043,
      "reward": 1.5,
      "reward_std": 0.26409149169921875,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.984375,
      "step": 47
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 424.578125,
      "epoch": 0.0384,
      "grad_norm": 1.361137450227213,
      "kl": 0.10595703125,
      "learning_rate": 4.9979789773795365e-06,
      "loss": 0.0042,
      "reward": 1.59375,
      "reward_std": 0.19149437546730042,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 48
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 471.046875,
      "epoch": 0.0392,
      "grad_norm": 1.3917125320283434,
      "kl": 0.1083984375,
      "learning_rate": 4.997893902872389e-06,
      "loss": 0.0043,
      "reward": 1.59375,
      "reward_std": 0.32195523381233215,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 49
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 489.34375,
      "epoch": 0.04,
      "grad_norm": 0.6143978746860285,
      "kl": 0.09765625,
      "learning_rate": 4.997807075247147e-06,
      "loss": 0.0039,
      "reward": 1.921875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 50
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 496.46875,
      "epoch": 0.0408,
      "grad_norm": 1.6577824499312852,
      "kl": 0.10205078125,
      "learning_rate": 4.997718494564747e-06,
      "loss": 0.0041,
      "reward": 1.546875,
      "reward_std": 0.3266732692718506,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 51
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 509.3125,
      "epoch": 0.0416,
      "grad_norm": 1.398115575311656,
      "kl": 0.09375,
      "learning_rate": 4.997628160887361e-06,
      "loss": 0.0038,
      "reward": 1.65625,
      "reward_std": 0.31300365924835205,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 52
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 475.9375,
      "epoch": 0.0424,
      "grad_norm": 1.289505457669401,
      "kl": 0.099609375,
      "learning_rate": 4.997536074278388e-06,
      "loss": 0.004,
      "reward": 1.78125,
      "reward_std": 0.3377464711666107,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 53
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 477.140625,
      "epoch": 0.0432,
      "grad_norm": 1.0274251674462114,
      "kl": 0.0927734375,
      "learning_rate": 4.9974422348024565e-06,
      "loss": 0.0037,
      "reward": 1.59375,
      "reward_std": 0.30509042739868164,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 54
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 457.828125,
      "epoch": 0.044,
      "grad_norm": 1.2328865824867006,
      "kl": 0.09521484375,
      "learning_rate": 4.997346642525429e-06,
      "loss": 0.0038,
      "reward": 1.6875,
      "reward_std": 0.33090677857398987,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 55
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 393.84375,
      "epoch": 0.0448,
      "grad_norm": 2.649900512145904,
      "kl": 0.1171875,
      "learning_rate": 4.9972492975143936e-06,
      "loss": 0.0047,
      "reward": 1.609375,
      "reward_std": 0.34717273712158203,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 56
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 435.984375,
      "epoch": 0.0456,
      "grad_norm": 1.5429448769370444,
      "kl": 0.10107421875,
      "learning_rate": 4.997150199837671e-06,
      "loss": 0.004,
      "reward": 1.640625,
      "reward_std": 0.2688094973564148,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 57
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 476.75,
      "epoch": 0.0464,
      "grad_norm": 0.7750728694989101,
      "kl": 0.10009765625,
      "learning_rate": 4.997049349564814e-06,
      "loss": 0.004,
      "reward": 1.453125,
      "reward_std": 0.1893727034330368,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.96875,
      "step": 58
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 416.84375,
      "epoch": 0.0472,
      "grad_norm": 1.1808569185472115,
      "kl": 0.10302734375,
      "learning_rate": 4.996946746766602e-06,
      "loss": 0.0041,
      "reward": 1.6875,
      "reward_std": 0.3061639666557312,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 59
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 436.359375,
      "epoch": 0.048,
      "grad_norm": 0.87560419549391,
      "kl": 0.10107421875,
      "learning_rate": 4.996842391515045e-06,
      "loss": 0.004,
      "reward": 1.703125,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 60
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 444.109375,
      "epoch": 0.0488,
      "grad_norm": 0.9983118033206102,
      "kl": 0.10595703125,
      "learning_rate": 4.996736283883382e-06,
      "loss": 0.0042,
      "reward": 1.46875,
      "reward_std": 0.28247910737991333,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 61
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 434.140625,
      "epoch": 0.0496,
      "grad_norm": 1.2522486907935333,
      "kl": 0.1044921875,
      "learning_rate": 4.9966284239460875e-06,
      "loss": 0.0042,
      "reward": 1.6875,
      "reward_std": 0.3697938621044159,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 62
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 456.828125,
      "epoch": 0.0504,
      "grad_norm": 1.9128830955616545,
      "kl": 0.0966796875,
      "learning_rate": 4.996518811778858e-06,
      "loss": 0.0039,
      "reward": 1.59375,
      "reward_std": 0.3104073107242584,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 63
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 442.734375,
      "epoch": 0.0512,
      "grad_norm": 1.5421042944955659,
      "kl": 0.10205078125,
      "learning_rate": 4.996407447458626e-06,
      "loss": 0.0041,
      "reward": 1.734375,
      "reward_std": 0.3845370411872864,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 64
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 413.796875,
      "epoch": 0.052,
      "grad_norm": 1.0558375983386774,
      "kl": 0.10302734375,
      "learning_rate": 4.99629433106355e-06,
      "loss": 0.0041,
      "reward": 1.59375,
      "reward_std": 0.2404065728187561,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 65
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 518.75,
      "epoch": 0.0528,
      "grad_norm": 0.6372842511275832,
      "kl": 0.08837890625,
      "learning_rate": 4.99617946267302e-06,
      "loss": 0.0035,
      "reward": 1.640625,
      "reward_std": 0.22673700749874115,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.96875,
      "step": 66
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 444.46875,
      "epoch": 0.0536,
      "grad_norm": 0.9039550027984934,
      "kl": 0.11181640625,
      "learning_rate": 4.996062842367655e-06,
      "loss": 0.0045,
      "reward": 1.734375,
      "reward_std": 0.2468603253364563,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 67
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 411.734375,
      "epoch": 0.0544,
      "grad_norm": 0.823442167835391,
      "kl": 0.10986328125,
      "learning_rate": 4.9959444702293025e-06,
      "loss": 0.0044,
      "reward": 1.625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 68
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 440.40625,
      "epoch": 0.0552,
      "grad_norm": 1.7659596418323726,
      "kl": 0.111328125,
      "learning_rate": 4.995824346341041e-06,
      "loss": 0.0045,
      "reward": 1.453125,
      "reward_std": 0.2993341088294983,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 69
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 419.703125,
      "epoch": 0.056,
      "grad_norm": 1.207292808967077,
      "kl": 0.107421875,
      "learning_rate": 4.99570247078718e-06,
      "loss": 0.0043,
      "reward": 1.703125,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 70
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 402.140625,
      "epoch": 0.0568,
      "grad_norm": 1.3768610203470437,
      "kl": 0.11328125,
      "learning_rate": 4.995578843653255e-06,
      "loss": 0.0045,
      "reward": 1.421875,
      "reward_std": 0.3403330445289612,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 71
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 432.078125,
      "epoch": 0.0576,
      "grad_norm": 1.7876726764336355,
      "kl": 0.1083984375,
      "learning_rate": 4.995453465026033e-06,
      "loss": 0.0043,
      "reward": 1.5,
      "reward_std": 0.2845909595489502,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.984375,
      "step": 72
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 440.9375,
      "epoch": 0.0584,
      "grad_norm": 1.20618885021594,
      "kl": 0.1044921875,
      "learning_rate": 4.995326334993508e-06,
      "loss": 0.0042,
      "reward": 1.640625,
      "reward_std": 0.24039676785469055,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 73
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 481.453125,
      "epoch": 0.0592,
      "grad_norm": 1.0694999127985507,
      "kl": 0.10888671875,
      "learning_rate": 4.9951974536449055e-06,
      "loss": 0.0044,
      "reward": 1.453125,
      "reward_std": 0.34259048104286194,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.9375,
      "step": 74
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 563.21875,
      "epoch": 0.06,
      "grad_norm": 0.6675463248539277,
      "kl": 0.091796875,
      "learning_rate": 4.9950668210706795e-06,
      "loss": 0.0037,
      "reward": 1.828125,
      "reward_std": 0.22340349853038788,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 75
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 418.25,
      "epoch": 0.0608,
      "grad_norm": 0.7716067192758925,
      "kl": 0.10791015625,
      "learning_rate": 4.994934437362513e-06,
      "loss": 0.0043,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 76
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 465.265625,
      "epoch": 0.0616,
      "grad_norm": 0.9137739572414142,
      "kl": 0.107421875,
      "learning_rate": 4.994800302613318e-06,
      "loss": 0.0043,
      "reward": 1.6875,
      "reward_std": 0.24251842498779297,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 77
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 468.03125,
      "epoch": 0.0624,
      "grad_norm": 1.4439553758293866,
      "kl": 0.10302734375,
      "learning_rate": 4.994664416917236e-06,
      "loss": 0.0041,
      "reward": 1.703125,
      "reward_std": 0.2688094973564148,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 78
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 495.625,
      "epoch": 0.0632,
      "grad_norm": 1.5610476959939437,
      "kl": 0.099609375,
      "learning_rate": 4.994526780369636e-06,
      "loss": 0.004,
      "reward": 1.5625,
      "reward_std": 0.4433611333370209,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 79
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 461.71875,
      "epoch": 0.064,
      "grad_norm": 1.4117659211308278,
      "kl": 0.10400390625,
      "learning_rate": 4.9943873930671175e-06,
      "loss": 0.0042,
      "reward": 1.78125,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 80
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 468.828125,
      "epoch": 0.0648,
      "grad_norm": 1.1585357618323595,
      "kl": 0.1103515625,
      "learning_rate": 4.994246255107506e-06,
      "loss": 0.0044,
      "reward": 1.609375,
      "reward_std": 0.2472364604473114,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 81
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 473.859375,
      "epoch": 0.0656,
      "grad_norm": 0.650633536981878,
      "kl": 0.1142578125,
      "learning_rate": 4.994103366589859e-06,
      "loss": 0.0046,
      "reward": 1.671875,
      "reward_std": 0.2289944440126419,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.96875,
      "step": 82
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 542.234375,
      "epoch": 0.0664,
      "grad_norm": 0.9189496334102202,
      "kl": 0.1005859375,
      "learning_rate": 4.993958727614462e-06,
      "loss": 0.004,
      "reward": 1.65625,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 83
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 432.5,
      "epoch": 0.0672,
      "grad_norm": 0.9665160891937301,
      "kl": 0.11328125,
      "learning_rate": 4.993812338282826e-06,
      "loss": 0.0045,
      "reward": 1.3125,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 84
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 491.40625,
      "epoch": 0.068,
      "grad_norm": 1.3058592035065932,
      "kl": 0.126953125,
      "learning_rate": 4.993664198697694e-06,
      "loss": 0.0051,
      "reward": 1.65625,
      "reward_std": 0.2893187999725342,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 85
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 549.328125,
      "epoch": 0.0688,
      "grad_norm": 0.9355477411550364,
      "kl": 0.1064453125,
      "learning_rate": 4.993514308963037e-06,
      "loss": 0.0043,
      "reward": 1.5625,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 86
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 468.53125,
      "epoch": 0.0696,
      "grad_norm": 2.916067462057221,
      "kl": 0.11572265625,
      "learning_rate": 4.993362669184051e-06,
      "loss": 0.0046,
      "reward": 1.53125,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 87
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 502.3125,
      "epoch": 0.0704,
      "grad_norm": 1.4264894854376802,
      "kl": 0.109375,
      "learning_rate": 4.993209279467164e-06,
      "loss": 0.0044,
      "reward": 1.78125,
      "reward_std": 0.3682710528373718,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 88
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 489.5625,
      "epoch": 0.0712,
      "grad_norm": 1.014116772780785,
      "kl": 0.107421875,
      "learning_rate": 4.993054139920031e-06,
      "loss": 0.0043,
      "reward": 1.625,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 89
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 502.875,
      "epoch": 0.072,
      "grad_norm": 1.0276948195058366,
      "kl": 0.11767578125,
      "learning_rate": 4.992897250651535e-06,
      "loss": 0.0047,
      "reward": 1.578125,
      "reward_std": 0.29826050996780396,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 90
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 435.234375,
      "epoch": 0.0728,
      "grad_norm": 2.4748299765258666,
      "kl": 0.1201171875,
      "learning_rate": 4.992738611771787e-06,
      "loss": 0.0048,
      "reward": 1.375,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 91
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 440.265625,
      "epoch": 0.0736,
      "grad_norm": 1.1630505917313283,
      "kl": 0.1142578125,
      "learning_rate": 4.992578223392124e-06,
      "loss": 0.0046,
      "reward": 1.65625,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 92
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 431.890625,
      "epoch": 0.0744,
      "grad_norm": 0.671672316043001,
      "kl": 0.10546875,
      "learning_rate": 4.992416085625115e-06,
      "loss": 0.0042,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 93
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 429.9375,
      "epoch": 0.0752,
      "grad_norm": 0.9166302480771739,
      "kl": 0.12060546875,
      "learning_rate": 4.992252198584554e-06,
      "loss": 0.0048,
      "reward": 1.75,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 94
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 427.390625,
      "epoch": 0.076,
      "grad_norm": 0.8938241675508535,
      "kl": 0.09521484375,
      "learning_rate": 4.992086562385462e-06,
      "loss": 0.0038,
      "reward": 1.609375,
      "reward_std": 0.24831004440784454,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 95
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 431.40625,
      "epoch": 0.0768,
      "grad_norm": 0.9472168820065903,
      "kl": 0.1064453125,
      "learning_rate": 4.9919191771440905e-06,
      "loss": 0.0043,
      "reward": 1.890625,
      "reward_std": 0.16887325048446655,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 96
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 486.171875,
      "epoch": 0.0776,
      "grad_norm": 1.53466631135331,
      "kl": 0.1123046875,
      "learning_rate": 4.9917500429779165e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.23827511072158813,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 97
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 425.484375,
      "epoch": 0.0784,
      "grad_norm": 0.9843285642666483,
      "kl": 0.107421875,
      "learning_rate": 4.991579160005644e-06,
      "loss": 0.0043,
      "reward": 1.734375,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 98
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 426.4375,
      "epoch": 0.0792,
      "grad_norm": 0.6754793866432567,
      "kl": 0.10546875,
      "learning_rate": 4.991406528347206e-06,
      "loss": 0.0042,
      "reward": 1.875,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 99
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 461.84375,
      "epoch": 0.08,
      "grad_norm": 0.7474505512475345,
      "kl": 0.11669921875,
      "learning_rate": 4.9912321481237616e-06,
      "loss": 0.0047,
      "reward": 1.5,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 100
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 427.0,
      "epoch": 0.0808,
      "grad_norm": 1.0264334044052603,
      "kl": 0.11279296875,
      "learning_rate": 4.991056019457697e-06,
      "loss": 0.0045,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 101
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 440.75,
      "epoch": 0.0816,
      "grad_norm": 1.30829061939001,
      "kl": 0.12255859375,
      "learning_rate": 4.990878142472628e-06,
      "loss": 0.0049,
      "reward": 1.53125,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 102
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 411.921875,
      "epoch": 0.0824,
      "grad_norm": 0.5577946664774175,
      "kl": 0.1181640625,
      "learning_rate": 4.990698517293394e-06,
      "loss": 0.0047,
      "reward": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 103
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 425.9375,
      "epoch": 0.0832,
      "grad_norm": 0.8685769518063424,
      "kl": 0.1083984375,
      "learning_rate": 4.9905171440460645e-06,
      "loss": 0.0043,
      "reward": 1.78125,
      "reward_std": 0.1962026059627533,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 104
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 466.546875,
      "epoch": 0.084,
      "grad_norm": 0.5092956743369942,
      "kl": 0.11181640625,
      "learning_rate": 4.990334022857932e-06,
      "loss": 0.0045,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 105
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 406.65625,
      "epoch": 0.0848,
      "grad_norm": 1.2897529204623364,
      "kl": 0.11474609375,
      "learning_rate": 4.990149153857519e-06,
      "loss": 0.0046,
      "reward": 1.421875,
      "reward_std": 0.3934885859489441,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 106
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 406.921875,
      "epoch": 0.0856,
      "grad_norm": 1.1754549401230758,
      "kl": 0.11572265625,
      "learning_rate": 4.989962537174573e-06,
      "loss": 0.0046,
      "reward": 1.75,
      "reward_std": 0.2961388826370239,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 107
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 440.75,
      "epoch": 0.0864,
      "grad_norm": 0.8386331634897316,
      "kl": 0.11572265625,
      "learning_rate": 4.989774172940071e-06,
      "loss": 0.0046,
      "reward": 1.625,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 108
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 448.296875,
      "epoch": 0.0872,
      "grad_norm": 0.7450142523211052,
      "kl": 0.10498046875,
      "learning_rate": 4.989584061286211e-06,
      "loss": 0.0042,
      "reward": 1.609375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 109
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 427.640625,
      "epoch": 0.088,
      "grad_norm": 0.08383802239107432,
      "kl": 0.0986328125,
      "learning_rate": 4.989392202346423e-06,
      "loss": 0.004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 110
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 428.46875,
      "epoch": 0.0888,
      "grad_norm": 1.2199438535090528,
      "kl": 0.12109375,
      "learning_rate": 4.989198596255361e-06,
      "loss": 0.0048,
      "reward": 1.75,
      "reward_std": 0.38452720642089844,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 111
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 571.71875,
      "epoch": 0.0896,
      "grad_norm": 0.9646884627154972,
      "kl": 0.103515625,
      "learning_rate": 4.989003243148904e-06,
      "loss": 0.0041,
      "reward": 1.5,
      "reward_std": 0.33090677857398987,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 112
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 431.6875,
      "epoch": 0.0904,
      "grad_norm": 0.8033337673718765,
      "kl": 0.103515625,
      "learning_rate": 4.988806143164159e-06,
      "loss": 0.0041,
      "reward": 1.796875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 113
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 447.421875,
      "epoch": 0.0912,
      "grad_norm": 0.46704702690421707,
      "kl": 0.1083984375,
      "learning_rate": 4.988607296439459e-06,
      "loss": 0.0043,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 114
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 487.40625,
      "epoch": 0.092,
      "grad_norm": 0.9691362664902082,
      "kl": 0.103515625,
      "learning_rate": 4.98840670311436e-06,
      "loss": 0.0041,
      "reward": 1.671875,
      "reward_std": 0.2688094973564148,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 115
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 436.5625,
      "epoch": 0.0928,
      "grad_norm": 0.7765141393387954,
      "kl": 0.1005859375,
      "learning_rate": 4.988204363329648e-06,
      "loss": 0.004,
      "reward": 1.84375,
      "reward_std": 0.22201895713806152,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 116
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 495.34375,
      "epoch": 0.0936,
      "grad_norm": 0.7147590654651254,
      "kl": 0.099609375,
      "learning_rate": 4.988000277227334e-06,
      "loss": 0.004,
      "reward": 1.65625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 117
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 495.984375,
      "epoch": 0.0944,
      "grad_norm": 1.677681916051708,
      "kl": 0.1103515625,
      "learning_rate": 4.987794444950651e-06,
      "loss": 0.0044,
      "reward": 1.546875,
      "reward_std": 0.3047240972518921,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 118
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 415.78125,
      "epoch": 0.0952,
      "grad_norm": 0.7635954540557363,
      "kl": 0.134765625,
      "learning_rate": 4.987586866644061e-06,
      "loss": 0.0054,
      "reward": 1.515625,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 119
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 464.953125,
      "epoch": 0.096,
      "grad_norm": 0.7514932681425285,
      "kl": 0.1103515625,
      "learning_rate": 4.9873775424532515e-06,
      "loss": 0.0044,
      "reward": 1.59375,
      "reward_std": 0.17570313811302185,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 120
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 424.84375,
      "epoch": 0.0968,
      "grad_norm": 1.0456069680305644,
      "kl": 0.12353515625,
      "learning_rate": 4.9871664725251314e-06,
      "loss": 0.0049,
      "reward": 1.46875,
      "reward_std": 0.2709311842918396,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 121
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 441.234375,
      "epoch": 0.0976,
      "grad_norm": 1.3131456197984361,
      "kl": 0.09912109375,
      "learning_rate": 4.986953657007841e-06,
      "loss": 0.004,
      "reward": 1.828125,
      "reward_std": 0.19044627249240875,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 122
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 468.234375,
      "epoch": 0.0984,
      "grad_norm": 4.478005441665192,
      "kl": 0.10791015625,
      "learning_rate": 4.98673909605074e-06,
      "loss": 0.0043,
      "reward": 1.734375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 123
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 433.21875,
      "epoch": 0.0992,
      "grad_norm": 1.006704967535853,
      "kl": 0.103515625,
      "learning_rate": 4.986522789804417e-06,
      "loss": 0.0041,
      "reward": 1.71875,
      "reward_std": 0.24251843988895416,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 124
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 411.03125,
      "epoch": 0.1,
      "grad_norm": 1.5202176841365798,
      "kl": 0.130859375,
      "learning_rate": 4.986304738420684e-06,
      "loss": 0.0052,
      "reward": 1.328125,
      "reward_std": 0.42256343364715576,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 0.984375,
      "step": 125
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 451.96875,
      "epoch": 0.1008,
      "grad_norm": 0.691520829690945,
      "kl": 0.1064453125,
      "learning_rate": 4.986084942052577e-06,
      "loss": 0.0043,
      "reward": 1.8125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 126
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 493.453125,
      "epoch": 0.1016,
      "grad_norm": 0.6394818378327461,
      "kl": 0.0966796875,
      "learning_rate": 4.9858634008543574e-06,
      "loss": 0.0039,
      "reward": 1.859375,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 127
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 420.140625,
      "epoch": 0.1024,
      "grad_norm": 1.122230685100362,
      "kl": 0.10595703125,
      "learning_rate": 4.985640114981513e-06,
      "loss": 0.0042,
      "reward": 1.6875,
      "reward_std": 0.24359199404716492,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 128
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 397.125,
      "epoch": 0.1032,
      "grad_norm": 1.0414783708435302,
      "kl": 0.103515625,
      "learning_rate": 4.985415084590752e-06,
      "loss": 0.0041,
      "reward": 1.84375,
      "reward_std": 0.1735912710428238,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 129
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 466.3125,
      "epoch": 0.104,
      "grad_norm": 1.2035770384664692,
      "kl": 0.09228515625,
      "learning_rate": 4.985188309840012e-06,
      "loss": 0.0037,
      "reward": 1.84375,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 130
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 383.203125,
      "epoch": 0.1048,
      "grad_norm": 1.0365072299262117,
      "kl": 0.1240234375,
      "learning_rate": 4.984959790888451e-06,
      "loss": 0.005,
      "reward": 1.5625,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 131
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 443.65625,
      "epoch": 0.1056,
      "grad_norm": 14.647926893861055,
      "kl": 0.1357421875,
      "learning_rate": 4.984729527896451e-06,
      "loss": 0.0054,
      "reward": 1.6875,
      "reward_std": 0.1962026059627533,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 132
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 402.140625,
      "epoch": 0.1064,
      "grad_norm": 11.978453161980177,
      "kl": 0.263671875,
      "learning_rate": 4.984497521025622e-06,
      "loss": 0.0105,
      "reward": 1.734375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 133
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 499.890625,
      "epoch": 0.1072,
      "grad_norm": 1.2891133348471002,
      "kl": 0.498046875,
      "learning_rate": 4.984263770438793e-06,
      "loss": 0.0199,
      "reward": 1.640625,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 134
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 596.953125,
      "epoch": 0.108,
      "grad_norm": 5.990687084060201,
      "kl": 1.0234375,
      "learning_rate": 4.984028276300021e-06,
      "loss": 0.0407,
      "reward": 1.8125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 135
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 495.765625,
      "epoch": 0.1088,
      "grad_norm": 0.7532120814342459,
      "kl": 0.103515625,
      "learning_rate": 4.983791038774585e-06,
      "loss": 0.0041,
      "reward": 1.890625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 136
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 442.171875,
      "epoch": 0.1096,
      "grad_norm": 0.866752394198012,
      "kl": 0.11572265625,
      "learning_rate": 4.983552058028985e-06,
      "loss": 0.0046,
      "reward": 1.734375,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 137
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 476.703125,
      "epoch": 0.1104,
      "grad_norm": 0.5921378125196634,
      "kl": 0.1103515625,
      "learning_rate": 4.9833113342309495e-06,
      "loss": 0.0044,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 138
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 495.859375,
      "epoch": 0.1112,
      "grad_norm": 3.8570119454857057,
      "kl": 0.11328125,
      "learning_rate": 4.983068867549427e-06,
      "loss": 0.0045,
      "reward": 1.765625,
      "reward_std": 0.2824692726135254,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 139
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 432.078125,
      "epoch": 0.112,
      "grad_norm": 0.7479618451959271,
      "kl": 0.123046875,
      "learning_rate": 4.982824658154589e-06,
      "loss": 0.0049,
      "reward": 1.859375,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 140
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 477.375,
      "epoch": 0.1128,
      "grad_norm": 0.7128326137294528,
      "kl": 0.10009765625,
      "learning_rate": 4.9825787062178315e-06,
      "loss": 0.004,
      "reward": 1.75,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 141
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 445.53125,
      "epoch": 0.1136,
      "grad_norm": 1.6602439851000796,
      "kl": 0.115234375,
      "learning_rate": 4.982331011911774e-06,
      "loss": 0.0046,
      "reward": 1.6875,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 142
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 475.25,
      "epoch": 0.1144,
      "grad_norm": 0.946041800084751,
      "kl": 0.1064453125,
      "learning_rate": 4.982081575410256e-06,
      "loss": 0.0043,
      "reward": 1.6875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 143
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 473.34375,
      "epoch": 0.1152,
      "grad_norm": 0.997362591524624,
      "kl": 0.1064453125,
      "learning_rate": 4.9818303968883445e-06,
      "loss": 0.0043,
      "reward": 1.78125,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 144
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 473.625,
      "epoch": 0.116,
      "grad_norm": 0.9153829155861837,
      "kl": 0.1083984375,
      "learning_rate": 4.981577476522323e-06,
      "loss": 0.0043,
      "reward": 1.65625,
      "reward_std": 0.3208816647529602,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 145
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 520.578125,
      "epoch": 0.1168,
      "grad_norm": 0.6533677207164628,
      "kl": 0.10107421875,
      "learning_rate": 4.981322814489703e-06,
      "loss": 0.004,
      "reward": 1.765625,
      "reward_std": 0.23925508558750153,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 146
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 480.828125,
      "epoch": 0.1176,
      "grad_norm": 0.93405014645264,
      "kl": 0.11181640625,
      "learning_rate": 4.981066410969215e-06,
      "loss": 0.0045,
      "reward": 1.609375,
      "reward_std": 0.2688094973564148,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 147
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 508.25,
      "epoch": 0.1184,
      "grad_norm": 0.9990112151974295,
      "kl": 0.1015625,
      "learning_rate": 4.980808266140813e-06,
      "loss": 0.0041,
      "reward": 1.25,
      "reward_std": 0.31300365924835205,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 0.984375,
      "step": 148
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 480.359375,
      "epoch": 0.1192,
      "grad_norm": 0.9937737169099482,
      "kl": 0.15625,
      "learning_rate": 4.9805483801856744e-06,
      "loss": 0.0062,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 149
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 483.484375,
      "epoch": 0.12,
      "grad_norm": 0.9316226268565391,
      "kl": 0.10302734375,
      "learning_rate": 4.980286753286196e-06,
      "loss": 0.0041,
      "reward": 1.671875,
      "reward_std": 0.2414703369140625,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 150
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 490.984375,
      "epoch": 0.1208,
      "grad_norm": 0.9497020028783308,
      "kl": 0.130859375,
      "learning_rate": 4.980023385625996e-06,
      "loss": 0.0052,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 151
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 473.8125,
      "epoch": 0.1216,
      "grad_norm": 0.33921544153730604,
      "kl": 0.10888671875,
      "learning_rate": 4.979758277389919e-06,
      "loss": 0.0044,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 152
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 435.734375,
      "epoch": 0.1224,
      "grad_norm": 1.0254651580844363,
      "kl": 0.1171875,
      "learning_rate": 4.9794914287640264e-06,
      "loss": 0.0047,
      "reward": 1.421875,
      "reward_std": 0.19044627249240875,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 153
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 465.703125,
      "epoch": 0.1232,
      "grad_norm": 0.7503411726500947,
      "kl": 0.1435546875,
      "learning_rate": 4.979222839935602e-06,
      "loss": 0.0057,
      "reward": 1.640625,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 154
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 444.78125,
      "epoch": 0.124,
      "grad_norm": 0.6925732949036683,
      "kl": 0.10009765625,
      "learning_rate": 4.9789525110931545e-06,
      "loss": 0.004,
      "reward": 1.828125,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 155
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 438.28125,
      "epoch": 0.1248,
      "grad_norm": 0.6444485434393429,
      "kl": 0.1123046875,
      "learning_rate": 4.978680442426409e-06,
      "loss": 0.0045,
      "reward": 1.921875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 156
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 510.25,
      "epoch": 0.1256,
      "grad_norm": 0.7238164485839524,
      "kl": 0.09765625,
      "learning_rate": 4.978406634126315e-06,
      "loss": 0.0039,
      "reward": 1.828125,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 157
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 405.875,
      "epoch": 0.1264,
      "grad_norm": 0.7341274999066865,
      "kl": 0.1220703125,
      "learning_rate": 4.978131086385041e-06,
      "loss": 0.0049,
      "reward": 1.65625,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 158
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 441.984375,
      "epoch": 0.1272,
      "grad_norm": 0.632450681610908,
      "kl": 0.12109375,
      "learning_rate": 4.977853799395976e-06,
      "loss": 0.0048,
      "reward": 1.625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 159
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 400.453125,
      "epoch": 0.128,
      "grad_norm": 0.9113883106967559,
      "kl": 0.109375,
      "learning_rate": 4.977574773353732e-06,
      "loss": 0.0044,
      "reward": 1.8125,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 160
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 500.09375,
      "epoch": 0.1288,
      "grad_norm": 0.7766519579442204,
      "kl": 0.111328125,
      "learning_rate": 4.97729400845414e-06,
      "loss": 0.0044,
      "reward": 1.796875,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 161
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 503.734375,
      "epoch": 0.1296,
      "grad_norm": 0.8484901654440711,
      "kl": 0.10107421875,
      "learning_rate": 4.977011504894253e-06,
      "loss": 0.004,
      "reward": 1.71875,
      "reward_std": 0.23356689512729645,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 162
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 403.140625,
      "epoch": 0.1304,
      "grad_norm": 1.694602433143206,
      "kl": 0.11279296875,
      "learning_rate": 4.97672726287234e-06,
      "loss": 0.0045,
      "reward": 1.84375,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 163
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 431.734375,
      "epoch": 0.1312,
      "grad_norm": 1.9691320474006948,
      "kl": 0.1103515625,
      "learning_rate": 4.976441282587894e-06,
      "loss": 0.0044,
      "reward": 1.734375,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 164
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 436.3125,
      "epoch": 0.132,
      "grad_norm": 0.8571450813349311,
      "kl": 0.11083984375,
      "learning_rate": 4.9761535642416284e-06,
      "loss": 0.0044,
      "reward": 1.765625,
      "reward_std": 0.2472364604473114,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 165
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 422.296875,
      "epoch": 0.1328,
      "grad_norm": 0.8401543733494647,
      "kl": 0.10546875,
      "learning_rate": 4.9758641080354745e-06,
      "loss": 0.0042,
      "reward": 1.578125,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 166
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 454.875,
      "epoch": 0.1336,
      "grad_norm": 0.7715139058569292,
      "kl": 0.10546875,
      "learning_rate": 4.975572914172581e-06,
      "loss": 0.0042,
      "reward": 1.890625,
      "reward_std": 0.21778544783592224,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 167
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 360.328125,
      "epoch": 0.1344,
      "grad_norm": 1.033415950988869,
      "kl": 0.11474609375,
      "learning_rate": 4.975279982857324e-06,
      "loss": 0.0046,
      "reward": 1.71875,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 168
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 354.640625,
      "epoch": 0.1352,
      "grad_norm": 1.3592179501596968,
      "kl": 0.12060546875,
      "learning_rate": 4.97498531429529e-06,
      "loss": 0.0048,
      "reward": 1.46875,
      "reward_std": 0.25513994693756104,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 169
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 402.28125,
      "epoch": 0.136,
      "grad_norm": 1.0245659937809795,
      "kl": 0.1162109375,
      "learning_rate": 4.97468890869329e-06,
      "loss": 0.0047,
      "reward": 1.734375,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 170
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 441.65625,
      "epoch": 0.1368,
      "grad_norm": 0.8540055181158542,
      "kl": 0.1103515625,
      "learning_rate": 4.974390766259353e-06,
      "loss": 0.0044,
      "reward": 1.703125,
      "reward_std": 0.2597545385360718,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 171
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 454.734375,
      "epoch": 0.1376,
      "grad_norm": 0.5927539427136594,
      "kl": 0.1162109375,
      "learning_rate": 4.974090887202726e-06,
      "loss": 0.0047,
      "reward": 1.765625,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 172
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 446.171875,
      "epoch": 0.1384,
      "grad_norm": 0.8255796715852812,
      "kl": 0.119140625,
      "learning_rate": 4.973789271733877e-06,
      "loss": 0.0048,
      "reward": 1.796875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 173
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 464.390625,
      "epoch": 0.1392,
      "grad_norm": 0.08165960439046371,
      "kl": 0.10302734375,
      "learning_rate": 4.973485920064491e-06,
      "loss": 0.0041,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 174
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 430.578125,
      "epoch": 0.14,
      "grad_norm": 1.2656076828941412,
      "kl": 0.12353515625,
      "learning_rate": 4.973180832407471e-06,
      "loss": 0.0049,
      "reward": 1.703125,
      "reward_std": 0.38664889335632324,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 175
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 395.46875,
      "epoch": 0.1408,
      "grad_norm": 0.9967437669996866,
      "kl": 0.130859375,
      "learning_rate": 4.97287400897694e-06,
      "loss": 0.0052,
      "reward": 1.59375,
      "reward_std": 0.2709311842918396,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 176
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 408.203125,
      "epoch": 0.1416,
      "grad_norm": 1.2284270222943057,
      "kl": 0.1376953125,
      "learning_rate": 4.972565449988238e-06,
      "loss": 0.0055,
      "reward": 1.640625,
      "reward_std": 0.2687061131000519,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 177
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 415.90625,
      "epoch": 0.1424,
      "grad_norm": 1.3762326715144548,
      "kl": 0.140625,
      "learning_rate": 4.972255155657925e-06,
      "loss": 0.0056,
      "reward": 1.328125,
      "reward_std": 0.2824692726135254,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 178
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 459.703125,
      "epoch": 0.1432,
      "grad_norm": 0.6842166715194086,
      "kl": 0.111328125,
      "learning_rate": 4.9719431262037755e-06,
      "loss": 0.0044,
      "reward": 1.6875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 179
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 460.546875,
      "epoch": 0.144,
      "grad_norm": 0.8638732688380146,
      "kl": 0.12890625,
      "learning_rate": 4.971629361844785e-06,
      "loss": 0.0052,
      "reward": 1.65625,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 180
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 422.921875,
      "epoch": 0.1448,
      "grad_norm": 1.1470878317233275,
      "kl": 0.123046875,
      "learning_rate": 4.971313862801166e-06,
      "loss": 0.0049,
      "reward": 1.640625,
      "reward_std": 0.24831004440784454,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 181
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 467.40625,
      "epoch": 0.1456,
      "grad_norm": 1.2298228677872869,
      "kl": 0.1259765625,
      "learning_rate": 4.9709966292943455e-06,
      "loss": 0.005,
      "reward": 1.8125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 182
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 502.609375,
      "epoch": 0.1464,
      "grad_norm": 5.045132796818451,
      "kl": 0.11767578125,
      "learning_rate": 4.970677661546972e-06,
      "loss": 0.0047,
      "reward": 1.71875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 183
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 520.8125,
      "epoch": 0.1472,
      "grad_norm": 0.9572797889763173,
      "kl": 0.109375,
      "learning_rate": 4.970356959782909e-06,
      "loss": 0.0044,
      "reward": 1.859375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 184
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 558.609375,
      "epoch": 0.148,
      "grad_norm": 0.8604921513908739,
      "kl": 0.111328125,
      "learning_rate": 4.970034524227239e-06,
      "loss": 0.0044,
      "reward": 1.84375,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 185
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 510.625,
      "epoch": 0.1488,
      "grad_norm": 1.1377042956515393,
      "kl": 0.130859375,
      "learning_rate": 4.969710355106256e-06,
      "loss": 0.0052,
      "reward": 1.4375,
      "reward_std": 0.3424547016620636,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 186
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 545.859375,
      "epoch": 0.1496,
      "grad_norm": 0.7689950103002073,
      "kl": 0.11572265625,
      "learning_rate": 4.969384452647477e-06,
      "loss": 0.0046,
      "reward": 1.734375,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.96875,
      "step": 187
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 536.71875,
      "epoch": 0.1504,
      "grad_norm": 0.8912646596770333,
      "kl": 0.107421875,
      "learning_rate": 4.969056817079633e-06,
      "loss": 0.0043,
      "reward": 1.84375,
      "reward_std": 0.19149437546730042,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 188
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 509.21875,
      "epoch": 0.1512,
      "grad_norm": 1.002615987813076,
      "kl": 0.1298828125,
      "learning_rate": 4.968727448632669e-06,
      "loss": 0.0052,
      "reward": 1.53125,
      "reward_std": 0.25513994693756104,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 189
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 555.1875,
      "epoch": 0.152,
      "grad_norm": 1.6278003157000074,
      "kl": 0.123046875,
      "learning_rate": 4.968396347537751e-06,
      "loss": 0.0049,
      "reward": 1.765625,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 190
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 518.546875,
      "epoch": 0.1528,
      "grad_norm": 0.8831829921163867,
      "kl": 0.1142578125,
      "learning_rate": 4.968063514027258e-06,
      "loss": 0.0046,
      "reward": 1.78125,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 191
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 570.671875,
      "epoch": 0.1536,
      "grad_norm": 0.7708673303650967,
      "kl": 0.12451171875,
      "learning_rate": 4.967728948334784e-06,
      "loss": 0.005,
      "reward": 1.734375,
      "reward_std": 0.20873048901557922,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 192
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 599.734375,
      "epoch": 0.1544,
      "grad_norm": 0.9116648653068451,
      "kl": 0.10546875,
      "learning_rate": 4.967392650695141e-06,
      "loss": 0.0042,
      "reward": 1.625,
      "reward_std": 0.2756394147872925,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.953125,
      "step": 193
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 482.828125,
      "epoch": 0.1552,
      "grad_norm": 1.185885422942044,
      "kl": 0.126953125,
      "learning_rate": 4.967054621344356e-06,
      "loss": 0.0051,
      "reward": 1.625,
      "reward_std": 0.3514062762260437,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.96875,
      "step": 194
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 499.515625,
      "epoch": 0.156,
      "grad_norm": 0.8173946853176801,
      "kl": 0.115234375,
      "learning_rate": 4.96671486051967e-06,
      "loss": 0.0046,
      "reward": 1.65625,
      "reward_std": 0.1962025910615921,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 195
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 553.515625,
      "epoch": 0.1568,
      "grad_norm": 0.7455450159086521,
      "kl": 0.11865234375,
      "learning_rate": 4.966373368459542e-06,
      "loss": 0.0048,
      "reward": 1.578125,
      "reward_std": 0.26735979318618774,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 196
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 459.140625,
      "epoch": 0.1576,
      "grad_norm": 0.9292556255485958,
      "kl": 0.12158203125,
      "learning_rate": 4.966030145403642e-06,
      "loss": 0.0049,
      "reward": 1.59375,
      "reward_std": 0.27019423246383667,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 197
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 562.765625,
      "epoch": 0.1584,
      "grad_norm": 0.7768499393593932,
      "kl": 0.1298828125,
      "learning_rate": 4.965685191592859e-06,
      "loss": 0.0052,
      "reward": 1.796875,
      "reward_std": 0.2902791500091553,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 198
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 431.5625,
      "epoch": 0.1592,
      "grad_norm": 0.6450068913255618,
      "kl": 0.11328125,
      "learning_rate": 4.9653385072692935e-06,
      "loss": 0.0045,
      "reward": 1.78125,
      "reward_std": 0.14480239152908325,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 199
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 447.375,
      "epoch": 0.16,
      "grad_norm": 2.1890081897816867,
      "kl": 0.12890625,
      "learning_rate": 4.964990092676263e-06,
      "loss": 0.0051,
      "reward": 1.640625,
      "reward_std": 0.38664889335632324,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 200
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 408.921875,
      "epoch": 0.1608,
      "grad_norm": 0.9503745916839271,
      "kl": 0.1220703125,
      "learning_rate": 4.964639948058297e-06,
      "loss": 0.0049,
      "reward": 1.84375,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 201
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 426.609375,
      "epoch": 0.1616,
      "grad_norm": 0.6309582728790826,
      "kl": 0.107421875,
      "learning_rate": 4.964288073661142e-06,
      "loss": 0.0043,
      "reward": 1.84375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 202
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 488.328125,
      "epoch": 0.1624,
      "grad_norm": 0.6938703856880744,
      "kl": 0.119140625,
      "learning_rate": 4.963934469731756e-06,
      "loss": 0.0048,
      "reward": 1.765625,
      "reward_std": 0.1893727034330368,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 203
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 471.640625,
      "epoch": 0.1632,
      "grad_norm": 1.0495615953561261,
      "kl": 0.11181640625,
      "learning_rate": 4.963579136518312e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.26409149169921875,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 204
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 461.875,
      "epoch": 0.164,
      "grad_norm": 0.8342885507359887,
      "kl": 0.1162109375,
      "learning_rate": 4.963222074270197e-06,
      "loss": 0.0046,
      "reward": 1.59375,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 205
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 466.0,
      "epoch": 0.1648,
      "grad_norm": 0.5996858178874089,
      "kl": 0.130859375,
      "learning_rate": 4.962863283238011e-06,
      "loss": 0.0052,
      "reward": 1.734375,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 206
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 462.984375,
      "epoch": 0.1656,
      "grad_norm": 15.684679168053783,
      "kl": 0.11474609375,
      "learning_rate": 4.962502763673566e-06,
      "loss": 0.0046,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 207
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 411.453125,
      "epoch": 0.1664,
      "grad_norm": 1.2234997917416828,
      "kl": 0.1240234375,
      "learning_rate": 4.96214051582989e-06,
      "loss": 0.005,
      "reward": 1.609375,
      "reward_std": 0.2519446909427643,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 208
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 469.046875,
      "epoch": 0.1672,
      "grad_norm": 3.6075728764664015,
      "kl": 0.12158203125,
      "learning_rate": 4.961776539961222e-06,
      "loss": 0.0049,
      "reward": 1.71875,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 209
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 467.609375,
      "epoch": 0.168,
      "grad_norm": 1.0093347119497476,
      "kl": 0.12109375,
      "learning_rate": 4.961410836323014e-06,
      "loss": 0.0048,
      "reward": 1.546875,
      "reward_std": 0.2777610421180725,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 210
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 481.53125,
      "epoch": 0.1688,
      "grad_norm": 1.4890072026116983,
      "kl": 0.11279296875,
      "learning_rate": 4.961043405171931e-06,
      "loss": 0.0045,
      "reward": 1.671875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 211
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 477.5625,
      "epoch": 0.1696,
      "grad_norm": 0.9493457601594587,
      "kl": 0.10888671875,
      "learning_rate": 4.9606742467658505e-06,
      "loss": 0.0044,
      "reward": 1.5625,
      "reward_std": 0.2540663480758667,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 212
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 395.40625,
      "epoch": 0.1704,
      "grad_norm": 1.2679381611265614,
      "kl": 0.1240234375,
      "learning_rate": 4.960303361363863e-06,
      "loss": 0.0049,
      "reward": 1.78125,
      "reward_std": 0.2540663480758667,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 213
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 485.53125,
      "epoch": 0.1712,
      "grad_norm": 0.9454250792077608,
      "kl": 0.11083984375,
      "learning_rate": 4.959930749226269e-06,
      "loss": 0.0044,
      "reward": 1.640625,
      "reward_std": 0.3352486789226532,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.953125,
      "step": 214
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 541.421875,
      "epoch": 0.172,
      "grad_norm": 0.7956760671345361,
      "kl": 0.1083984375,
      "learning_rate": 4.9595564106145825e-06,
      "loss": 0.0043,
      "reward": 1.796875,
      "reward_std": 0.25726157426834106,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 215
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 485.796875,
      "epoch": 0.1728,
      "grad_norm": 0.9430121875636792,
      "kl": 0.109375,
      "learning_rate": 4.959180345791528e-06,
      "loss": 0.0044,
      "reward": 1.828125,
      "reward_std": 0.28460076451301575,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 216
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 480.90625,
      "epoch": 0.1736,
      "grad_norm": 0.7171185322557593,
      "kl": 0.0986328125,
      "learning_rate": 4.958802555021042e-06,
      "loss": 0.004,
      "reward": 1.875,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 217
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 473.796875,
      "epoch": 0.1744,
      "grad_norm": 0.5353140087390248,
      "kl": 0.115234375,
      "learning_rate": 4.958423038568274e-06,
      "loss": 0.0046,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 218
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 485.4375,
      "epoch": 0.1752,
      "grad_norm": 0.5570596760557941,
      "kl": 0.103515625,
      "learning_rate": 4.958041796699583e-06,
      "loss": 0.0041,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 219
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 448.890625,
      "epoch": 0.176,
      "grad_norm": 0.7544119083860452,
      "kl": 0.11328125,
      "learning_rate": 4.957658829682539e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 220
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 437.34375,
      "epoch": 0.1768,
      "grad_norm": 1.1584516801614175,
      "kl": 0.126953125,
      "learning_rate": 4.9572741377859225e-06,
      "loss": 0.0051,
      "reward": 1.78125,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 221
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 511.46875,
      "epoch": 0.1776,
      "grad_norm": 1.0657478691247289,
      "kl": 0.10595703125,
      "learning_rate": 4.956887721279726e-06,
      "loss": 0.0042,
      "reward": 1.734375,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 222
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 480.546875,
      "epoch": 0.1784,
      "grad_norm": 0.6005686701039716,
      "kl": 0.10888671875,
      "learning_rate": 4.95649958043515e-06,
      "loss": 0.0043,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 223
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 557.609375,
      "epoch": 0.1792,
      "grad_norm": 0.7167623609538366,
      "kl": 0.09912109375,
      "learning_rate": 4.956109715524609e-06,
      "loss": 0.004,
      "reward": 1.828125,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 224
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 445.765625,
      "epoch": 0.18,
      "grad_norm": 1.1325882965070015,
      "kl": 0.1083984375,
      "learning_rate": 4.9557181268217225e-06,
      "loss": 0.0043,
      "reward": 1.71875,
      "reward_std": 0.3104073107242584,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 225
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 462.71875,
      "epoch": 0.1808,
      "grad_norm": 1.1803208487766932,
      "kl": 0.119140625,
      "learning_rate": 4.955324814601324e-06,
      "loss": 0.0048,
      "reward": 1.6875,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 226
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 446.671875,
      "epoch": 0.1816,
      "grad_norm": 0.6396309044211909,
      "kl": 0.10791015625,
      "learning_rate": 4.954929779139455e-06,
      "loss": 0.0043,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 227
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 485.78125,
      "epoch": 0.1824,
      "grad_norm": 0.9122035852929374,
      "kl": 0.10400390625,
      "learning_rate": 4.954533020713367e-06,
      "loss": 0.0042,
      "reward": 1.828125,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 228
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.625,
      "completion_length": 443.296875,
      "epoch": 0.1832,
      "grad_norm": 0.43387655485080506,
      "kl": 0.123046875,
      "learning_rate": 4.954134539601519e-06,
      "loss": 0.0049,
      "reward": 1.359375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 229
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 452.015625,
      "epoch": 0.184,
      "grad_norm": 0.8096554413694654,
      "kl": 0.1298828125,
      "learning_rate": 4.953734336083582e-06,
      "loss": 0.0052,
      "reward": 1.71875,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 230
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 426.5625,
      "epoch": 0.1848,
      "grad_norm": 0.6413121537106756,
      "kl": 0.1142578125,
      "learning_rate": 4.953332410440434e-06,
      "loss": 0.0046,
      "reward": 1.6875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 231
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 524.484375,
      "epoch": 0.1856,
      "grad_norm": 1.03331732968214,
      "kl": 0.1015625,
      "learning_rate": 4.952928762954161e-06,
      "loss": 0.0041,
      "reward": 1.515625,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 232
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 475.0625,
      "epoch": 0.1864,
      "grad_norm": 0.37423563486247585,
      "kl": 0.10107421875,
      "learning_rate": 4.952523393908059e-06,
      "loss": 0.004,
      "reward": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 233
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 415.375,
      "epoch": 0.1872,
      "grad_norm": 0.778531578943487,
      "kl": 0.115234375,
      "learning_rate": 4.952116303586631e-06,
      "loss": 0.0046,
      "reward": 1.734375,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 234
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 536.109375,
      "epoch": 0.188,
      "grad_norm": 0.7253260385102834,
      "kl": 0.0966796875,
      "learning_rate": 4.951707492275589e-06,
      "loss": 0.0039,
      "reward": 1.875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 235
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 507.40625,
      "epoch": 0.1888,
      "grad_norm": 0.778897690244807,
      "kl": 0.10986328125,
      "learning_rate": 4.951296960261853e-06,
      "loss": 0.0044,
      "reward": 1.796875,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 236
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 484.21875,
      "epoch": 0.1896,
      "grad_norm": 0.9712714231037451,
      "kl": 0.1123046875,
      "learning_rate": 4.95088470783355e-06,
      "loss": 0.0045,
      "reward": 1.578125,
      "reward_std": 0.2777610421180725,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 237
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 543.5,
      "epoch": 0.1904,
      "grad_norm": 0.7958523077567334,
      "kl": 0.10595703125,
      "learning_rate": 4.950470735280013e-06,
      "loss": 0.0042,
      "reward": 1.6875,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 238
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 507.09375,
      "epoch": 0.1912,
      "grad_norm": 0.09715496428711683,
      "kl": 0.0986328125,
      "learning_rate": 4.950055042891786e-06,
      "loss": 0.004,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 239
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 536.703125,
      "epoch": 0.192,
      "grad_norm": 1.0485606281171582,
      "kl": 0.111328125,
      "learning_rate": 4.949637630960618e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 240
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 506.53125,
      "epoch": 0.1928,
      "grad_norm": 0.8799111721141154,
      "kl": 0.11767578125,
      "learning_rate": 4.949218499779462e-06,
      "loss": 0.0047,
      "reward": 1.609375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 241
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 503.9375,
      "epoch": 0.1936,
      "grad_norm": 1.1528474685863739,
      "kl": 0.115234375,
      "learning_rate": 4.948797649642484e-06,
      "loss": 0.0046,
      "reward": 1.53125,
      "reward_std": 0.19149437546730042,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 242
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 583.390625,
      "epoch": 0.1944,
      "grad_norm": 0.635291825122285,
      "kl": 0.11279296875,
      "learning_rate": 4.94837508084505e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.26124268770217896,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 243
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 556.90625,
      "epoch": 0.1952,
      "grad_norm": 0.708666658482898,
      "kl": 0.11669921875,
      "learning_rate": 4.9479507936837364e-06,
      "loss": 0.0047,
      "reward": 1.84375,
      "reward_std": 0.19149437546730042,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 244
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 474.171875,
      "epoch": 0.196,
      "grad_norm": 0.843492251351777,
      "kl": 0.11572265625,
      "learning_rate": 4.947524788456325e-06,
      "loss": 0.0046,
      "reward": 1.640625,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 245
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 515.984375,
      "epoch": 0.1968,
      "grad_norm": 1.128475107879419,
      "kl": 0.1142578125,
      "learning_rate": 4.947097065461801e-06,
      "loss": 0.0046,
      "reward": 1.8125,
      "reward_std": 0.22558549046516418,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.96875,
      "step": 246
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 478.703125,
      "epoch": 0.1976,
      "grad_norm": 1.3156818737169922,
      "kl": 0.12255859375,
      "learning_rate": 4.946667625000358e-06,
      "loss": 0.0049,
      "reward": 1.578125,
      "reward_std": 0.2472364604473114,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 247
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 480.96875,
      "epoch": 0.1984,
      "grad_norm": 0.5624094172009639,
      "kl": 0.1064453125,
      "learning_rate": 4.946236467373392e-06,
      "loss": 0.0043,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 248
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 470.671875,
      "epoch": 0.1992,
      "grad_norm": 0.7739082371087889,
      "kl": 0.11865234375,
      "learning_rate": 4.945803592883509e-06,
      "loss": 0.0047,
      "reward": 1.78125,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 249
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 480.828125,
      "epoch": 0.2,
      "grad_norm": 0.6262767300134252,
      "kl": 0.1005859375,
      "learning_rate": 4.9453690018345144e-06,
      "loss": 0.004,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 250
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 539.578125,
      "epoch": 0.2008,
      "grad_norm": 0.5808817933264756,
      "kl": 0.09375,
      "learning_rate": 4.944932694531423e-06,
      "loss": 0.0037,
      "reward": 1.8125,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.96875,
      "step": 251
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 436.484375,
      "epoch": 0.2016,
      "grad_norm": 1.002185621612387,
      "kl": 0.10107421875,
      "learning_rate": 4.94449467128045e-06,
      "loss": 0.0041,
      "reward": 1.859375,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 252
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 474.703125,
      "epoch": 0.2024,
      "grad_norm": 0.7756433117997087,
      "kl": 0.09912109375,
      "learning_rate": 4.944054932389018e-06,
      "loss": 0.004,
      "reward": 1.6875,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 253
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 477.34375,
      "epoch": 0.2032,
      "grad_norm": 0.976580739419774,
      "kl": 0.10546875,
      "learning_rate": 4.943613478165753e-06,
      "loss": 0.0042,
      "reward": 1.71875,
      "reward_std": 0.34929442405700684,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 254
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 457.78125,
      "epoch": 0.204,
      "grad_norm": 1.0544302690718284,
      "kl": 0.1142578125,
      "learning_rate": 4.943170308920484e-06,
      "loss": 0.0046,
      "reward": 1.546875,
      "reward_std": 0.34208837151527405,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 255
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 471.0625,
      "epoch": 0.2048,
      "grad_norm": 0.9717316604564764,
      "kl": 0.1005859375,
      "learning_rate": 4.9427254249642445e-06,
      "loss": 0.004,
      "reward": 1.515625,
      "reward_std": 0.3021277189254761,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.96875,
      "step": 256
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 519.953125,
      "epoch": 0.2056,
      "grad_norm": 0.7824221426335534,
      "kl": 0.1064453125,
      "learning_rate": 4.942278826609272e-06,
      "loss": 0.0043,
      "reward": 1.59375,
      "reward_std": 0.28883427381515503,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 257
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 446.75,
      "epoch": 0.2064,
      "grad_norm": 0.8664602612014396,
      "kl": 0.10009765625,
      "learning_rate": 4.9418305141690045e-06,
      "loss": 0.004,
      "reward": 1.671875,
      "reward_std": 0.23568853735923767,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 258
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 417.265625,
      "epoch": 0.2072,
      "grad_norm": 0.5245421561960985,
      "kl": 0.09716796875,
      "learning_rate": 4.9413804879580865e-06,
      "loss": 0.0039,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 259
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 378.28125,
      "epoch": 0.208,
      "grad_norm": 0.5647553609893893,
      "kl": 0.11376953125,
      "learning_rate": 4.940928748292363e-06,
      "loss": 0.0046,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 260
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 446.234375,
      "epoch": 0.2088,
      "grad_norm": 0.6984961037767852,
      "kl": 0.1025390625,
      "learning_rate": 4.940475295488882e-06,
      "loss": 0.0041,
      "reward": 1.765625,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 261
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 484.65625,
      "epoch": 0.2096,
      "grad_norm": 0.6968168736839139,
      "kl": 0.099609375,
      "learning_rate": 4.940020129865895e-06,
      "loss": 0.004,
      "reward": 1.78125,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 262
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 426.328125,
      "epoch": 0.2104,
      "grad_norm": 0.9782982162086745,
      "kl": 0.103515625,
      "learning_rate": 4.9395632517428546e-06,
      "loss": 0.0041,
      "reward": 1.578125,
      "reward_std": 0.3608325123786926,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 263
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 398.625,
      "epoch": 0.2112,
      "grad_norm": 0.4489442102308692,
      "kl": 0.09912109375,
      "learning_rate": 4.939104661440415e-06,
      "loss": 0.004,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 264
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 414.53125,
      "epoch": 0.212,
      "grad_norm": 0.8169469654738788,
      "kl": 0.1171875,
      "learning_rate": 4.938644359280433e-06,
      "loss": 0.0047,
      "reward": 1.71875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 265
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 411.359375,
      "epoch": 0.2128,
      "grad_norm": 1.2032151789560201,
      "kl": 0.11181640625,
      "learning_rate": 4.938182345585967e-06,
      "loss": 0.0045,
      "reward": 1.546875,
      "reward_std": 0.32407689094543457,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 266
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 465.46875,
      "epoch": 0.2136,
      "grad_norm": 0.5442903107655314,
      "kl": 0.10009765625,
      "learning_rate": 4.937718620681273e-06,
      "loss": 0.004,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 267
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 418.90625,
      "epoch": 0.2144,
      "grad_norm": 0.9270822595209119,
      "kl": 0.1064453125,
      "learning_rate": 4.9372531848918145e-06,
      "loss": 0.0043,
      "reward": 1.5,
      "reward_std": 0.2573973536491394,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 268
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 442.75,
      "epoch": 0.2152,
      "grad_norm": 0.5784340447468568,
      "kl": 0.10498046875,
      "learning_rate": 4.936786038544251e-06,
      "loss": 0.0042,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 269
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 493.53125,
      "epoch": 0.216,
      "grad_norm": 0.5053126414260094,
      "kl": 0.10009765625,
      "learning_rate": 4.9363171819664434e-06,
      "loss": 0.004,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 270
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 443.9375,
      "epoch": 0.2168,
      "grad_norm": 0.7760369870867544,
      "kl": 0.10107421875,
      "learning_rate": 4.9358466154874535e-06,
      "loss": 0.004,
      "reward": 1.734375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 271
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 420.125,
      "epoch": 0.2176,
      "grad_norm": 1.14786779748916,
      "kl": 0.11669921875,
      "learning_rate": 4.935374339437543e-06,
      "loss": 0.0047,
      "reward": 1.5,
      "reward_std": 0.2540663480758667,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.984375,
      "step": 272
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 481.609375,
      "epoch": 0.2184,
      "grad_norm": 1.0288369680601825,
      "kl": 0.103515625,
      "learning_rate": 4.934900354148173e-06,
      "loss": 0.0041,
      "reward": 1.8125,
      "reward_std": 0.2708277702331543,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 273
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 499.609375,
      "epoch": 0.2192,
      "grad_norm": 0.9289192454082107,
      "kl": 0.0966796875,
      "learning_rate": 4.934424659952006e-06,
      "loss": 0.0039,
      "reward": 1.515625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 274
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 468.28125,
      "epoch": 0.22,
      "grad_norm": 0.6845760409425516,
      "kl": 0.09765625,
      "learning_rate": 4.933947257182901e-06,
      "loss": 0.0039,
      "reward": 1.734375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 275
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 439.25,
      "epoch": 0.2208,
      "grad_norm": 0.5347567869972326,
      "kl": 0.10205078125,
      "learning_rate": 4.933468146175918e-06,
      "loss": 0.0041,
      "reward": 1.65625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 276
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 444.65625,
      "epoch": 0.2216,
      "grad_norm": 0.5296343921953094,
      "kl": 0.09716796875,
      "learning_rate": 4.932987327267317e-06,
      "loss": 0.0039,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 277
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 458.546875,
      "epoch": 0.2224,
      "grad_norm": 0.7431199028703543,
      "kl": 0.171875,
      "learning_rate": 4.932504800794553e-06,
      "loss": 0.0069,
      "reward": 1.671875,
      "reward_std": 0.26735979318618774,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 278
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 420.46875,
      "epoch": 0.2232,
      "grad_norm": 0.7454584011248475,
      "kl": 0.10205078125,
      "learning_rate": 4.9320205670962815e-06,
      "loss": 0.0041,
      "reward": 1.828125,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 279
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 462.09375,
      "epoch": 0.224,
      "grad_norm": 0.9510667076624594,
      "kl": 0.1103515625,
      "learning_rate": 4.931534626512359e-06,
      "loss": 0.0044,
      "reward": 1.578125,
      "reward_std": 0.2824692726135254,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 280
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 400.28125,
      "epoch": 0.2248,
      "grad_norm": 0.7687280891983295,
      "kl": 0.11083984375,
      "learning_rate": 4.931046979383836e-06,
      "loss": 0.0044,
      "reward": 1.859375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 281
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 411.84375,
      "epoch": 0.2256,
      "grad_norm": 0.7574221662954138,
      "kl": 0.111328125,
      "learning_rate": 4.930557626052961e-06,
      "loss": 0.0045,
      "reward": 1.65625,
      "reward_std": 0.1962026059627533,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 282
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 418.75,
      "epoch": 0.2264,
      "grad_norm": 0.8783907034453455,
      "kl": 0.1044921875,
      "learning_rate": 4.930066566863182e-06,
      "loss": 0.0042,
      "reward": 1.78125,
      "reward_std": 0.2540663480758667,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 283
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 389.96875,
      "epoch": 0.2272,
      "grad_norm": 0.9774193184682641,
      "kl": 0.115234375,
      "learning_rate": 4.929573802159143e-06,
      "loss": 0.0046,
      "reward": 1.640625,
      "reward_std": 0.34717273712158203,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 284
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 447.40625,
      "epoch": 0.228,
      "grad_norm": 1.0777899832373057,
      "kl": 0.115234375,
      "learning_rate": 4.929079332286685e-06,
      "loss": 0.0046,
      "reward": 1.546875,
      "reward_std": 0.40822190046310425,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 285
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 442.828125,
      "epoch": 0.2288,
      "grad_norm": 0.9185695394739847,
      "kl": 0.1298828125,
      "learning_rate": 4.928583157592846e-06,
      "loss": 0.0052,
      "reward": 1.71875,
      "reward_std": 0.32195520401000977,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 286
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 413.1875,
      "epoch": 0.2296,
      "grad_norm": 1.0338072709985884,
      "kl": 0.10693359375,
      "learning_rate": 4.928085278425862e-06,
      "loss": 0.0043,
      "reward": 1.59375,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 287
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 461.390625,
      "epoch": 0.2304,
      "grad_norm": 0.8021566880229952,
      "kl": 0.1103515625,
      "learning_rate": 4.927585695135162e-06,
      "loss": 0.0044,
      "reward": 1.6875,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 288
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 387.6875,
      "epoch": 0.2312,
      "grad_norm": 0.9297237542138639,
      "kl": 0.1044921875,
      "learning_rate": 4.9270844080713735e-06,
      "loss": 0.0042,
      "reward": 1.8125,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 289
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 435.71875,
      "epoch": 0.232,
      "grad_norm": 0.8541926001316015,
      "kl": 0.09765625,
      "learning_rate": 4.926581417586319e-06,
      "loss": 0.0039,
      "reward": 1.609375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 290
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 364.390625,
      "epoch": 0.2328,
      "grad_norm": 1.124317503956043,
      "kl": 0.109375,
      "learning_rate": 4.926076724033016e-06,
      "loss": 0.0044,
      "reward": 1.65625,
      "reward_std": 0.26409149169921875,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 291
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 410.328125,
      "epoch": 0.2336,
      "grad_norm": 0.4502816614200057,
      "kl": 0.11279296875,
      "learning_rate": 4.925570327765678e-06,
      "loss": 0.0045,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 292
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 448.171875,
      "epoch": 0.2344,
      "grad_norm": 0.45357608231069996,
      "kl": 0.087890625,
      "learning_rate": 4.9250622291397144e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 293
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 437.8125,
      "epoch": 0.2352,
      "grad_norm": 0.9337827458070511,
      "kl": 0.10888671875,
      "learning_rate": 4.924552428511727e-06,
      "loss": 0.0044,
      "reward": 1.703125,
      "reward_std": 0.24039676785469055,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 294
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 429.875,
      "epoch": 0.236,
      "grad_norm": 0.9199067291128734,
      "kl": 0.11083984375,
      "learning_rate": 4.924040926239515e-06,
      "loss": 0.0044,
      "reward": 1.71875,
      "reward_std": 0.28247910737991333,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 295
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 423.578125,
      "epoch": 0.2368,
      "grad_norm": 1.6806458457179179,
      "kl": 0.095703125,
      "learning_rate": 4.92352772268207e-06,
      "loss": 0.0038,
      "reward": 1.6875,
      "reward_std": 0.1825428307056427,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 296
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 426.359375,
      "epoch": 0.2376,
      "grad_norm": 0.46539335382214253,
      "kl": 0.11865234375,
      "learning_rate": 4.923012818199576e-06,
      "loss": 0.0047,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 297
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 437.6875,
      "epoch": 0.2384,
      "grad_norm": 0.7415496459961164,
      "kl": 0.10302734375,
      "learning_rate": 4.922496213153416e-06,
      "loss": 0.0041,
      "reward": 1.78125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 298
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 398.328125,
      "epoch": 0.2392,
      "grad_norm": 0.8651793057898831,
      "kl": 0.10693359375,
      "learning_rate": 4.921977907906161e-06,
      "loss": 0.0043,
      "reward": 1.75,
      "reward_std": 0.22201895713806152,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 299
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 455.734375,
      "epoch": 0.24,
      "grad_norm": 1.0135729622759693,
      "kl": 0.1298828125,
      "learning_rate": 4.921457902821578e-06,
      "loss": 0.0052,
      "reward": 1.5,
      "reward_std": 0.29143065214157104,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 300
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 414.984375,
      "epoch": 0.2408,
      "grad_norm": 0.8119903503196184,
      "kl": 0.1181640625,
      "learning_rate": 4.9209361982646275e-06,
      "loss": 0.0047,
      "reward": 1.640625,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 301
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 502.09375,
      "epoch": 0.2416,
      "grad_norm": 0.6331152922884491,
      "kl": 0.10009765625,
      "learning_rate": 4.920412794601461e-06,
      "loss": 0.004,
      "reward": 1.546875,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 302
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 480.53125,
      "epoch": 0.2424,
      "grad_norm": 0.36110170160289956,
      "kl": 0.10693359375,
      "learning_rate": 4.919887692199423e-06,
      "loss": 0.0043,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 303
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 454.59375,
      "epoch": 0.2432,
      "grad_norm": 0.650522167585543,
      "kl": 0.1171875,
      "learning_rate": 4.9193608914270515e-06,
      "loss": 0.0047,
      "reward": 1.546875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 304
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 478.546875,
      "epoch": 0.244,
      "grad_norm": 0.40448868486799067,
      "kl": 0.1083984375,
      "learning_rate": 4.918832392654075e-06,
      "loss": 0.0043,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 305
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 463.84375,
      "epoch": 0.2448,
      "grad_norm": 0.38978931826325497,
      "kl": 0.11962890625,
      "learning_rate": 4.9183021962514145e-06,
      "loss": 0.0048,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 306
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 415.375,
      "epoch": 0.2456,
      "grad_norm": 1.0986331700929117,
      "kl": 0.1162109375,
      "learning_rate": 4.917770302591183e-06,
      "loss": 0.0046,
      "reward": 1.59375,
      "reward_std": 0.3198433816432953,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 307
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 591.125,
      "epoch": 0.2464,
      "grad_norm": 1.1125510217790837,
      "kl": 0.11572265625,
      "learning_rate": 4.917236712046682e-06,
      "loss": 0.0046,
      "reward": 1.828125,
      "reward_std": 0.21320319175720215,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 308
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 512.453125,
      "epoch": 0.2472,
      "grad_norm": 1.0540625145093168,
      "kl": 0.10595703125,
      "learning_rate": 4.9167014249924075e-06,
      "loss": 0.0042,
      "reward": 1.578125,
      "reward_std": 0.2561880052089691,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 309
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 462.84375,
      "epoch": 0.248,
      "grad_norm": 0.8686096717928228,
      "kl": 0.162109375,
      "learning_rate": 4.916164441804044e-06,
      "loss": 0.0065,
      "reward": 1.640625,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 310
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 486.40625,
      "epoch": 0.2488,
      "grad_norm": 0.7270541313642246,
      "kl": 0.1103515625,
      "learning_rate": 4.915625762858467e-06,
      "loss": 0.0044,
      "reward": 1.71875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 311
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 463.75,
      "epoch": 0.2496,
      "grad_norm": 1.0498750119950635,
      "kl": 0.1318359375,
      "learning_rate": 4.915085388533743e-06,
      "loss": 0.0053,
      "reward": 1.609375,
      "reward_std": 0.31983357667922974,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 312
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 494.828125,
      "epoch": 0.2504,
      "grad_norm": 0.7317483564705757,
      "kl": 0.12353515625,
      "learning_rate": 4.914543319209126e-06,
      "loss": 0.0049,
      "reward": 1.640625,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 313
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 458.15625,
      "epoch": 0.2512,
      "grad_norm": 0.9310724758442198,
      "kl": 0.12255859375,
      "learning_rate": 4.913999555265062e-06,
      "loss": 0.0049,
      "reward": 1.59375,
      "reward_std": 0.32805800437927246,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 314
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 545.65625,
      "epoch": 0.252,
      "grad_norm": 18.925354049851833,
      "kl": 0.0947265625,
      "learning_rate": 4.913454097083185e-06,
      "loss": 0.0038,
      "reward": 1.71875,
      "reward_std": 0.240030437707901,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 315
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 480.4375,
      "epoch": 0.2528,
      "grad_norm": 0.6723579513353247,
      "kl": 0.10498046875,
      "learning_rate": 4.912906945046319e-06,
      "loss": 0.0042,
      "reward": 1.78125,
      "reward_std": 0.17570313811302185,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 316
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 542.640625,
      "epoch": 0.2536,
      "grad_norm": 0.7046884052855756,
      "kl": 0.09912109375,
      "learning_rate": 4.912358099538476e-06,
      "loss": 0.004,
      "reward": 1.6875,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 317
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 506.96875,
      "epoch": 0.2544,
      "grad_norm": 0.7943111076995014,
      "kl": 0.11474609375,
      "learning_rate": 4.911807560944858e-06,
      "loss": 0.0046,
      "reward": 1.453125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 318
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 509.59375,
      "epoch": 0.2552,
      "grad_norm": 0.7505085892345437,
      "kl": 0.09375,
      "learning_rate": 4.911255329651852e-06,
      "loss": 0.0038,
      "reward": 1.796875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 319
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 467.328125,
      "epoch": 0.256,
      "grad_norm": 1.3011643228506782,
      "kl": 0.1162109375,
      "learning_rate": 4.910701406047037e-06,
      "loss": 0.0047,
      "reward": 1.578125,
      "reward_std": 0.41388463973999023,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 320
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 445.453125,
      "epoch": 0.2568,
      "grad_norm": 0.6195217573777856,
      "kl": 0.111328125,
      "learning_rate": 4.910145790519177e-06,
      "loss": 0.0045,
      "reward": 1.703125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 321
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 565.296875,
      "epoch": 0.2576,
      "grad_norm": 1.4057130532982316,
      "kl": 0.09326171875,
      "learning_rate": 4.9095884834582256e-06,
      "loss": 0.0037,
      "reward": 1.734375,
      "reward_std": 0.22636085748672485,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 322
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 506.890625,
      "epoch": 0.2584,
      "grad_norm": 0.9573420494635505,
      "kl": 0.10791015625,
      "learning_rate": 4.909029485255321e-06,
      "loss": 0.0043,
      "reward": 1.859375,
      "reward_std": 0.2688094973564148,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 323
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 469.40625,
      "epoch": 0.2592,
      "grad_norm": 1.0911521485399704,
      "kl": 0.10546875,
      "learning_rate": 4.90846879630279e-06,
      "loss": 0.0042,
      "reward": 1.578125,
      "reward_std": 0.2867126166820526,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 324
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 511.453125,
      "epoch": 0.26,
      "grad_norm": 0.40648896945998414,
      "kl": 0.08837890625,
      "learning_rate": 4.907906416994146e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 325
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 483.5625,
      "epoch": 0.2608,
      "grad_norm": 1.0194468410692472,
      "kl": 0.10009765625,
      "learning_rate": 4.907342347724088e-06,
      "loss": 0.004,
      "reward": 1.671875,
      "reward_std": 0.23531240224838257,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 326
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 423.328125,
      "epoch": 0.2616,
      "grad_norm": 0.6452948851174004,
      "kl": 0.11328125,
      "learning_rate": 4.906776588888502e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 327
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 531.59375,
      "epoch": 0.2624,
      "grad_norm": 0.9072527178998807,
      "kl": 0.08935546875,
      "learning_rate": 4.906209140884459e-06,
      "loss": 0.0036,
      "reward": 1.71875,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 328
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 496.84375,
      "epoch": 0.2632,
      "grad_norm": 1.0696754259727737,
      "kl": 0.09716796875,
      "learning_rate": 4.905640004110216e-06,
      "loss": 0.0039,
      "reward": 1.515625,
      "reward_std": 0.255811870098114,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 329
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 495.578125,
      "epoch": 0.264,
      "grad_norm": 0.8335515525216668,
      "kl": 0.1142578125,
      "learning_rate": 4.905069178965215e-06,
      "loss": 0.0046,
      "reward": 1.5,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 330
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 488.734375,
      "epoch": 0.2648,
      "grad_norm": 0.9957266368937032,
      "kl": 0.1064453125,
      "learning_rate": 4.904496665850083e-06,
      "loss": 0.0043,
      "reward": 1.515625,
      "reward_std": 0.34208837151527405,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.96875,
      "step": 331
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 453.734375,
      "epoch": 0.2656,
      "grad_norm": 1.1223237724051203,
      "kl": 0.11669921875,
      "learning_rate": 4.903922465166633e-06,
      "loss": 0.0047,
      "reward": 1.59375,
      "reward_std": 0.35400262475013733,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 332
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 511.5,
      "epoch": 0.2664,
      "grad_norm": 1.2579776360335044,
      "kl": 0.10595703125,
      "learning_rate": 4.903346577317859e-06,
      "loss": 0.0042,
      "reward": 1.546875,
      "reward_std": 0.34564992785453796,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 333
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 477.71875,
      "epoch": 0.2672,
      "grad_norm": 1.2349668070290505,
      "kl": 0.125,
      "learning_rate": 4.902769002707942e-06,
      "loss": 0.005,
      "reward": 1.8125,
      "reward_std": 0.3039487302303314,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.96875,
      "step": 334
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 450.390625,
      "epoch": 0.268,
      "grad_norm": 0.8960974727205545,
      "kl": 0.12109375,
      "learning_rate": 4.902189741742247e-06,
      "loss": 0.0049,
      "reward": 1.6875,
      "reward_std": 0.3139738440513611,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.953125,
      "step": 335
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 514.34375,
      "epoch": 0.2688,
      "grad_norm": 0.706708681387227,
      "kl": 0.11865234375,
      "learning_rate": 4.901608794827321e-06,
      "loss": 0.0047,
      "reward": 1.6875,
      "reward_std": 0.2407432198524475,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 336
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 468.75,
      "epoch": 0.2696,
      "grad_norm": 1.5034128585970563,
      "kl": 0.1259765625,
      "learning_rate": 4.9010261623708945e-06,
      "loss": 0.005,
      "reward": 1.90625,
      "reward_std": 0.15769661962985992,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.96875,
      "step": 337
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 469.15625,
      "epoch": 0.2704,
      "grad_norm": 0.9957358517939294,
      "kl": 0.28515625,
      "learning_rate": 4.900441844781882e-06,
      "loss": 0.0114,
      "reward": 1.53125,
      "reward_std": 0.27281248569488525,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.921875,
      "step": 338
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 442.921875,
      "epoch": 0.2712,
      "grad_norm": 1.03838448773179,
      "kl": 0.126953125,
      "learning_rate": 4.89985584247038e-06,
      "loss": 0.0051,
      "reward": 1.546875,
      "reward_std": 0.26977968215942383,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 339
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 463.515625,
      "epoch": 0.272,
      "grad_norm": 1.1184747989813613,
      "kl": 0.232421875,
      "learning_rate": 4.899268155847667e-06,
      "loss": 0.0093,
      "reward": 1.59375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.953125,
      "step": 340
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 452.140625,
      "epoch": 0.2728,
      "grad_norm": 1.446286599289767,
      "kl": 0.20703125,
      "learning_rate": 4.898678785326205e-06,
      "loss": 0.0083,
      "reward": 1.609375,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 341
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 406.5,
      "epoch": 0.2736,
      "grad_norm": 1.0667490276319753,
      "kl": 0.1337890625,
      "learning_rate": 4.898087731319637e-06,
      "loss": 0.0053,
      "reward": 1.6875,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 342
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 414.25,
      "epoch": 0.2744,
      "grad_norm": 1.8075013197728957,
      "kl": 0.56640625,
      "learning_rate": 4.8974949942427854e-06,
      "loss": 0.0226,
      "reward": 1.3125,
      "reward_std": 0.7984259128570557,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.703125,
      "step": 343
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 409.453125,
      "epoch": 0.2752,
      "grad_norm": 2.0612645399792253,
      "kl": 0.78125,
      "learning_rate": 4.896900574511657e-06,
      "loss": 0.0313,
      "reward": 1.1875,
      "reward_std": 0.6475251913070679,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.734375,
      "step": 344
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 351.328125,
      "epoch": 0.276,
      "grad_norm": 5.117351232455169,
      "kl": 0.54296875,
      "learning_rate": 4.89630447254344e-06,
      "loss": 0.0217,
      "reward": 1.25,
      "reward_std": 0.5729714632034302,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.765625,
      "step": 345
    },
    {
      "all_correct": 0.0,
      "all_wrong": 1.0,
      "completion_length": 247.0,
      "epoch": 0.2768,
      "grad_norm": 24.876604888296548,
      "kl": 6.625,
      "learning_rate": 4.8957066887565005e-06,
      "loss": 0.2649,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "step": 346
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 640.671875,
      "epoch": 0.2776,
      "grad_norm": 201.1175403256699,
      "kl": 4.625,
      "learning_rate": 4.895107223570386e-06,
      "loss": 0.1846,
      "reward": 0.4375,
      "reward_std": 0.6013320684432983,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 0.1875,
      "step": 347
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 353.6875,
      "epoch": 0.2784,
      "grad_norm": 33.61976097623809,
      "kl": 2.4375,
      "learning_rate": 4.894506077405824e-06,
      "loss": 0.0977,
      "reward": 0.859375,
      "reward_std": 0.8206709027290344,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.546875,
      "step": 348
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.875,
      "completion_length": 451.484375,
      "epoch": 0.2792,
      "grad_norm": 626.7174881514043,
      "kl": 5.09375,
      "learning_rate": 4.893903250684723e-06,
      "loss": 0.2029,
      "reward": 0.171875,
      "reward_std": 0.3107786178588867,
      "rewards/accuracy_reward": 0.140625,
      "rewards/format_reward": 0.03125,
      "step": 349
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 332.546875,
      "epoch": 0.28,
      "grad_norm": 28.91852693203601,
      "kl": 0.498046875,
      "learning_rate": 4.893298743830168e-06,
      "loss": 0.0199,
      "reward": 1.640625,
      "reward_std": 0.3841980993747711,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.90625,
      "step": 350
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 343.0625,
      "epoch": 0.2808,
      "grad_norm": 7.171144824263768,
      "kl": 1.40625,
      "learning_rate": 4.892692557266429e-06,
      "loss": 0.0561,
      "reward": 1.1875,
      "reward_std": 0.7015436887741089,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.65625,
      "step": 351
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 255.765625,
      "epoch": 0.2816,
      "grad_norm": 7.446387142941203,
      "kl": 2.46875,
      "learning_rate": 4.8920846914189465e-06,
      "loss": 0.0987,
      "reward": 1.0,
      "reward_std": 1.0010052919387817,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.484375,
      "step": 352
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 279.5,
      "epoch": 0.2824,
      "grad_norm": 7.482555980879127,
      "kl": 2.46875,
      "learning_rate": 4.891475146714348e-06,
      "loss": 0.0987,
      "reward": 1.0,
      "reward_std": 0.9171967506408691,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.546875,
      "step": 353
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 314.140625,
      "epoch": 0.2832,
      "grad_norm": 186.2241805566282,
      "kl": 4.84375,
      "learning_rate": 4.8908639235804324e-06,
      "loss": 0.1935,
      "reward": 0.625,
      "reward_std": 0.7367569804191589,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.34375,
      "step": 354
    },
    {
      "all_correct": 0.0,
      "all_wrong": 1.0,
      "completion_length": 531.34375,
      "epoch": 0.284,
      "grad_norm": 2163.964903161328,
      "kl": 58.0,
      "learning_rate": 4.890251022446181e-06,
      "loss": 2.3272,
      "reward": 0.09375,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.046875,
      "rewards/format_reward": 0.046875,
      "step": 355
    },
    {
      "all_correct": 0.0,
      "all_wrong": 1.0,
      "completion_length": 599.875,
      "epoch": 0.2848,
      "grad_norm": 30.636793031414754,
      "kl": 0.79296875,
      "learning_rate": 4.889636443741752e-06,
      "loss": 0.0317,
      "reward": 0.109375,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.078125,
      "rewards/format_reward": 0.03125,
      "step": 356
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 208.96875,
      "epoch": 0.2856,
      "grad_norm": 5.97530784882832,
      "kl": 2.375,
      "learning_rate": 4.88902018789848e-06,
      "loss": 0.0949,
      "reward": 0.640625,
      "reward_std": 0.7411497831344604,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 0.28125,
      "step": 357
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.75,
      "completion_length": 176.25,
      "epoch": 0.2864,
      "grad_norm": 19.34351483851466,
      "kl": 7.4375,
      "learning_rate": 4.888402255348877e-06,
      "loss": 0.297,
      "reward": 0.234375,
      "reward_std": 0.4060066342353821,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 0.109375,
      "step": 358
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.375,
      "completion_length": 304.015625,
      "epoch": 0.2872,
      "grad_norm": 12.216881362937261,
      "kl": 2.375,
      "learning_rate": 4.887782646526631e-06,
      "loss": 0.0951,
      "reward": 0.90625,
      "reward_std": 0.5779029130935669,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.34375,
      "step": 359
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.75,
      "completion_length": 293.453125,
      "epoch": 0.288,
      "grad_norm": 1.8055399813270507,
      "kl": 0.435546875,
      "learning_rate": 4.887161361866608e-06,
      "loss": 0.0175,
      "reward": 0.546875,
      "reward_std": 0.4071483612060547,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.140625,
      "step": 360
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 294.875,
      "epoch": 0.2888,
      "grad_norm": 2.2464320445436123,
      "kl": 0.2353515625,
      "learning_rate": 4.8865384018048494e-06,
      "loss": 0.0094,
      "reward": 1.046875,
      "reward_std": 0.6361694931983948,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.4375,
      "step": 361
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 277.6875,
      "epoch": 0.2896,
      "grad_norm": 1.5168721375251013,
      "kl": 0.2001953125,
      "learning_rate": 4.8859137667785735e-06,
      "loss": 0.008,
      "reward": 1.390625,
      "reward_std": 0.39560043811798096,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.890625,
      "step": 362
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 369.484375,
      "epoch": 0.2904,
      "grad_norm": 2.9135374435837598,
      "kl": 0.1845703125,
      "learning_rate": 4.8852874572261715e-06,
      "loss": 0.0074,
      "reward": 1.4375,
      "reward_std": 0.4606751799583435,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.671875,
      "step": 363
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 317.03125,
      "epoch": 0.2912,
      "grad_norm": 1.6946308584406236,
      "kl": 0.1884765625,
      "learning_rate": 4.884659473587213e-06,
      "loss": 0.0075,
      "reward": 1.515625,
      "reward_std": 0.4728219509124756,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.78125,
      "step": 364
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 409.3125,
      "epoch": 0.292,
      "grad_norm": 1.9890104705809388,
      "kl": 0.205078125,
      "learning_rate": 4.884029816302441e-06,
      "loss": 0.0082,
      "reward": 1.546875,
      "reward_std": 0.4024401307106018,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.78125,
      "step": 365
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.375,
      "completion_length": 447.046875,
      "epoch": 0.2928,
      "grad_norm": 3.4786297634905567,
      "kl": 0.302734375,
      "learning_rate": 4.883398485813772e-06,
      "loss": 0.0121,
      "reward": 1.15625,
      "reward_std": 0.4918765723705292,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.59375,
      "step": 366
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 599.75,
      "epoch": 0.2936,
      "grad_norm": 16.311891900467852,
      "kl": 0.89453125,
      "learning_rate": 4.8827654825642984e-06,
      "loss": 0.0358,
      "reward": 1.078125,
      "reward_std": 0.6761397123336792,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.421875,
      "step": 367
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 553.78125,
      "epoch": 0.2944,
      "grad_norm": 11.37452807904692,
      "kl": 1.8125,
      "learning_rate": 4.882130806998287e-06,
      "loss": 0.0726,
      "reward": 0.984375,
      "reward_std": 0.7113250494003296,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.484375,
      "step": 368
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 397.421875,
      "epoch": 0.2952,
      "grad_norm": 8.426860834578568,
      "kl": 1.640625,
      "learning_rate": 4.881494459561177e-06,
      "loss": 0.0656,
      "reward": 1.21875,
      "reward_std": 0.7671608924865723,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.78125,
      "step": 369
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 386.421875,
      "epoch": 0.296,
      "grad_norm": 5.772641810637551,
      "kl": 4.125,
      "learning_rate": 4.880856440699582e-06,
      "loss": 0.1649,
      "reward": 1.375,
      "reward_std": 0.708383321762085,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.8125,
      "step": 370
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 293.734375,
      "epoch": 0.2968,
      "grad_norm": 22.04847857153818,
      "kl": 11.5,
      "learning_rate": 4.880216750861288e-06,
      "loss": 0.4595,
      "reward": 1.078125,
      "reward_std": 0.9076728820800781,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.640625,
      "step": 371
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 305.09375,
      "epoch": 0.2976,
      "grad_norm": 16.139049061859996,
      "kl": 12.875,
      "learning_rate": 4.879575390495254e-06,
      "loss": 0.5157,
      "reward": 1.140625,
      "reward_std": 0.8692222833633423,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.65625,
      "step": 372
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 439.140625,
      "epoch": 0.2984,
      "grad_norm": 5.483376733053643,
      "kl": 4.28125,
      "learning_rate": 4.878932360051611e-06,
      "loss": 0.1714,
      "reward": 1.734375,
      "reward_std": 0.5622838735580444,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.890625,
      "step": 373
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 385.96875,
      "epoch": 0.2992,
      "grad_norm": 22.744631135693638,
      "kl": 17.125,
      "learning_rate": 4.878287659981663e-06,
      "loss": 0.6852,
      "reward": 0.984375,
      "reward_std": 0.9501945972442627,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.53125,
      "step": 374
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 631.65625,
      "epoch": 0.3,
      "grad_norm": 10.663644117055918,
      "kl": 5.25,
      "learning_rate": 4.8776412907378845e-06,
      "loss": 0.2108,
      "reward": 1.25,
      "reward_std": 0.7919073700904846,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.734375,
      "step": 375
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 591.59375,
      "epoch": 0.3008,
      "grad_norm": 5.836849178969912,
      "kl": 7.875,
      "learning_rate": 4.876993252773923e-06,
      "loss": 0.315,
      "reward": 1.078125,
      "reward_std": 0.5978821516036987,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.65625,
      "step": 376
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 640.234375,
      "epoch": 0.3016,
      "grad_norm": 4.0425819468893645,
      "kl": 5.3125,
      "learning_rate": 4.876343546544596e-06,
      "loss": 0.2119,
      "reward": 1.25,
      "reward_std": 0.662531316280365,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.734375,
      "step": 377
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 606.0,
      "epoch": 0.3024,
      "grad_norm": 4.353961170477641,
      "kl": 5.625,
      "learning_rate": 4.8756921725058935e-06,
      "loss": 0.2251,
      "reward": 1.40625,
      "reward_std": 0.7833096981048584,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.765625,
      "step": 378
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 520.671875,
      "epoch": 0.3032,
      "grad_norm": 3.907345388335812,
      "kl": 7.9375,
      "learning_rate": 4.875039131114975e-06,
      "loss": 0.3178,
      "reward": 1.46875,
      "reward_std": 0.7728353142738342,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.796875,
      "step": 379
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 609.09375,
      "epoch": 0.304,
      "grad_norm": 3.0197647012699838,
      "kl": 7.0625,
      "learning_rate": 4.8743844228301676e-06,
      "loss": 0.282,
      "reward": 1.34375,
      "reward_std": 0.7501465082168579,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.8125,
      "step": 380
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 630.890625,
      "epoch": 0.3048,
      "grad_norm": 3.328362838541217,
      "kl": 4.75,
      "learning_rate": 4.873728048110973e-06,
      "loss": 0.1908,
      "reward": 1.34375,
      "reward_std": 0.669731855392456,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.828125,
      "step": 381
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 553.75,
      "epoch": 0.3056,
      "grad_norm": 4.545235460323257,
      "kl": 6.78125,
      "learning_rate": 4.873070007418059e-06,
      "loss": 0.2708,
      "reward": 1.296875,
      "reward_std": 0.7942386269569397,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.78125,
      "step": 382
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 543.34375,
      "epoch": 0.3064,
      "grad_norm": 11.300499138026463,
      "kl": 2.296875,
      "learning_rate": 4.872410301213265e-06,
      "loss": 0.092,
      "reward": 1.5,
      "reward_std": 0.5112500190734863,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.859375,
      "step": 383
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 479.453125,
      "epoch": 0.3072,
      "grad_norm": 2.5224974439131205,
      "kl": 1.7421875,
      "learning_rate": 4.871748929959598e-06,
      "loss": 0.0695,
      "reward": 1.546875,
      "reward_std": 0.30617380142211914,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.9375,
      "step": 384
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 522.390625,
      "epoch": 0.308,
      "grad_norm": 8.861165553607586,
      "kl": 2.40625,
      "learning_rate": 4.871085894121234e-06,
      "loss": 0.096,
      "reward": 1.515625,
      "reward_std": 0.37378886342048645,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.90625,
      "step": 385
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 415.375,
      "epoch": 0.3088,
      "grad_norm": 156.19365589168777,
      "kl": 6.8125,
      "learning_rate": 4.870421194163515e-06,
      "loss": 0.2727,
      "reward": 1.578125,
      "reward_std": 0.3775939345359802,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.96875,
      "step": 386
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 406.75,
      "epoch": 0.3096,
      "grad_norm": 27.153773633031346,
      "kl": 0.345703125,
      "learning_rate": 4.869754830552956e-06,
      "loss": 0.0138,
      "reward": 1.796875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 387
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 408.453125,
      "epoch": 0.3104,
      "grad_norm": 1030.1027037042702,
      "kl": 111.5,
      "learning_rate": 4.869086803757235e-06,
      "loss": 4.4696,
      "reward": 1.46875,
      "reward_std": 0.34669801592826843,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.9375,
      "step": 388
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 450.984375,
      "epoch": 0.3112,
      "grad_norm": 7.507009527485788,
      "kl": 0.34375,
      "learning_rate": 4.868417114245199e-06,
      "loss": 0.0137,
      "reward": 1.75,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.9375,
      "step": 389
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 479.765625,
      "epoch": 0.312,
      "grad_norm": 0.4155073762935163,
      "kl": 0.216796875,
      "learning_rate": 4.867745762486862e-06,
      "loss": 0.0087,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 390
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 454.40625,
      "epoch": 0.3128,
      "grad_norm": 17.285174184609716,
      "kl": 0.6171875,
      "learning_rate": 4.8670727489534035e-06,
      "loss": 0.0247,
      "reward": 1.609375,
      "reward_std": 0.40701988339424133,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.90625,
      "step": 391
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 393.25,
      "epoch": 0.3136,
      "grad_norm": 876.5609922977326,
      "kl": 134.0,
      "learning_rate": 4.866398074117173e-06,
      "loss": 5.3539,
      "reward": 1.515625,
      "reward_std": 0.3132994771003723,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.9375,
      "step": 392
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 506.859375,
      "epoch": 0.3144,
      "grad_norm": 0.8228708456368266,
      "kl": 0.248046875,
      "learning_rate": 4.86572173845168e-06,
      "loss": 0.0099,
      "reward": 1.484375,
      "reward_std": 0.35352790355682373,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.984375,
      "step": 393
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 419.296875,
      "epoch": 0.3152,
      "grad_norm": 35.19059370326572,
      "kl": 0.306640625,
      "learning_rate": 4.865043742431605e-06,
      "loss": 0.0122,
      "reward": 1.671875,
      "reward_std": 0.39809340238571167,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.9375,
      "step": 394
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 371.359375,
      "epoch": 0.316,
      "grad_norm": 0.8201978988422712,
      "kl": 0.228515625,
      "learning_rate": 4.864364086532792e-06,
      "loss": 0.0091,
      "reward": 1.640625,
      "reward_std": 0.22707363963127136,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 395
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 454.890625,
      "epoch": 0.3168,
      "grad_norm": 111.8832716959935,
      "kl": 5.0,
      "learning_rate": 4.863682771232249e-06,
      "loss": 0.1995,
      "reward": 1.453125,
      "reward_std": 0.5471649169921875,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.875,
      "step": 396
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 401.109375,
      "epoch": 0.3176,
      "grad_norm": 0.5286369705491609,
      "kl": 0.197265625,
      "learning_rate": 4.862999797008149e-06,
      "loss": 0.0079,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 397
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 479.84375,
      "epoch": 0.3184,
      "grad_norm": 22.95638431876077,
      "kl": 1.453125,
      "learning_rate": 4.862315164339829e-06,
      "loss": 0.0581,
      "reward": 1.734375,
      "reward_std": 0.38578349351882935,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.90625,
      "step": 398
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 466.390625,
      "epoch": 0.3192,
      "grad_norm": 4.65125203900269,
      "kl": 1.6640625,
      "learning_rate": 4.861628873707792e-06,
      "loss": 0.0664,
      "reward": 1.46875,
      "reward_std": 0.5292564630508423,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.890625,
      "step": 399
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.625,
      "completion_length": 393.796875,
      "epoch": 0.32,
      "grad_norm": 1.085220273805104,
      "kl": 0.2265625,
      "learning_rate": 4.860940925593703e-06,
      "loss": 0.009,
      "reward": 1.34375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 400
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 470.21875,
      "epoch": 0.3208,
      "grad_norm": 0.6245969728254261,
      "kl": 0.1767578125,
      "learning_rate": 4.86025132048039e-06,
      "loss": 0.0071,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 401
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 455.5,
      "epoch": 0.3216,
      "grad_norm": 0.7126066128293126,
      "kl": 0.201171875,
      "learning_rate": 4.859560058851844e-06,
      "loss": 0.0081,
      "reward": 1.75,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 402
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 409.4375,
      "epoch": 0.3224,
      "grad_norm": 0.761201915189821,
      "kl": 0.201171875,
      "learning_rate": 4.8588671411932195e-06,
      "loss": 0.008,
      "reward": 1.671875,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 403
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 447.75,
      "epoch": 0.3232,
      "grad_norm": 1.1718892639708978,
      "kl": 0.1865234375,
      "learning_rate": 4.858172567990832e-06,
      "loss": 0.0075,
      "reward": 1.734375,
      "reward_std": 0.29788440465927124,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 404
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 445.3125,
      "epoch": 0.324,
      "grad_norm": 0.8366164839673635,
      "kl": 0.185546875,
      "learning_rate": 4.857476339732162e-06,
      "loss": 0.0074,
      "reward": 1.78125,
      "reward_std": 0.2756393849849701,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 405
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 398.6875,
      "epoch": 0.3248,
      "grad_norm": 0.6147289990534973,
      "kl": 0.1923828125,
      "learning_rate": 4.856778456905846e-06,
      "loss": 0.0077,
      "reward": 1.6875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 406
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 388.484375,
      "epoch": 0.3256,
      "grad_norm": 0.7855381422211676,
      "kl": 0.21875,
      "learning_rate": 4.856078920001689e-06,
      "loss": 0.0087,
      "reward": 1.421875,
      "reward_std": 0.22636085748672485,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.984375,
      "step": 407
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 417.15625,
      "epoch": 0.3264,
      "grad_norm": 2.275516154603678,
      "kl": 2.3125,
      "learning_rate": 4.855377729510648e-06,
      "loss": 0.0929,
      "reward": 1.40625,
      "reward_std": 0.2924008071422577,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.953125,
      "step": 408
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 406.921875,
      "epoch": 0.3272,
      "grad_norm": 9.369765022864346,
      "kl": 1.4921875,
      "learning_rate": 4.8546748859248504e-06,
      "loss": 0.06,
      "reward": 1.609375,
      "reward_std": 0.2993340790271759,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 409
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 401.140625,
      "epoch": 0.328,
      "grad_norm": 1.4444623821966434,
      "kl": 2.265625,
      "learning_rate": 4.853970389737576e-06,
      "loss": 0.0904,
      "reward": 1.515625,
      "reward_std": 0.33351296186447144,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.953125,
      "step": 410
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 429.09375,
      "epoch": 0.3288,
      "grad_norm": 2.549204946324485,
      "kl": 3.0,
      "learning_rate": 4.8532642414432675e-06,
      "loss": 0.1202,
      "reward": 1.609375,
      "reward_std": 0.4376729428768158,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.921875,
      "step": 411
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 417.71875,
      "epoch": 0.3296,
      "grad_norm": 2.656722449352898,
      "kl": 3.34375,
      "learning_rate": 4.852556441537528e-06,
      "loss": 0.1337,
      "reward": 1.25,
      "reward_std": 0.5111948251724243,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 0.90625,
      "step": 412
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 423.28125,
      "epoch": 0.3304,
      "grad_norm": 0.9276053292737674,
      "kl": 1.3359375,
      "learning_rate": 4.851846990517118e-06,
      "loss": 0.0534,
      "reward": 1.65625,
      "reward_std": 0.4088938534259796,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.96875,
      "step": 413
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 475.5,
      "epoch": 0.3312,
      "grad_norm": 1.2562446558868838,
      "kl": 1.5,
      "learning_rate": 4.851135888879958e-06,
      "loss": 0.0601,
      "reward": 1.71875,
      "reward_std": 0.2709311842918396,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.953125,
      "step": 414
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 457.390625,
      "epoch": 0.332,
      "grad_norm": 1.5340077204292817,
      "kl": 1.0,
      "learning_rate": 4.850423137125126e-06,
      "loss": 0.0399,
      "reward": 1.78125,
      "reward_std": 0.3061639666557312,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 415
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 434.6875,
      "epoch": 0.3328,
      "grad_norm": 0.9409909612683872,
      "kl": 0.5859375,
      "learning_rate": 4.8497087357528585e-06,
      "loss": 0.0234,
      "reward": 1.9375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 416
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 425.046875,
      "epoch": 0.3336,
      "grad_norm": 0.545076811233695,
      "kl": 0.1669921875,
      "learning_rate": 4.8489926852645505e-06,
      "loss": 0.0067,
      "reward": 1.703125,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 417
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 409.515625,
      "epoch": 0.3344,
      "grad_norm": 0.9556631606557033,
      "kl": 0.796875,
      "learning_rate": 4.848274986162754e-06,
      "loss": 0.0318,
      "reward": 1.546875,
      "reward_std": 0.28778618574142456,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 418
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 420.34375,
      "epoch": 0.3352,
      "grad_norm": 0.7867848801233414,
      "kl": 0.36328125,
      "learning_rate": 4.847555638951177e-06,
      "loss": 0.0145,
      "reward": 1.828125,
      "reward_std": 0.2662131190299988,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 419
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 368.890625,
      "epoch": 0.336,
      "grad_norm": 0.5755910960305857,
      "kl": 0.1533203125,
      "learning_rate": 4.846834644134686e-06,
      "loss": 0.0061,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 420
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 396.03125,
      "epoch": 0.3368,
      "grad_norm": 0.5315593964989055,
      "kl": 0.1552734375,
      "learning_rate": 4.846112002219301e-06,
      "loss": 0.0062,
      "reward": 1.53125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 421
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 423.734375,
      "epoch": 0.3376,
      "grad_norm": 0.545127285092993,
      "kl": 0.201171875,
      "learning_rate": 4.845387713712203e-06,
      "loss": 0.0081,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 422
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 364.78125,
      "epoch": 0.3384,
      "grad_norm": 1.312387736907392,
      "kl": 0.1533203125,
      "learning_rate": 4.844661779121723e-06,
      "loss": 0.0061,
      "reward": 1.546875,
      "reward_std": 0.28930896520614624,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 423
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 400.71875,
      "epoch": 0.3392,
      "grad_norm": 2.1562952500677324,
      "kl": 0.8515625,
      "learning_rate": 4.843934198957351e-06,
      "loss": 0.034,
      "reward": 1.578125,
      "reward_std": 0.2805546522140503,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 424
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 362.40625,
      "epoch": 0.34,
      "grad_norm": 3.3751659314634805,
      "kl": 0.171875,
      "learning_rate": 4.84320497372973e-06,
      "loss": 0.0069,
      "reward": 1.625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 425
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 402.171875,
      "epoch": 0.3408,
      "grad_norm": 0.8741894143777079,
      "kl": 0.13671875,
      "learning_rate": 4.842474103950658e-06,
      "loss": 0.0055,
      "reward": 1.921875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 426
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 396.1875,
      "epoch": 0.3416,
      "grad_norm": 0.8174260769621657,
      "kl": 0.494140625,
      "learning_rate": 4.841741590133089e-06,
      "loss": 0.0197,
      "reward": 1.59375,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 427
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 369.109375,
      "epoch": 0.3424,
      "grad_norm": 0.5209696523338851,
      "kl": 0.1396484375,
      "learning_rate": 4.841007432791129e-06,
      "loss": 0.0056,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 428
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 428.875,
      "epoch": 0.3432,
      "grad_norm": 0.8153419848070561,
      "kl": 0.169921875,
      "learning_rate": 4.8402716324400375e-06,
      "loss": 0.0068,
      "reward": 1.8125,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 429
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 415.6875,
      "epoch": 0.344,
      "grad_norm": 1.0666154124765679,
      "kl": 0.78125,
      "learning_rate": 4.839534189596228e-06,
      "loss": 0.0311,
      "reward": 1.59375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.96875,
      "step": 430
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 395.859375,
      "epoch": 0.3448,
      "grad_norm": 0.44393856411551624,
      "kl": 0.169921875,
      "learning_rate": 4.8387951047772656e-06,
      "loss": 0.0068,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 431
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 419.28125,
      "epoch": 0.3456,
      "grad_norm": 1.1161358537344634,
      "kl": 2.203125,
      "learning_rate": 4.838054378501868e-06,
      "loss": 0.0883,
      "reward": 1.890625,
      "reward_std": 0.20873048901557922,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.953125,
      "step": 432
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 376.890625,
      "epoch": 0.3464,
      "grad_norm": 0.4075823163043336,
      "kl": 0.1435546875,
      "learning_rate": 4.837312011289907e-06,
      "loss": 0.0057,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 433
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 359.421875,
      "epoch": 0.3472,
      "grad_norm": 1.0816196211554479,
      "kl": 0.1474609375,
      "learning_rate": 4.836568003662403e-06,
      "loss": 0.0059,
      "reward": 1.609375,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 434
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 420.015625,
      "epoch": 0.348,
      "grad_norm": 0.7898993144441228,
      "kl": 1.5,
      "learning_rate": 4.8358223561415304e-06,
      "loss": 0.06,
      "reward": 1.765625,
      "reward_std": 0.20290403068065643,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.953125,
      "step": 435
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 407.921875,
      "epoch": 0.3488,
      "grad_norm": 1.578568176968985,
      "kl": 0.130859375,
      "learning_rate": 4.835075069250613e-06,
      "loss": 0.0053,
      "reward": 1.640625,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 436
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 395.953125,
      "epoch": 0.3496,
      "grad_norm": 0.8569826158727925,
      "kl": 0.1328125,
      "learning_rate": 4.8343261435141245e-06,
      "loss": 0.0053,
      "reward": 1.78125,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 437
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 381.625,
      "epoch": 0.3504,
      "grad_norm": 0.5041244227154292,
      "kl": 0.203125,
      "learning_rate": 4.833575579457691e-06,
      "loss": 0.0081,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 438
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 439.984375,
      "epoch": 0.3512,
      "grad_norm": 0.544866660057303,
      "kl": 0.1357421875,
      "learning_rate": 4.832823377608088e-06,
      "loss": 0.0054,
      "reward": 1.4375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 439
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 403.140625,
      "epoch": 0.352,
      "grad_norm": 1.0531205514657995,
      "kl": 0.1396484375,
      "learning_rate": 4.832069538493237e-06,
      "loss": 0.0056,
      "reward": 1.78125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 440
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 425.703125,
      "epoch": 0.3528,
      "grad_norm": 0.6199752182002176,
      "kl": 0.2041015625,
      "learning_rate": 4.831314062642213e-06,
      "loss": 0.0082,
      "reward": 1.640625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 441
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 453.0625,
      "epoch": 0.3536,
      "grad_norm": 1.4048917434449617,
      "kl": 0.2890625,
      "learning_rate": 4.830556950585239e-06,
      "loss": 0.0116,
      "reward": 1.65625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 442
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 441.734375,
      "epoch": 0.3544,
      "grad_norm": 0.4946040299292862,
      "kl": 0.328125,
      "learning_rate": 4.829798202853683e-06,
      "loss": 0.0131,
      "reward": 1.75,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 443
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 432.59375,
      "epoch": 0.3552,
      "grad_norm": 0.7484184221012916,
      "kl": 0.365234375,
      "learning_rate": 4.829037819980065e-06,
      "loss": 0.0146,
      "reward": 1.546875,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 444
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 372.46875,
      "epoch": 0.356,
      "grad_norm": 0.6506183134020523,
      "kl": 0.1669921875,
      "learning_rate": 4.828275802498051e-06,
      "loss": 0.0067,
      "reward": 1.625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 445
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 430.9375,
      "epoch": 0.3568,
      "grad_norm": 0.9208137570875184,
      "kl": 0.138671875,
      "learning_rate": 4.827512150942454e-06,
      "loss": 0.0056,
      "reward": 1.609375,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 446
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 428.296875,
      "epoch": 0.3576,
      "grad_norm": 1.2098713467531075,
      "kl": 0.4921875,
      "learning_rate": 4.8267468658492335e-06,
      "loss": 0.0197,
      "reward": 1.5,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.984375,
      "step": 447
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 467.296875,
      "epoch": 0.3584,
      "grad_norm": 0.7150741939065385,
      "kl": 0.1220703125,
      "learning_rate": 4.825979947755496e-06,
      "loss": 0.0049,
      "reward": 1.84375,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 448
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 421.4375,
      "epoch": 0.3592,
      "grad_norm": 1.1223282700977706,
      "kl": 0.1416015625,
      "learning_rate": 4.8252113971994955e-06,
      "loss": 0.0057,
      "reward": 1.765625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 449
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 383.265625,
      "epoch": 0.36,
      "grad_norm": 0.5208765265104575,
      "kl": 0.515625,
      "learning_rate": 4.824441214720629e-06,
      "loss": 0.0207,
      "reward": 1.703125,
      "reward_std": 0.13868528604507446,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 450
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 479.421875,
      "epoch": 0.3608,
      "grad_norm": 1.1245698395617494,
      "kl": 0.85546875,
      "learning_rate": 4.823669400859441e-06,
      "loss": 0.0342,
      "reward": 1.71875,
      "reward_std": 0.3618124723434448,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 451
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 496.6875,
      "epoch": 0.3616,
      "grad_norm": 0.7035661119729969,
      "kl": 0.357421875,
      "learning_rate": 4.8228959561576195e-06,
      "loss": 0.0143,
      "reward": 1.578125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 452
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 409.140625,
      "epoch": 0.3624,
      "grad_norm": 0.8392469138583998,
      "kl": 1.3125,
      "learning_rate": 4.822120881157998e-06,
      "loss": 0.0526,
      "reward": 1.6875,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.953125,
      "step": 453
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 478.75,
      "epoch": 0.3632,
      "grad_norm": 1.173091688779587,
      "kl": 0.93359375,
      "learning_rate": 4.821344176404554e-06,
      "loss": 0.0374,
      "reward": 1.421875,
      "reward_std": 0.2867125868797302,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.953125,
      "step": 454
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 397.53125,
      "epoch": 0.364,
      "grad_norm": 1.002096287669717,
      "kl": 0.416015625,
      "learning_rate": 4.820565842442408e-06,
      "loss": 0.0167,
      "reward": 1.703125,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 455
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 510.15625,
      "epoch": 0.3648,
      "grad_norm": 0.9178416021368806,
      "kl": 0.58984375,
      "learning_rate": 4.819785879817827e-06,
      "loss": 0.0236,
      "reward": 1.34375,
      "reward_std": 0.35786980390548706,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.9375,
      "step": 456
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 490.609375,
      "epoch": 0.3656,
      "grad_norm": 2.3311408484899823,
      "kl": 1.8046875,
      "learning_rate": 4.819004289078217e-06,
      "loss": 0.072,
      "reward": 1.546875,
      "reward_std": 0.3183838725090027,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.890625,
      "step": 457
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 534.90625,
      "epoch": 0.3664,
      "grad_norm": 1.4680518444754158,
      "kl": 1.4609375,
      "learning_rate": 4.818221070772129e-06,
      "loss": 0.0584,
      "reward": 1.40625,
      "reward_std": 0.292504221200943,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.9375,
      "step": 458
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 485.171875,
      "epoch": 0.3672,
      "grad_norm": 1.0154822113166795,
      "kl": 0.39453125,
      "learning_rate": 4.8174362254492555e-06,
      "loss": 0.0159,
      "reward": 1.8125,
      "reward_std": 0.1825428307056427,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 459
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 447.453125,
      "epoch": 0.368,
      "grad_norm": 1.6372782134672146,
      "kl": 1.78125,
      "learning_rate": 4.816649753660431e-06,
      "loss": 0.0712,
      "reward": 1.5,
      "reward_std": 0.5082703232765198,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.921875,
      "step": 460
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 475.1875,
      "epoch": 0.3688,
      "grad_norm": 1.5660653871347994,
      "kl": 0.8828125,
      "learning_rate": 4.815861655957632e-06,
      "loss": 0.0353,
      "reward": 1.671875,
      "reward_std": 0.3624844253063202,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.96875,
      "step": 461
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 489.21875,
      "epoch": 0.3696,
      "grad_norm": 0.9808014696511161,
      "kl": 1.6484375,
      "learning_rate": 4.815071932893976e-06,
      "loss": 0.0661,
      "reward": 1.65625,
      "reward_std": 0.3139738440513611,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.953125,
      "step": 462
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 444.0625,
      "epoch": 0.3704,
      "grad_norm": 0.7216301646582304,
      "kl": 1.2578125,
      "learning_rate": 4.81428058502372e-06,
      "loss": 0.0504,
      "reward": 1.84375,
      "reward_std": 0.3049619495868683,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.953125,
      "step": 463
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 432.796875,
      "epoch": 0.3712,
      "grad_norm": 0.8390030730948433,
      "kl": 0.1484375,
      "learning_rate": 4.813487612902265e-06,
      "loss": 0.0059,
      "reward": 1.671875,
      "reward_std": 0.1893727034330368,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 464
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 476.640625,
      "epoch": 0.372,
      "grad_norm": 0.5245197613735183,
      "kl": 0.1748046875,
      "learning_rate": 4.812693017086145e-06,
      "loss": 0.007,
      "reward": 1.53125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 465
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 448.40625,
      "epoch": 0.3728,
      "grad_norm": 1.5094922067340875,
      "kl": 1.078125,
      "learning_rate": 4.811896798133042e-06,
      "loss": 0.043,
      "reward": 1.71875,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 466
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 519.328125,
      "epoch": 0.3736,
      "grad_norm": 0.9716355528132773,
      "kl": 2.328125,
      "learning_rate": 4.811098956601772e-06,
      "loss": 0.0934,
      "reward": 1.671875,
      "reward_std": 0.44749903678894043,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.890625,
      "step": 467
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 397.421875,
      "epoch": 0.3744,
      "grad_norm": 1.1380947906135268,
      "kl": 0.341796875,
      "learning_rate": 4.810299493052289e-06,
      "loss": 0.0137,
      "reward": 1.46875,
      "reward_std": 0.2756394147872925,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 468
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 388.015625,
      "epoch": 0.3752,
      "grad_norm": 1.6899780677744536,
      "kl": 0.56640625,
      "learning_rate": 4.809498408045691e-06,
      "loss": 0.0227,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 469
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 359.234375,
      "epoch": 0.376,
      "grad_norm": 0.6075789265258572,
      "kl": 0.3515625,
      "learning_rate": 4.808695702144206e-06,
      "loss": 0.0141,
      "reward": 1.515625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 470
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 345.421875,
      "epoch": 0.3768,
      "grad_norm": 0.8841111221529462,
      "kl": 0.1650390625,
      "learning_rate": 4.807891375911207e-06,
      "loss": 0.0066,
      "reward": 1.859375,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 471
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 407.25,
      "epoch": 0.3776,
      "grad_norm": 5.8807950669550415,
      "kl": 0.5625,
      "learning_rate": 4.8070854299112e-06,
      "loss": 0.0225,
      "reward": 1.6875,
      "reward_std": 0.3715250492095947,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.953125,
      "step": 472
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 304.453125,
      "epoch": 0.3784,
      "grad_norm": 1.2945351267840253,
      "kl": 0.48046875,
      "learning_rate": 4.806277864709828e-06,
      "loss": 0.0193,
      "reward": 1.734375,
      "reward_std": 0.24831002950668335,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 473
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 375.546875,
      "epoch": 0.3792,
      "grad_norm": 1.3149418973552802,
      "kl": 1.4453125,
      "learning_rate": 4.805468680873874e-06,
      "loss": 0.0581,
      "reward": 1.375,
      "reward_std": 0.27439430356025696,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.9375,
      "step": 474
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 379.71875,
      "epoch": 0.38,
      "grad_norm": 2.7455582598650285,
      "kl": 1.09375,
      "learning_rate": 4.804657878971252e-06,
      "loss": 0.0439,
      "reward": 1.71875,
      "reward_std": 0.19506090879440308,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.953125,
      "step": 475
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 366.921875,
      "epoch": 0.3808,
      "grad_norm": 3.0891644648564505,
      "kl": 2.015625,
      "learning_rate": 4.803845459571014e-06,
      "loss": 0.0806,
      "reward": 1.6875,
      "reward_std": 0.5475407242774963,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.890625,
      "step": 476
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 445.15625,
      "epoch": 0.3816,
      "grad_norm": 5.880200757491912,
      "kl": 7.6875,
      "learning_rate": 4.803031423243349e-06,
      "loss": 0.309,
      "reward": 1.125,
      "reward_std": 0.6859990358352661,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.671875,
      "step": 477
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 451.234375,
      "epoch": 0.3824,
      "grad_norm": 10.03130773454054,
      "kl": 5.34375,
      "learning_rate": 4.802215770559578e-06,
      "loss": 0.2142,
      "reward": 1.484375,
      "reward_std": 0.4931156635284424,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.8125,
      "step": 478
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 455.046875,
      "epoch": 0.3832,
      "grad_norm": 2.1507203179173766,
      "kl": 2.703125,
      "learning_rate": 4.801398502092156e-06,
      "loss": 0.108,
      "reward": 1.578125,
      "reward_std": 0.3764522075653076,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.84375,
      "step": 479
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 321.890625,
      "epoch": 0.384,
      "grad_norm": 1.021505949210174,
      "kl": 0.318359375,
      "learning_rate": 4.800579618414677e-06,
      "loss": 0.0128,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 480
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 359.53125,
      "epoch": 0.3848,
      "grad_norm": 0.6694409753576069,
      "kl": 0.1943359375,
      "learning_rate": 4.799759120101861e-06,
      "loss": 0.0077,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 481
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 311.375,
      "epoch": 0.3856,
      "grad_norm": 0.4359363974042517,
      "kl": 0.1357421875,
      "learning_rate": 4.798937007729568e-06,
      "loss": 0.0054,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 482
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 350.828125,
      "epoch": 0.3864,
      "grad_norm": 1.5162481213831498,
      "kl": 0.134765625,
      "learning_rate": 4.798113281874788e-06,
      "loss": 0.0054,
      "reward": 1.671875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 483
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 321.4375,
      "epoch": 0.3872,
      "grad_norm": 0.4597374200983244,
      "kl": 0.142578125,
      "learning_rate": 4.797287943115642e-06,
      "loss": 0.0057,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 484
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 396.8125,
      "epoch": 0.388,
      "grad_norm": 0.7607387695800587,
      "kl": 0.1201171875,
      "learning_rate": 4.796460992031386e-06,
      "loss": 0.0048,
      "reward": 1.65625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 485
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 322.0625,
      "epoch": 0.3888,
      "grad_norm": 0.4831613708707655,
      "kl": 0.1416015625,
      "learning_rate": 4.7956324292024045e-06,
      "loss": 0.0057,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 486
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 383.40625,
      "epoch": 0.3896,
      "grad_norm": 0.4081897198615544,
      "kl": 0.12255859375,
      "learning_rate": 4.794802255210217e-06,
      "loss": 0.0049,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 487
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 373.765625,
      "epoch": 0.3904,
      "grad_norm": 0.5795253658984987,
      "kl": 0.1298828125,
      "learning_rate": 4.793970470637469e-06,
      "loss": 0.0052,
      "reward": 1.65625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 488
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 379.296875,
      "epoch": 0.3912,
      "grad_norm": 0.38536641281872774,
      "kl": 0.140625,
      "learning_rate": 4.7931370760679415e-06,
      "loss": 0.0056,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 489
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 389.265625,
      "epoch": 0.392,
      "grad_norm": 0.884095503790402,
      "kl": 0.1240234375,
      "learning_rate": 4.792302072086542e-06,
      "loss": 0.005,
      "reward": 1.46875,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 490
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 387.1875,
      "epoch": 0.3928,
      "grad_norm": 0.9184220362440718,
      "kl": 0.1279296875,
      "learning_rate": 4.7914654592793065e-06,
      "loss": 0.0051,
      "reward": 1.71875,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 491
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 379.71875,
      "epoch": 0.3936,
      "grad_norm": 0.9116583207306123,
      "kl": 0.1298828125,
      "learning_rate": 4.790627238233405e-06,
      "loss": 0.0052,
      "reward": 1.609375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 492
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 410.296875,
      "epoch": 0.3944,
      "grad_norm": 0.42981223850322453,
      "kl": 0.109375,
      "learning_rate": 4.789787409537131e-06,
      "loss": 0.0044,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 493
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 408.34375,
      "epoch": 0.3952,
      "grad_norm": 0.3384378532478862,
      "kl": 0.1240234375,
      "learning_rate": 4.7889459737799105e-06,
      "loss": 0.0049,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 494
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 404.125,
      "epoch": 0.396,
      "grad_norm": 0.9063049215223403,
      "kl": 0.12890625,
      "learning_rate": 4.788102931552294e-06,
      "loss": 0.0052,
      "reward": 1.703125,
      "reward_std": 0.28930896520614624,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 495
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 434.78125,
      "epoch": 0.3968,
      "grad_norm": 0.7924641044749103,
      "kl": 0.11328125,
      "learning_rate": 4.787258283445962e-06,
      "loss": 0.0045,
      "reward": 1.53125,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 496
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 376.15625,
      "epoch": 0.3976,
      "grad_norm": 0.5730328890520717,
      "kl": 0.1318359375,
      "learning_rate": 4.786412030053721e-06,
      "loss": 0.0053,
      "reward": 1.796875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 497
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 433.03125,
      "epoch": 0.3984,
      "grad_norm": 0.4940836808042512,
      "kl": 0.12109375,
      "learning_rate": 4.785564171969503e-06,
      "loss": 0.0048,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 498
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 384.609375,
      "epoch": 0.3992,
      "grad_norm": 1.014219373765608,
      "kl": 0.140625,
      "learning_rate": 4.784714709788368e-06,
      "loss": 0.0056,
      "reward": 1.796875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 499
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 403.859375,
      "epoch": 0.4,
      "grad_norm": 0.543273438917351,
      "kl": 0.12158203125,
      "learning_rate": 4.783863644106502e-06,
      "loss": 0.0049,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 500
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 398.46875,
      "epoch": 0.4008,
      "grad_norm": 0.09114864037290937,
      "kl": 0.12890625,
      "learning_rate": 4.783010975521216e-06,
      "loss": 0.0051,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 501
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 408.96875,
      "epoch": 0.4016,
      "grad_norm": 0.17449315291315534,
      "kl": 0.1328125,
      "learning_rate": 4.782156704630944e-06,
      "loss": 0.0053,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 502
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 433.734375,
      "epoch": 0.4024,
      "grad_norm": 0.5055626056608999,
      "kl": 0.115234375,
      "learning_rate": 4.7813008320352475e-06,
      "loss": 0.0046,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 503
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 364.359375,
      "epoch": 0.4032,
      "grad_norm": 0.7827907023189481,
      "kl": 0.1474609375,
      "learning_rate": 4.78044335833481e-06,
      "loss": 0.0059,
      "reward": 1.59375,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 504
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 426.484375,
      "epoch": 0.404,
      "grad_norm": 0.4589464961908402,
      "kl": 0.115234375,
      "learning_rate": 4.77958428413144e-06,
      "loss": 0.0046,
      "reward": 1.921875,
      "reward_std": 0.13868528604507446,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 505
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 462.15625,
      "epoch": 0.4048,
      "grad_norm": 0.6544346911739349,
      "kl": 0.12353515625,
      "learning_rate": 4.7787236100280685e-06,
      "loss": 0.0049,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 506
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 452.515625,
      "epoch": 0.4056,
      "grad_norm": 1.3105254135159445,
      "kl": 0.1240234375,
      "learning_rate": 4.777861336628751e-06,
      "loss": 0.005,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 507
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 449.171875,
      "epoch": 0.4064,
      "grad_norm": 0.8220089849713779,
      "kl": 0.1416015625,
      "learning_rate": 4.7769974645386616e-06,
      "loss": 0.0057,
      "reward": 1.5,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 508
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 432.546875,
      "epoch": 0.4072,
      "grad_norm": 0.7873270142931792,
      "kl": 0.140625,
      "learning_rate": 4.776131994364102e-06,
      "loss": 0.0056,
      "reward": 1.671875,
      "reward_std": 0.1983242630958557,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 509
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 452.078125,
      "epoch": 0.408,
      "grad_norm": 1.2832406872401583,
      "kl": 0.12890625,
      "learning_rate": 4.775264926712489e-06,
      "loss": 0.0052,
      "reward": 1.75,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 510
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 437.03125,
      "epoch": 0.4088,
      "grad_norm": 0.5124697706361917,
      "kl": 0.12451171875,
      "learning_rate": 4.774396262192368e-06,
      "loss": 0.005,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 511
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 488.8125,
      "epoch": 0.4096,
      "grad_norm": 0.4250072391435047,
      "kl": 0.1201171875,
      "learning_rate": 4.7735260014133986e-06,
      "loss": 0.0048,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 512
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 417.34375,
      "epoch": 0.4104,
      "grad_norm": 0.40503428079574494,
      "kl": 0.14453125,
      "learning_rate": 4.772654144986364e-06,
      "loss": 0.0058,
      "reward": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 513
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 462.703125,
      "epoch": 0.4112,
      "grad_norm": 0.7881543976365165,
      "kl": 0.1279296875,
      "learning_rate": 4.7717806935231665e-06,
      "loss": 0.0051,
      "reward": 1.8125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 514
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 424.65625,
      "epoch": 0.412,
      "grad_norm": 0.5766837892479305,
      "kl": 0.154296875,
      "learning_rate": 4.770905647636828e-06,
      "loss": 0.0062,
      "reward": 1.53125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 515
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 419.375,
      "epoch": 0.4128,
      "grad_norm": 0.44407401013076075,
      "kl": 0.138671875,
      "learning_rate": 4.77002900794149e-06,
      "loss": 0.0055,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 516
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 470.59375,
      "epoch": 0.4136,
      "grad_norm": 0.5235926160400373,
      "kl": 0.12890625,
      "learning_rate": 4.769150775052411e-06,
      "loss": 0.0051,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 517
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 444.578125,
      "epoch": 0.4144,
      "grad_norm": 0.9967752126166552,
      "kl": 0.1279296875,
      "learning_rate": 4.768270949585968e-06,
      "loss": 0.0051,
      "reward": 1.78125,
      "reward_std": 0.23356689512729645,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 518
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 410.09375,
      "epoch": 0.4152,
      "grad_norm": 0.8675042446430157,
      "kl": 0.1376953125,
      "learning_rate": 4.767389532159659e-06,
      "loss": 0.0055,
      "reward": 1.5,
      "reward_std": 0.2630178928375244,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 519
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 396.46875,
      "epoch": 0.416,
      "grad_norm": 0.8224038768492201,
      "kl": 0.1494140625,
      "learning_rate": 4.766506523392095e-06,
      "loss": 0.006,
      "reward": 1.359375,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 520
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 448.890625,
      "epoch": 0.4168,
      "grad_norm": 0.5425410755487962,
      "kl": 0.125,
      "learning_rate": 4.765621923903005e-06,
      "loss": 0.005,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 521
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 486.34375,
      "epoch": 0.4176,
      "grad_norm": 1.128073055749155,
      "kl": 0.130859375,
      "learning_rate": 4.764735734313236e-06,
      "loss": 0.0052,
      "reward": 1.75,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 522
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 426.1875,
      "epoch": 0.4184,
      "grad_norm": 0.4031109956060146,
      "kl": 0.11669921875,
      "learning_rate": 4.763847955244749e-06,
      "loss": 0.0047,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 523
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 480.203125,
      "epoch": 0.4192,
      "grad_norm": 0.90433099369881,
      "kl": 0.11474609375,
      "learning_rate": 4.762958587320623e-06,
      "loss": 0.0046,
      "reward": 1.734375,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 524
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 388.640625,
      "epoch": 0.42,
      "grad_norm": 0.6687933553064049,
      "kl": 0.1865234375,
      "learning_rate": 4.762067631165049e-06,
      "loss": 0.0075,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 525
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 445.078125,
      "epoch": 0.4208,
      "grad_norm": 0.6786375875150067,
      "kl": 0.1181640625,
      "learning_rate": 4.761175087403336e-06,
      "loss": 0.0047,
      "reward": 1.84375,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 526
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 438.96875,
      "epoch": 0.4216,
      "grad_norm": 0.7303677491642302,
      "kl": 0.1240234375,
      "learning_rate": 4.760280956661904e-06,
      "loss": 0.005,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 527
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 416.53125,
      "epoch": 0.4224,
      "grad_norm": 0.7593473495539002,
      "kl": 0.1318359375,
      "learning_rate": 4.75938523956829e-06,
      "loss": 0.0053,
      "reward": 1.703125,
      "reward_std": 0.2519446909427643,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 528
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 410.75,
      "epoch": 0.4232,
      "grad_norm": 0.330189456706158,
      "kl": 0.1298828125,
      "learning_rate": 4.75848793675114e-06,
      "loss": 0.0052,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 529
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 401.1875,
      "epoch": 0.424,
      "grad_norm": 0.6018398606145906,
      "kl": 0.1396484375,
      "learning_rate": 4.757589048840219e-06,
      "loss": 0.0056,
      "reward": 1.78125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 530
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 456.4375,
      "epoch": 0.4248,
      "grad_norm": 0.9966756944312201,
      "kl": 0.142578125,
      "learning_rate": 4.756688576466398e-06,
      "loss": 0.0057,
      "reward": 1.75,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 531
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 410.59375,
      "epoch": 0.4256,
      "grad_norm": 0.5972720922208659,
      "kl": 0.1328125,
      "learning_rate": 4.755786520261666e-06,
      "loss": 0.0053,
      "reward": 1.765625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 532
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 464.234375,
      "epoch": 0.4264,
      "grad_norm": 0.9903609430647151,
      "kl": 0.15234375,
      "learning_rate": 4.75488288085912e-06,
      "loss": 0.0061,
      "reward": 1.890625,
      "reward_std": 0.22707363963127136,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.953125,
      "step": 533
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 502.03125,
      "epoch": 0.4272,
      "grad_norm": 0.7985987687798536,
      "kl": 0.146484375,
      "learning_rate": 4.753977658892967e-06,
      "loss": 0.0058,
      "reward": 1.65625,
      "reward_std": 0.2404065728187561,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 534
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 454.25,
      "epoch": 0.428,
      "grad_norm": 0.7620615509609708,
      "kl": 0.13671875,
      "learning_rate": 4.753070854998529e-06,
      "loss": 0.0055,
      "reward": 1.6875,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 535
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 466.875,
      "epoch": 0.4288,
      "grad_norm": 1.2767021083048151,
      "kl": 0.134765625,
      "learning_rate": 4.752162469812234e-06,
      "loss": 0.0054,
      "reward": 1.703125,
      "reward_std": 0.27883461117744446,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.9375,
      "step": 536
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 382.40625,
      "epoch": 0.4296,
      "grad_norm": 0.6046278635344334,
      "kl": 0.1806640625,
      "learning_rate": 4.751252503971624e-06,
      "loss": 0.0072,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 537
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 485.984375,
      "epoch": 0.4304,
      "grad_norm": 1.0901821078716176,
      "kl": 0.19140625,
      "learning_rate": 4.750340958115346e-06,
      "loss": 0.0077,
      "reward": 1.5,
      "reward_std": 0.2798827290534973,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.953125,
      "step": 538
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 429.78125,
      "epoch": 0.4312,
      "grad_norm": 1.1995065520907704,
      "kl": 0.220703125,
      "learning_rate": 4.749427832883158e-06,
      "loss": 0.0088,
      "reward": 1.796875,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 539
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 443.078125,
      "epoch": 0.432,
      "grad_norm": 1.9491122287555656,
      "kl": 0.484375,
      "learning_rate": 4.748513128915928e-06,
      "loss": 0.0194,
      "reward": 1.609375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.921875,
      "step": 540
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 474.0625,
      "epoch": 0.4328,
      "grad_norm": 3.1890070394788874,
      "kl": 0.6171875,
      "learning_rate": 4.747596846855629e-06,
      "loss": 0.0246,
      "reward": 1.59375,
      "reward_std": 0.26409149169921875,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.9375,
      "step": 541
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 486.125,
      "epoch": 0.4336,
      "grad_norm": 2.0043216789347746,
      "kl": 0.60546875,
      "learning_rate": 4.7466789873453446e-06,
      "loss": 0.0242,
      "reward": 1.53125,
      "reward_std": 0.3412526845932007,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.953125,
      "step": 542
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 490.640625,
      "epoch": 0.4344,
      "grad_norm": 2.580392206072138,
      "kl": 2.140625,
      "learning_rate": 4.7457595510292615e-06,
      "loss": 0.0856,
      "reward": 1.671875,
      "reward_std": 0.4963078498840332,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.859375,
      "step": 543
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 398.0625,
      "epoch": 0.4352,
      "grad_norm": 2.597169689699224,
      "kl": 1.1328125,
      "learning_rate": 4.744838538552678e-06,
      "loss": 0.0454,
      "reward": 1.625,
      "reward_std": 0.22558549046516418,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.921875,
      "step": 544
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 391.46875,
      "epoch": 0.436,
      "grad_norm": 1.0571142153194588,
      "kl": 1.546875,
      "learning_rate": 4.7439159505619946e-06,
      "loss": 0.0618,
      "reward": 1.890625,
      "reward_std": 0.20290401577949524,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.953125,
      "step": 545
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 407.71875,
      "epoch": 0.4368,
      "grad_norm": 2.0231863556745306,
      "kl": 1.6875,
      "learning_rate": 4.74299178770472e-06,
      "loss": 0.0675,
      "reward": 1.65625,
      "reward_std": 0.3596561551094055,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.921875,
      "step": 546
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 424.0625,
      "epoch": 0.4376,
      "grad_norm": 1.5551994849968498,
      "kl": 3.046875,
      "learning_rate": 4.742066050629465e-06,
      "loss": 0.1219,
      "reward": 1.484375,
      "reward_std": 0.39774754643440247,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.921875,
      "step": 547
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 360.84375,
      "epoch": 0.4384,
      "grad_norm": 3.919915861139782,
      "kl": 3.1875,
      "learning_rate": 4.741138739985951e-06,
      "loss": 0.1278,
      "reward": 1.421875,
      "reward_std": 0.45242589712142944,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.90625,
      "step": 548
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 389.46875,
      "epoch": 0.4392,
      "grad_norm": 23.12505474924314,
      "kl": 14.625,
      "learning_rate": 4.740209856424998e-06,
      "loss": 0.5854,
      "reward": 1.21875,
      "reward_std": 0.712020993232727,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.578125,
      "step": 549
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 339.640625,
      "epoch": 0.44,
      "grad_norm": 1.9920077078374177,
      "kl": 2.578125,
      "learning_rate": 4.7392794005985324e-06,
      "loss": 0.1031,
      "reward": 1.59375,
      "reward_std": 0.4238043427467346,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.90625,
      "step": 550
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 404.171875,
      "epoch": 0.4408,
      "grad_norm": 2.8098958853055827,
      "kl": 4.75,
      "learning_rate": 4.738347373159585e-06,
      "loss": 0.1898,
      "reward": 1.578125,
      "reward_std": 0.5108742117881775,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.828125,
      "step": 551
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 311.0625,
      "epoch": 0.4416,
      "grad_norm": 1.752527803564222,
      "kl": 0.435546875,
      "learning_rate": 4.737413774762287e-06,
      "loss": 0.0175,
      "reward": 1.71875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 552
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 324.328125,
      "epoch": 0.4424,
      "grad_norm": 0.623290617592855,
      "kl": 0.1298828125,
      "learning_rate": 4.736478606061876e-06,
      "loss": 0.0052,
      "reward": 1.640625,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 553
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 382.984375,
      "epoch": 0.4432,
      "grad_norm": 0.9636215109224563,
      "kl": 0.373046875,
      "learning_rate": 4.735541867714687e-06,
      "loss": 0.0149,
      "reward": 1.65625,
      "reward_std": 0.21917018294334412,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 554
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 372.0625,
      "epoch": 0.444,
      "grad_norm": 0.4643488621551407,
      "kl": 0.1220703125,
      "learning_rate": 4.73460356037816e-06,
      "loss": 0.0049,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 555
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 349.78125,
      "epoch": 0.4448,
      "grad_norm": 0.4106290847161274,
      "kl": 0.10888671875,
      "learning_rate": 4.733663684710835e-06,
      "loss": 0.0043,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 556
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 357.859375,
      "epoch": 0.4456,
      "grad_norm": 0.8089250709426993,
      "kl": 0.12109375,
      "learning_rate": 4.732722241372354e-06,
      "loss": 0.0048,
      "reward": 1.6875,
      "reward_std": 0.17570313811302185,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 557
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 363.03125,
      "epoch": 0.4464,
      "grad_norm": 0.46368035433701,
      "kl": 0.10498046875,
      "learning_rate": 4.731779231023456e-06,
      "loss": 0.0042,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 558
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 301.84375,
      "epoch": 0.4472,
      "grad_norm": 0.9322080245234828,
      "kl": 0.13671875,
      "learning_rate": 4.730834654325984e-06,
      "loss": 0.0055,
      "reward": 1.671875,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 559
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 358.171875,
      "epoch": 0.448,
      "grad_norm": 0.39997720342036575,
      "kl": 0.125,
      "learning_rate": 4.729888511942877e-06,
      "loss": 0.005,
      "reward": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 560
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 287.71875,
      "epoch": 0.4488,
      "grad_norm": 0.9548420650515719,
      "kl": 0.126953125,
      "learning_rate": 4.728940804538176e-06,
      "loss": 0.0051,
      "reward": 1.625,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 561
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 361.078125,
      "epoch": 0.4496,
      "grad_norm": 3.617701918943756,
      "kl": 0.1240234375,
      "learning_rate": 4.727991532777016e-06,
      "loss": 0.005,
      "reward": 1.8125,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 562
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 333.625,
      "epoch": 0.4504,
      "grad_norm": 0.7436500355949118,
      "kl": 0.130859375,
      "learning_rate": 4.727040697325634e-06,
      "loss": 0.0052,
      "reward": 1.59375,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 563
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 359.921875,
      "epoch": 0.4512,
      "grad_norm": 0.8053192115050734,
      "kl": 0.12109375,
      "learning_rate": 4.726088298851362e-06,
      "loss": 0.0049,
      "reward": 1.59375,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 564
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 376.40625,
      "epoch": 0.452,
      "grad_norm": 0.8159629873340042,
      "kl": 0.1123046875,
      "learning_rate": 4.725134338022631e-06,
      "loss": 0.0045,
      "reward": 1.65625,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 565
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 322.578125,
      "epoch": 0.4528,
      "grad_norm": 4.563543622614359,
      "kl": 0.134765625,
      "learning_rate": 4.724178815508967e-06,
      "loss": 0.0054,
      "reward": 1.6875,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 566
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 368.265625,
      "epoch": 0.4536,
      "grad_norm": 0.3786388782574309,
      "kl": 0.1171875,
      "learning_rate": 4.723221731980993e-06,
      "loss": 0.0047,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 567
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 357.828125,
      "epoch": 0.4544,
      "grad_norm": 0.8553088933063837,
      "kl": 0.1201171875,
      "learning_rate": 4.722263088110426e-06,
      "loss": 0.0048,
      "reward": 1.53125,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 568
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 368.6875,
      "epoch": 0.4552,
      "grad_norm": 0.4770339948697236,
      "kl": 0.130859375,
      "learning_rate": 4.721302884570079e-06,
      "loss": 0.0052,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 569
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 352.921875,
      "epoch": 0.456,
      "grad_norm": 1.0804985777028955,
      "kl": 0.1171875,
      "learning_rate": 4.720341122033862e-06,
      "loss": 0.0047,
      "reward": 1.78125,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 570
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 386.78125,
      "epoch": 0.4568,
      "grad_norm": 0.35084366597904426,
      "kl": 0.1162109375,
      "learning_rate": 4.719377801176774e-06,
      "loss": 0.0046,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 571
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 327.75,
      "epoch": 0.4576,
      "grad_norm": 0.4010624118550041,
      "kl": 0.140625,
      "learning_rate": 4.718412922674913e-06,
      "loss": 0.0056,
      "reward": 1.59375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 572
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 361.28125,
      "epoch": 0.4584,
      "grad_norm": 0.5170960923198824,
      "kl": 0.1298828125,
      "learning_rate": 4.717446487205466e-06,
      "loss": 0.0052,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 573
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 383.25,
      "epoch": 0.4592,
      "grad_norm": 1.9945218987513769,
      "kl": 0.15234375,
      "learning_rate": 4.716478495446717e-06,
      "loss": 0.0061,
      "reward": 1.828125,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 574
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 410.40625,
      "epoch": 0.46,
      "grad_norm": 2.895751664424082,
      "kl": 0.19140625,
      "learning_rate": 4.715508948078037e-06,
      "loss": 0.0076,
      "reward": 1.546875,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.9375,
      "step": 575
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 376.1875,
      "epoch": 0.4608,
      "grad_norm": 1.638065505747679,
      "kl": 0.400390625,
      "learning_rate": 4.714537845779894e-06,
      "loss": 0.0161,
      "reward": 1.890625,
      "reward_std": 0.19654905796051025,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.921875,
      "step": 576
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 349.734375,
      "epoch": 0.4616,
      "grad_norm": 2.5247233544787684,
      "kl": 0.54296875,
      "learning_rate": 4.7135651892338445e-06,
      "loss": 0.0217,
      "reward": 1.609375,
      "reward_std": 0.38194066286087036,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.921875,
      "step": 577
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.5,
      "completion_length": 338.265625,
      "epoch": 0.4624,
      "grad_norm": 2.7984982639925433,
      "kl": 0.75390625,
      "learning_rate": 4.712590979122534e-06,
      "loss": 0.0301,
      "reward": 1.328125,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.90625,
      "step": 578
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 381.15625,
      "epoch": 0.4632,
      "grad_norm": 2.735569317045563,
      "kl": 1.375,
      "learning_rate": 4.7116152161297045e-06,
      "loss": 0.055,
      "reward": 1.390625,
      "reward_std": 0.39916694164276123,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.875,
      "step": 579
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 442.53125,
      "epoch": 0.464,
      "grad_norm": 3.1388133340699715,
      "kl": 4.5625,
      "learning_rate": 4.710637900940181e-06,
      "loss": 0.1821,
      "reward": 1.25,
      "reward_std": 0.6596025228500366,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.734375,
      "step": 580
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 365.3125,
      "epoch": 0.4648,
      "grad_norm": 2.807083376416559,
      "kl": 2.328125,
      "learning_rate": 4.7096590342398825e-06,
      "loss": 0.0932,
      "reward": 1.703125,
      "reward_std": 0.44655635952949524,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.921875,
      "step": 581
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 387.5,
      "epoch": 0.4656,
      "grad_norm": 2.8132496897579014,
      "kl": 4.125,
      "learning_rate": 4.708678616715815e-06,
      "loss": 0.1651,
      "reward": 1.578125,
      "reward_std": 0.4879796802997589,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.875,
      "step": 582
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 344.5,
      "epoch": 0.4664,
      "grad_norm": 1.8018781601902947,
      "kl": 1.046875,
      "learning_rate": 4.707696649056073e-06,
      "loss": 0.0419,
      "reward": 1.5,
      "reward_std": 0.3648670017719269,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 583
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 373.484375,
      "epoch": 0.4672,
      "grad_norm": 0.760952215794593,
      "kl": 1.140625,
      "learning_rate": 4.706713131949839e-06,
      "loss": 0.0456,
      "reward": 1.796875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 584
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 365.390625,
      "epoch": 0.468,
      "grad_norm": 0.58701159191976,
      "kl": 0.69921875,
      "learning_rate": 4.705728066087384e-06,
      "loss": 0.028,
      "reward": 1.765625,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 585
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 359.875,
      "epoch": 0.4688,
      "grad_norm": 0.5590360711726854,
      "kl": 0.87109375,
      "learning_rate": 4.704741452160064e-06,
      "loss": 0.0347,
      "reward": 1.375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 0.984375,
      "step": 586
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 336.140625,
      "epoch": 0.4696,
      "grad_norm": 0.37962287263134975,
      "kl": 0.13671875,
      "learning_rate": 4.703753290860323e-06,
      "loss": 0.0055,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 587
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 321.515625,
      "epoch": 0.4704,
      "grad_norm": 0.7498115796865782,
      "kl": 0.77734375,
      "learning_rate": 4.702763582881692e-06,
      "loss": 0.0311,
      "reward": 1.859375,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 588
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 359.125,
      "epoch": 0.4712,
      "grad_norm": 1.149421698917399,
      "kl": 1.75,
      "learning_rate": 4.701772328918784e-06,
      "loss": 0.07,
      "reward": 1.78125,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 589
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 365.5625,
      "epoch": 0.472,
      "grad_norm": 0.4679289139806567,
      "kl": 0.1318359375,
      "learning_rate": 4.700779529667301e-06,
      "loss": 0.0053,
      "reward": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 590
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 341.84375,
      "epoch": 0.4728,
      "grad_norm": 1.4087039371104781,
      "kl": 0.12890625,
      "learning_rate": 4.699785185824026e-06,
      "loss": 0.0052,
      "reward": 1.53125,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 591
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 348.71875,
      "epoch": 0.4736,
      "grad_norm": 0.7139348028378941,
      "kl": 0.1474609375,
      "learning_rate": 4.69878929808683e-06,
      "loss": 0.0059,
      "reward": 1.8125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 592
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 348.796875,
      "epoch": 0.4744,
      "grad_norm": 0.09219704742523589,
      "kl": 0.11962890625,
      "learning_rate": 4.6977918671546635e-06,
      "loss": 0.0048,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 593
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 332.84375,
      "epoch": 0.4752,
      "grad_norm": 0.45624501772723347,
      "kl": 0.1435546875,
      "learning_rate": 4.696792893727562e-06,
      "loss": 0.0057,
      "reward": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 594
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.5,
      "completion_length": 304.5,
      "epoch": 0.476,
      "grad_norm": 0.9202164978785126,
      "kl": 0.1650390625,
      "learning_rate": 4.695792378506645e-06,
      "loss": 0.0066,
      "reward": 1.421875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 595
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 332.828125,
      "epoch": 0.4768,
      "grad_norm": 0.6614322723537502,
      "kl": 0.140625,
      "learning_rate": 4.694790322194111e-06,
      "loss": 0.0056,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 596
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 394.515625,
      "epoch": 0.4776,
      "grad_norm": 0.2163278343277934,
      "kl": 0.123046875,
      "learning_rate": 4.693786725493242e-06,
      "loss": 0.0049,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 597
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 353.6875,
      "epoch": 0.4784,
      "grad_norm": 1.0025384969215045,
      "kl": 0.146484375,
      "learning_rate": 4.692781589108402e-06,
      "loss": 0.0059,
      "reward": 1.6875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 598
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 304.203125,
      "epoch": 0.4792,
      "grad_norm": 0.8555086375496709,
      "kl": 0.1474609375,
      "learning_rate": 4.691774913745033e-06,
      "loss": 0.0059,
      "reward": 1.59375,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 599
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 388.4375,
      "epoch": 0.48,
      "grad_norm": 0.6762760084546305,
      "kl": 0.12890625,
      "learning_rate": 4.690766700109659e-06,
      "loss": 0.0052,
      "reward": 1.71875,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 600
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 323.890625,
      "epoch": 0.4808,
      "grad_norm": 0.8763816444356921,
      "kl": 0.146484375,
      "learning_rate": 4.689756948909884e-06,
      "loss": 0.0059,
      "reward": 1.65625,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 601
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 351.65625,
      "epoch": 0.4816,
      "grad_norm": 0.8698407297285081,
      "kl": 0.158203125,
      "learning_rate": 4.688745660854388e-06,
      "loss": 0.0063,
      "reward": 1.515625,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 602
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 359.6875,
      "epoch": 0.4824,
      "grad_norm": 0.7746532613310444,
      "kl": 0.1435546875,
      "learning_rate": 4.687732836652935e-06,
      "loss": 0.0058,
      "reward": 1.875,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.984375,
      "step": 603
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 342.328125,
      "epoch": 0.4832,
      "grad_norm": 0.9476309548652774,
      "kl": 0.138671875,
      "learning_rate": 4.686718477016361e-06,
      "loss": 0.0056,
      "reward": 1.625,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 604
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 372.625,
      "epoch": 0.484,
      "grad_norm": 0.6907800503076698,
      "kl": 0.1416015625,
      "learning_rate": 4.6857025826565845e-06,
      "loss": 0.0056,
      "reward": 1.390625,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 605
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 333.875,
      "epoch": 0.4848,
      "grad_norm": 0.7036792594591179,
      "kl": 0.1474609375,
      "learning_rate": 4.684685154286599e-06,
      "loss": 0.0059,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 606
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 334.3125,
      "epoch": 0.4856,
      "grad_norm": 0.7784029972183542,
      "kl": 0.150390625,
      "learning_rate": 4.683666192620474e-06,
      "loss": 0.006,
      "reward": 1.640625,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 607
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 381.234375,
      "epoch": 0.4864,
      "grad_norm": 0.3276931766459097,
      "kl": 0.1337890625,
      "learning_rate": 4.682645698373357e-06,
      "loss": 0.0053,
      "reward": 1.390625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 608
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 385.4375,
      "epoch": 0.4872,
      "grad_norm": 1.0244585337473353,
      "kl": 0.15234375,
      "learning_rate": 4.6816236722614694e-06,
      "loss": 0.0061,
      "reward": 1.578125,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 609
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 384.28125,
      "epoch": 0.488,
      "grad_norm": 0.6623725278278989,
      "kl": 0.1298828125,
      "learning_rate": 4.680600115002109e-06,
      "loss": 0.0052,
      "reward": 1.734375,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 610
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 410.140625,
      "epoch": 0.4888,
      "grad_norm": 1.1182525515241812,
      "kl": 0.345703125,
      "learning_rate": 4.679575027313649e-06,
      "loss": 0.0139,
      "reward": 1.75,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 611
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 402.421875,
      "epoch": 0.4896,
      "grad_norm": 1.1473152123947696,
      "kl": 0.12890625,
      "learning_rate": 4.6785484099155324e-06,
      "loss": 0.0052,
      "reward": 1.765625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 612
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 403.703125,
      "epoch": 0.4904,
      "grad_norm": 0.37850839844547285,
      "kl": 0.1484375,
      "learning_rate": 4.67752026352828e-06,
      "loss": 0.0059,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 613
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 431.734375,
      "epoch": 0.4912,
      "grad_norm": 0.680113184405759,
      "kl": 0.1357421875,
      "learning_rate": 4.676490588873486e-06,
      "loss": 0.0054,
      "reward": 1.765625,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 614
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 463.28125,
      "epoch": 0.492,
      "grad_norm": 0.9847351909602291,
      "kl": 0.13671875,
      "learning_rate": 4.675459386673815e-06,
      "loss": 0.0055,
      "reward": 1.625,
      "reward_std": 0.2756393849849701,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 615
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 466.578125,
      "epoch": 0.4928,
      "grad_norm": 0.45510120379863056,
      "kl": 0.150390625,
      "learning_rate": 4.674426657653003e-06,
      "loss": 0.006,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 616
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 347.984375,
      "epoch": 0.4936,
      "grad_norm": 0.7723889384366126,
      "kl": 0.1396484375,
      "learning_rate": 4.67339240253586e-06,
      "loss": 0.0056,
      "reward": 1.453125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 617
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 421.75,
      "epoch": 0.4944,
      "grad_norm": 0.537734608716655,
      "kl": 0.140625,
      "learning_rate": 4.672356622048266e-06,
      "loss": 0.0056,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 618
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 356.71875,
      "epoch": 0.4952,
      "grad_norm": 0.671441619255682,
      "kl": 0.140625,
      "learning_rate": 4.671319316917172e-06,
      "loss": 0.0056,
      "reward": 1.46875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 619
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 395.765625,
      "epoch": 0.496,
      "grad_norm": 0.7721416598091413,
      "kl": 0.7734375,
      "learning_rate": 4.670280487870599e-06,
      "loss": 0.0308,
      "reward": 1.6875,
      "reward_std": 0.21556037664413452,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 620
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 390.46875,
      "epoch": 0.4968,
      "grad_norm": 0.9680757407929496,
      "kl": 0.169921875,
      "learning_rate": 4.669240135637635e-06,
      "loss": 0.0068,
      "reward": 1.53125,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 621
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 378.234375,
      "epoch": 0.4976,
      "grad_norm": 0.4582522192089543,
      "kl": 0.1396484375,
      "learning_rate": 4.668198260948442e-06,
      "loss": 0.0056,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 622
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 448.0625,
      "epoch": 0.4984,
      "grad_norm": 2.4735346905444797,
      "kl": 2.546875,
      "learning_rate": 4.667154864534245e-06,
      "loss": 0.1017,
      "reward": 1.625,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.9375,
      "step": 623
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 453.34375,
      "epoch": 0.4992,
      "grad_norm": 2.3522533719660346,
      "kl": 1.1328125,
      "learning_rate": 4.666109947127343e-06,
      "loss": 0.0452,
      "reward": 1.875,
      "reward_std": 0.1825428307056427,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 624
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 371.96875,
      "epoch": 0.5,
      "grad_norm": 0.67998421474587,
      "kl": 0.134765625,
      "learning_rate": 4.665063509461098e-06,
      "loss": 0.0054,
      "reward": 1.53125,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 625
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 399.28125,
      "epoch": 0.5008,
      "grad_norm": 0.6331284378912473,
      "kl": 1.7265625,
      "learning_rate": 4.664015552269938e-06,
      "loss": 0.069,
      "reward": 1.5625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.96875,
      "step": 626
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 433.71875,
      "epoch": 0.5016,
      "grad_norm": 0.9447568979983324,
      "kl": 0.6015625,
      "learning_rate": 4.662966076289363e-06,
      "loss": 0.0241,
      "reward": 1.9375,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 627
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 408.09375,
      "epoch": 0.5024,
      "grad_norm": 0.448770492187416,
      "kl": 0.1298828125,
      "learning_rate": 4.661915082255932e-06,
      "loss": 0.0052,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 628
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 372.140625,
      "epoch": 0.5032,
      "grad_norm": 1.0622563594989127,
      "kl": 0.150390625,
      "learning_rate": 4.6608625709072766e-06,
      "loss": 0.006,
      "reward": 1.515625,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 629
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 430.5625,
      "epoch": 0.504,
      "grad_norm": 0.7294440628390886,
      "kl": 0.1328125,
      "learning_rate": 4.659808542982089e-06,
      "loss": 0.0053,
      "reward": 1.65625,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 630
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 394.90625,
      "epoch": 0.5048,
      "grad_norm": 0.7076799509181766,
      "kl": 0.890625,
      "learning_rate": 4.658752999220125e-06,
      "loss": 0.0356,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 631
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 434.0,
      "epoch": 0.5056,
      "grad_norm": 0.7658790385590288,
      "kl": 0.251953125,
      "learning_rate": 4.657695940362207e-06,
      "loss": 0.0101,
      "reward": 1.625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 632
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 394.765625,
      "epoch": 0.5064,
      "grad_norm": 0.8367046057258256,
      "kl": 0.66796875,
      "learning_rate": 4.65663736715022e-06,
      "loss": 0.0268,
      "reward": 1.421875,
      "reward_std": 0.3054368495941162,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.984375,
      "step": 633
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 363.171875,
      "epoch": 0.5072,
      "grad_norm": 0.7394620159296568,
      "kl": 0.21875,
      "learning_rate": 4.65557728032711e-06,
      "loss": 0.0087,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 634
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 438.46875,
      "epoch": 0.508,
      "grad_norm": 0.9007138649455658,
      "kl": 0.388671875,
      "learning_rate": 4.654515680636888e-06,
      "loss": 0.0155,
      "reward": 1.9375,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 635
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 407.859375,
      "epoch": 0.5088,
      "grad_norm": 0.8812664025932062,
      "kl": 1.3046875,
      "learning_rate": 4.653452568824625e-06,
      "loss": 0.0525,
      "reward": 1.78125,
      "reward_std": 0.3319803476333618,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 636
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.625,
      "completion_length": 371.84375,
      "epoch": 0.5096,
      "grad_norm": 1.9274762594921737,
      "kl": 0.494140625,
      "learning_rate": 4.652387945636454e-06,
      "loss": 0.0198,
      "reward": 1.28125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 0.984375,
      "step": 637
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 475.71875,
      "epoch": 0.5104,
      "grad_norm": 0.5049969313742863,
      "kl": 0.12158203125,
      "learning_rate": 4.651321811819568e-06,
      "loss": 0.0049,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 638
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 363.796875,
      "epoch": 0.5112,
      "grad_norm": 0.683187960721247,
      "kl": 0.1357421875,
      "learning_rate": 4.650254168122222e-06,
      "loss": 0.0054,
      "reward": 1.78125,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 639
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 468.65625,
      "epoch": 0.512,
      "grad_norm": 2.3577243207279666,
      "kl": 0.828125,
      "learning_rate": 4.649185015293728e-06,
      "loss": 0.0331,
      "reward": 1.53125,
      "reward_std": 0.22451192140579224,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 640
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 389.734375,
      "epoch": 0.5128,
      "grad_norm": 0.8012985914421464,
      "kl": 1.2109375,
      "learning_rate": 4.64811435408446e-06,
      "loss": 0.0483,
      "reward": 1.6875,
      "reward_std": 0.22558549046516418,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 641
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 449.625,
      "epoch": 0.5136,
      "grad_norm": 1.2406518104534963,
      "kl": 1.65625,
      "learning_rate": 4.647042185245848e-06,
      "loss": 0.0664,
      "reward": 1.640625,
      "reward_std": 0.22707363963127136,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.953125,
      "step": 642
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 409.09375,
      "epoch": 0.5144,
      "grad_norm": 0.7995409575096425,
      "kl": 0.734375,
      "learning_rate": 4.645968509530381e-06,
      "loss": 0.0293,
      "reward": 1.5625,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 643
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 405.140625,
      "epoch": 0.5152,
      "grad_norm": 0.9728174896744552,
      "kl": 0.123046875,
      "learning_rate": 4.644893327691608e-06,
      "loss": 0.0049,
      "reward": 1.71875,
      "reward_std": 0.3208816647529602,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 644
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 355.171875,
      "epoch": 0.516,
      "grad_norm": 1.2186157898665686,
      "kl": 1.5078125,
      "learning_rate": 4.6438166404841316e-06,
      "loss": 0.0603,
      "reward": 1.53125,
      "reward_std": 0.47431010007858276,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.96875,
      "step": 645
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 469.75,
      "epoch": 0.5168,
      "grad_norm": 1.2018790641173411,
      "kl": 0.86328125,
      "learning_rate": 4.6427384486636115e-06,
      "loss": 0.0347,
      "reward": 1.640625,
      "reward_std": 0.3187600076198578,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.953125,
      "step": 646
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 415.734375,
      "epoch": 0.5176,
      "grad_norm": 0.5962496013183518,
      "kl": 0.11181640625,
      "learning_rate": 4.6416587529867665e-06,
      "loss": 0.0045,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 647
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 362.5625,
      "epoch": 0.5184,
      "grad_norm": 0.5950301304457719,
      "kl": 0.1083984375,
      "learning_rate": 4.640577554211366e-06,
      "loss": 0.0043,
      "reward": 1.875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 648
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 420.78125,
      "epoch": 0.5192,
      "grad_norm": 0.7874157693561777,
      "kl": 0.1103515625,
      "learning_rate": 4.63949485309624e-06,
      "loss": 0.0044,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 649
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 355.5,
      "epoch": 0.52,
      "grad_norm": 1.5109255828936672,
      "kl": 0.1181640625,
      "learning_rate": 4.638410650401267e-06,
      "loss": 0.0047,
      "reward": 1.5,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 650
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 452.46875,
      "epoch": 0.5208,
      "grad_norm": 0.6834046639715636,
      "kl": 0.11767578125,
      "learning_rate": 4.637324946887384e-06,
      "loss": 0.0047,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 651
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 348.1875,
      "epoch": 0.5216,
      "grad_norm": 3.4529482216950655,
      "kl": 0.87109375,
      "learning_rate": 4.636237743316578e-06,
      "loss": 0.0348,
      "reward": 1.609375,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 652
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 424.84375,
      "epoch": 0.5224,
      "grad_norm": 0.731927864044106,
      "kl": 0.125,
      "learning_rate": 4.635149040451891e-06,
      "loss": 0.005,
      "reward": 1.453125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 653
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 418.328125,
      "epoch": 0.5232,
      "grad_norm": 0.7038942934042214,
      "kl": 0.11328125,
      "learning_rate": 4.634058839057417e-06,
      "loss": 0.0045,
      "reward": 1.640625,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 654
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 391.9375,
      "epoch": 0.524,
      "grad_norm": 0.7724484209864683,
      "kl": 0.1103515625,
      "learning_rate": 4.632967139898301e-06,
      "loss": 0.0044,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 655
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 419.609375,
      "epoch": 0.5248,
      "grad_norm": 0.7184964328574942,
      "kl": 0.11767578125,
      "learning_rate": 4.63187394374074e-06,
      "loss": 0.0047,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 656
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 415.90625,
      "epoch": 0.5256,
      "grad_norm": 0.3646902862710937,
      "kl": 0.10595703125,
      "learning_rate": 4.63077925135198e-06,
      "loss": 0.0042,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 657
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 432.1875,
      "epoch": 0.5264,
      "grad_norm": 0.670708203135066,
      "kl": 0.10009765625,
      "learning_rate": 4.629683063500319e-06,
      "loss": 0.004,
      "reward": 1.53125,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 658
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 372.828125,
      "epoch": 0.5272,
      "grad_norm": 0.7104430342537692,
      "kl": 0.98828125,
      "learning_rate": 4.628585380955104e-06,
      "loss": 0.0394,
      "reward": 1.625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 659
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 402.796875,
      "epoch": 0.528,
      "grad_norm": 0.8181530916132829,
      "kl": 0.91796875,
      "learning_rate": 4.62748620448673e-06,
      "loss": 0.0368,
      "reward": 1.640625,
      "reward_std": 0.2109457403421402,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 660
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 384.015625,
      "epoch": 0.5288,
      "grad_norm": 0.6460679965046602,
      "kl": 0.81640625,
      "learning_rate": 4.626385534866642e-06,
      "loss": 0.0327,
      "reward": 1.765625,
      "reward_std": 0.15981829166412354,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 661
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 428.0625,
      "epoch": 0.5296,
      "grad_norm": 0.8947737055186511,
      "kl": 0.11279296875,
      "learning_rate": 4.625283372867333e-06,
      "loss": 0.0045,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 662
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 413.65625,
      "epoch": 0.5304,
      "grad_norm": 0.6340310799448744,
      "kl": 0.10595703125,
      "learning_rate": 4.624179719262342e-06,
      "loss": 0.0042,
      "reward": 1.859375,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 663
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 397.96875,
      "epoch": 0.5312,
      "grad_norm": 0.9445756454310534,
      "kl": 0.890625,
      "learning_rate": 4.623074574826254e-06,
      "loss": 0.0355,
      "reward": 1.765625,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 664
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 366.71875,
      "epoch": 0.532,
      "grad_norm": 1.6832785858802943,
      "kl": 0.12353515625,
      "learning_rate": 4.621967940334705e-06,
      "loss": 0.0049,
      "reward": 1.609375,
      "reward_std": 0.2651395797729492,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 665
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 436.90625,
      "epoch": 0.5328,
      "grad_norm": 0.3725953073806499,
      "kl": 1.0859375,
      "learning_rate": 4.620859816564371e-06,
      "loss": 0.0434,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 666
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 369.609375,
      "epoch": 0.5336,
      "grad_norm": 2.10160684646745,
      "kl": 0.11572265625,
      "learning_rate": 4.619750204292978e-06,
      "loss": 0.0046,
      "reward": 1.59375,
      "reward_std": 0.1962026059627533,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 667
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 396.53125,
      "epoch": 0.5344,
      "grad_norm": 1.0811850376721042,
      "kl": 0.12060546875,
      "learning_rate": 4.618639104299294e-06,
      "loss": 0.0048,
      "reward": 1.71875,
      "reward_std": 0.28247907757759094,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 668
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 372.96875,
      "epoch": 0.5352,
      "grad_norm": 4.040796775648103,
      "kl": 0.365234375,
      "learning_rate": 4.6175265173631304e-06,
      "loss": 0.0147,
      "reward": 1.609375,
      "reward_std": 0.33351296186447144,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 669
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 367.046875,
      "epoch": 0.536,
      "grad_norm": 0.9364156193477713,
      "kl": 0.84765625,
      "learning_rate": 4.616412444265344e-06,
      "loss": 0.0338,
      "reward": 1.671875,
      "reward_std": 0.23531240224838257,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 670
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 377.015625,
      "epoch": 0.5368,
      "grad_norm": 0.7278856518684674,
      "kl": 0.11328125,
      "learning_rate": 4.6152968857878365e-06,
      "loss": 0.0045,
      "reward": 1.765625,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 671
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 367.828125,
      "epoch": 0.5376,
      "grad_norm": 1.907213125846056,
      "kl": 0.119140625,
      "learning_rate": 4.6141798427135475e-06,
      "loss": 0.0048,
      "reward": 1.6875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 672
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 419.390625,
      "epoch": 0.5384,
      "grad_norm": 0.6939537294722774,
      "kl": 0.115234375,
      "learning_rate": 4.6130613158264605e-06,
      "loss": 0.0046,
      "reward": 1.859375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 673
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 462.15625,
      "epoch": 0.5392,
      "grad_norm": 2.732658890054957,
      "kl": 1.125,
      "learning_rate": 4.611941305911602e-06,
      "loss": 0.0452,
      "reward": 1.875,
      "reward_std": 0.1243029236793518,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 674
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 388.640625,
      "epoch": 0.54,
      "grad_norm": 0.6753912416589343,
      "kl": 0.8515625,
      "learning_rate": 4.610819813755038e-06,
      "loss": 0.0339,
      "reward": 1.71875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 675
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 404.625,
      "epoch": 0.5408,
      "grad_norm": 1.9874445396659512,
      "kl": 0.1298828125,
      "learning_rate": 4.609696840143875e-06,
      "loss": 0.0052,
      "reward": 1.546875,
      "reward_std": 0.25726157426834106,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 676
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 396.46875,
      "epoch": 0.5416,
      "grad_norm": 0.7510630767554702,
      "kl": 0.9609375,
      "learning_rate": 4.6085723858662575e-06,
      "loss": 0.0383,
      "reward": 1.515625,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 677
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 455.5,
      "epoch": 0.5424,
      "grad_norm": 1.6421316050848016,
      "kl": 1.6953125,
      "learning_rate": 4.607446451711372e-06,
      "loss": 0.0678,
      "reward": 1.765625,
      "reward_std": 0.2414703369140625,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.96875,
      "step": 678
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 378.921875,
      "epoch": 0.5432,
      "grad_norm": 1.1696105580122462,
      "kl": 0.224609375,
      "learning_rate": 4.606319038469443e-06,
      "loss": 0.009,
      "reward": 1.5,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 679
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 421.375,
      "epoch": 0.544,
      "grad_norm": 0.6993453019710336,
      "kl": 0.84765625,
      "learning_rate": 4.605190146931731e-06,
      "loss": 0.0337,
      "reward": 1.71875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 680
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 364.59375,
      "epoch": 0.5448,
      "grad_norm": 1.0028918860428877,
      "kl": 0.55078125,
      "learning_rate": 4.604059777890537e-06,
      "loss": 0.0221,
      "reward": 1.515625,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 681
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 371.484375,
      "epoch": 0.5456,
      "grad_norm": 0.9951682898291854,
      "kl": 0.703125,
      "learning_rate": 4.602927932139197e-06,
      "loss": 0.0282,
      "reward": 1.671875,
      "reward_std": 0.20189079642295837,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 682
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 409.0625,
      "epoch": 0.5464,
      "grad_norm": 0.5659555810744075,
      "kl": 0.59765625,
      "learning_rate": 4.601794610472083e-06,
      "loss": 0.024,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 683
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 345.3125,
      "epoch": 0.5472,
      "grad_norm": 0.9820024220569756,
      "kl": 1.9453125,
      "learning_rate": 4.6006598136846056e-06,
      "loss": 0.0779,
      "reward": 1.703125,
      "reward_std": 0.3012469410896301,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.953125,
      "step": 684
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 398.5,
      "epoch": 0.548,
      "grad_norm": 1.6200290850707568,
      "kl": 1.46875,
      "learning_rate": 4.599523542573207e-06,
      "loss": 0.0586,
      "reward": 1.59375,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.96875,
      "step": 685
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 369.875,
      "epoch": 0.5488,
      "grad_norm": 1.1585081024064825,
      "kl": 0.404296875,
      "learning_rate": 4.598385797935368e-06,
      "loss": 0.0162,
      "reward": 1.53125,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 686
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 389.828125,
      "epoch": 0.5496,
      "grad_norm": 0.5636989474360153,
      "kl": 0.12353515625,
      "learning_rate": 4.5972465805696e-06,
      "loss": 0.0049,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 687
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 343.234375,
      "epoch": 0.5504,
      "grad_norm": 0.7489029107882763,
      "kl": 0.125,
      "learning_rate": 4.596105891275449e-06,
      "loss": 0.005,
      "reward": 1.65625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 688
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 366.140625,
      "epoch": 0.5512,
      "grad_norm": 0.07564483123034887,
      "kl": 0.11474609375,
      "learning_rate": 4.594963730853497e-06,
      "loss": 0.0046,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 689
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 329.015625,
      "epoch": 0.552,
      "grad_norm": 0.7715043962696903,
      "kl": 0.1318359375,
      "learning_rate": 4.593820100105355e-06,
      "loss": 0.0053,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 690
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 408.4375,
      "epoch": 0.5528,
      "grad_norm": 1.5903541288745378,
      "kl": 0.13671875,
      "learning_rate": 4.5926749998336665e-06,
      "loss": 0.0054,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 691
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 384.09375,
      "epoch": 0.5536,
      "grad_norm": 0.8409980617832177,
      "kl": 0.12890625,
      "learning_rate": 4.5915284308421075e-06,
      "loss": 0.0051,
      "reward": 1.71875,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 692
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 407.84375,
      "epoch": 0.5544,
      "grad_norm": 0.11505513163004885,
      "kl": 0.11767578125,
      "learning_rate": 4.590380393935383e-06,
      "loss": 0.0047,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 693
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 337.8125,
      "epoch": 0.5552,
      "grad_norm": 1.0015722524145403,
      "kl": 0.134765625,
      "learning_rate": 4.589230889919232e-06,
      "loss": 0.0054,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 694
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 407.96875,
      "epoch": 0.556,
      "grad_norm": 0.7181094305585296,
      "kl": 0.7421875,
      "learning_rate": 4.588079919600419e-06,
      "loss": 0.0297,
      "reward": 1.578125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 695
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 383.515625,
      "epoch": 0.5568,
      "grad_norm": 1.032777656021463,
      "kl": 0.80859375,
      "learning_rate": 4.586927483786739e-06,
      "loss": 0.0324,
      "reward": 1.65625,
      "reward_std": 0.20266614854335785,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 696
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 399.96875,
      "epoch": 0.5576,
      "grad_norm": 0.7345604816782041,
      "kl": 0.30859375,
      "learning_rate": 4.585773583287017e-06,
      "loss": 0.0124,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 697
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 417.484375,
      "epoch": 0.5584,
      "grad_norm": 1.0895463669750132,
      "kl": 0.73828125,
      "learning_rate": 4.584618218911104e-06,
      "loss": 0.0296,
      "reward": 1.578125,
      "reward_std": 0.25726157426834106,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 698
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 351.46875,
      "epoch": 0.5592,
      "grad_norm": 1.597638201896758,
      "kl": 0.16015625,
      "learning_rate": 4.583461391469879e-06,
      "loss": 0.0064,
      "reward": 1.546875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 699
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 374.21875,
      "epoch": 0.56,
      "grad_norm": 0.5662371958432129,
      "kl": 0.119140625,
      "learning_rate": 4.582303101775249e-06,
      "loss": 0.0048,
      "reward": 1.65625,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 700
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 388.984375,
      "epoch": 0.5608,
      "grad_norm": 2.0291530281532735,
      "kl": 0.87890625,
      "learning_rate": 4.581143350640146e-06,
      "loss": 0.0352,
      "reward": 1.828125,
      "reward_std": 0.2993340790271759,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 701
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 401.328125,
      "epoch": 0.5616,
      "grad_norm": 0.6348448126263817,
      "kl": 0.5625,
      "learning_rate": 4.579982138878527e-06,
      "loss": 0.0224,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 702
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 416.28125,
      "epoch": 0.5624,
      "grad_norm": 0.9368449426361695,
      "kl": 2.140625,
      "learning_rate": 4.578819467305375e-06,
      "loss": 0.0855,
      "reward": 1.625,
      "reward_std": 0.3049619197845459,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.953125,
      "step": 703
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 421.015625,
      "epoch": 0.5632,
      "grad_norm": 0.913545916055359,
      "kl": 2.15625,
      "learning_rate": 4.5776553367367e-06,
      "loss": 0.086,
      "reward": 1.5625,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 704
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 418.625,
      "epoch": 0.564,
      "grad_norm": 0.6110781288263009,
      "kl": 0.12353515625,
      "learning_rate": 4.576489747989532e-06,
      "loss": 0.0049,
      "reward": 1.671875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 705
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 383.5625,
      "epoch": 0.5648,
      "grad_norm": 0.755254769811337,
      "kl": 0.74609375,
      "learning_rate": 4.575322701881926e-06,
      "loss": 0.0297,
      "reward": 1.46875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.984375,
      "step": 706
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 425.125,
      "epoch": 0.5656,
      "grad_norm": 1.247168348852021,
      "kl": 0.953125,
      "learning_rate": 4.57415419923296e-06,
      "loss": 0.0381,
      "reward": 1.578125,
      "reward_std": 0.2414703369140625,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 707
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 424.328125,
      "epoch": 0.5664,
      "grad_norm": 0.6235588189660318,
      "kl": 0.1259765625,
      "learning_rate": 4.572984240862733e-06,
      "loss": 0.0051,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 708
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 388.453125,
      "epoch": 0.5672,
      "grad_norm": 1.547770321678808,
      "kl": 4.03125,
      "learning_rate": 4.57181282759237e-06,
      "loss": 0.1616,
      "reward": 1.46875,
      "reward_std": 0.5385891795158386,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.890625,
      "step": 709
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 406.375,
      "epoch": 0.568,
      "grad_norm": 0.7482968349890383,
      "kl": 1.0,
      "learning_rate": 4.570639960244011e-06,
      "loss": 0.0401,
      "reward": 1.546875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 710
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 440.59375,
      "epoch": 0.5688,
      "grad_norm": 0.8559596436138244,
      "kl": 0.111328125,
      "learning_rate": 4.56946563964082e-06,
      "loss": 0.0045,
      "reward": 1.859375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 711
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 409.96875,
      "epoch": 0.5696,
      "grad_norm": 1.0352416021885698,
      "kl": 0.126953125,
      "learning_rate": 4.5682898666069815e-06,
      "loss": 0.0051,
      "reward": 1.671875,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 712
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 403.21875,
      "epoch": 0.5704,
      "grad_norm": 1.0665806428938223,
      "kl": 0.6875,
      "learning_rate": 4.567112641967697e-06,
      "loss": 0.0274,
      "reward": 1.578125,
      "reward_std": 0.2662131190299988,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 713
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 397.5625,
      "epoch": 0.5712,
      "grad_norm": 1.4321423266016051,
      "kl": 0.1279296875,
      "learning_rate": 4.5659339665491894e-06,
      "loss": 0.0051,
      "reward": 1.6875,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 714
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 435.9375,
      "epoch": 0.572,
      "grad_norm": 0.6344378789069257,
      "kl": 0.11328125,
      "learning_rate": 4.5647538411786965e-06,
      "loss": 0.0045,
      "reward": 1.546875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 715
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 402.0,
      "epoch": 0.5728,
      "grad_norm": 0.6581202971018658,
      "kl": 0.1259765625,
      "learning_rate": 4.563572266684478e-06,
      "loss": 0.005,
      "reward": 1.84375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 716
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 442.890625,
      "epoch": 0.5736,
      "grad_norm": 1.1452932072382693,
      "kl": 0.478515625,
      "learning_rate": 4.562389243895807e-06,
      "loss": 0.0191,
      "reward": 1.546875,
      "reward_std": 0.19044627249240875,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 717
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 432.90625,
      "epoch": 0.5744,
      "grad_norm": 0.5898405189255272,
      "kl": 0.703125,
      "learning_rate": 4.561204773642974e-06,
      "loss": 0.0282,
      "reward": 1.703125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 718
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.5,
      "completion_length": 388.796875,
      "epoch": 0.5752,
      "grad_norm": 0.7806688656878049,
      "kl": 0.1455078125,
      "learning_rate": 4.5600188567572874e-06,
      "loss": 0.0058,
      "reward": 1.375,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 719
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 430.78125,
      "epoch": 0.576,
      "grad_norm": 0.8776508767164352,
      "kl": 2.28125,
      "learning_rate": 4.558831494071069e-06,
      "loss": 0.0909,
      "reward": 1.640625,
      "reward_std": 0.329858660697937,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.953125,
      "step": 720
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 399.203125,
      "epoch": 0.5768,
      "grad_norm": 1.0941455692436421,
      "kl": 1.046875,
      "learning_rate": 4.557642686417654e-06,
      "loss": 0.0418,
      "reward": 1.6875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 721
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 457.28125,
      "epoch": 0.5776,
      "grad_norm": 0.6712978048188565,
      "kl": 1.9140625,
      "learning_rate": 4.556452434631396e-06,
      "loss": 0.0764,
      "reward": 1.671875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.953125,
      "step": 722
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 443.734375,
      "epoch": 0.5784,
      "grad_norm": 0.7381385221923685,
      "kl": 0.134765625,
      "learning_rate": 4.555260739547657e-06,
      "loss": 0.0054,
      "reward": 1.5625,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 723
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 423.421875,
      "epoch": 0.5792,
      "grad_norm": 0.688014727599496,
      "kl": 0.6484375,
      "learning_rate": 4.554067602002815e-06,
      "loss": 0.0259,
      "reward": 1.640625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 724
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 419.796875,
      "epoch": 0.58,
      "grad_norm": 1.8472595817554485,
      "kl": 0.84375,
      "learning_rate": 4.55287302283426e-06,
      "loss": 0.0335,
      "reward": 1.609375,
      "reward_std": 0.19044627249240875,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 725
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 460.46875,
      "epoch": 0.5808,
      "grad_norm": 0.8264857423169704,
      "kl": 2.21875,
      "learning_rate": 4.551677002880395e-06,
      "loss": 0.0891,
      "reward": 1.875,
      "reward_std": 0.2619796395301819,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.953125,
      "step": 726
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 417.484375,
      "epoch": 0.5816,
      "grad_norm": 0.785843803400338,
      "kl": 0.1181640625,
      "learning_rate": 4.550479542980632e-06,
      "loss": 0.0047,
      "reward": 1.859375,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 727
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 424.0,
      "epoch": 0.5824,
      "grad_norm": 0.8086127282101715,
      "kl": 0.69921875,
      "learning_rate": 4.549280643975394e-06,
      "loss": 0.0279,
      "reward": 1.71875,
      "reward_std": 0.2177756279706955,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 728
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 429.9375,
      "epoch": 0.5832,
      "grad_norm": 0.761908604882218,
      "kl": 1.2890625,
      "learning_rate": 4.548080306706114e-06,
      "loss": 0.0516,
      "reward": 1.578125,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.96875,
      "step": 729
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 383.625,
      "epoch": 0.584,
      "grad_norm": 0.9352748214370604,
      "kl": 0.453125,
      "learning_rate": 4.5468785320152365e-06,
      "loss": 0.0182,
      "reward": 1.703125,
      "reward_std": 0.25726157426834106,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 730
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 369.546875,
      "epoch": 0.5848,
      "grad_norm": 1.199317088782103,
      "kl": 0.359375,
      "learning_rate": 4.545675320746212e-06,
      "loss": 0.0144,
      "reward": 1.75,
      "reward_std": 0.3197399973869324,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 731
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 386.796875,
      "epoch": 0.5856,
      "grad_norm": 1.0920792282092384,
      "kl": 0.11962890625,
      "learning_rate": 4.544470673743502e-06,
      "loss": 0.0048,
      "reward": 1.328125,
      "reward_std": 0.2472364753484726,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 732
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 476.21875,
      "epoch": 0.5864,
      "grad_norm": 4.64403574438647,
      "kl": 0.11767578125,
      "learning_rate": 4.543264591852572e-06,
      "loss": 0.0047,
      "reward": 1.65625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 733
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 404.21875,
      "epoch": 0.5872,
      "grad_norm": 0.5808856463061708,
      "kl": 0.671875,
      "learning_rate": 4.542057075919898e-06,
      "loss": 0.0269,
      "reward": 1.796875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 734
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 408.96875,
      "epoch": 0.588,
      "grad_norm": 0.9289541622106935,
      "kl": 0.88671875,
      "learning_rate": 4.54084812679296e-06,
      "loss": 0.0353,
      "reward": 1.484375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.984375,
      "step": 735
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 399.09375,
      "epoch": 0.5888,
      "grad_norm": 0.5923941542606462,
      "kl": 0.1337890625,
      "learning_rate": 4.539637745320247e-06,
      "loss": 0.0053,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 736
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 420.046875,
      "epoch": 0.5896,
      "grad_norm": 0.5623431689697669,
      "kl": 0.11474609375,
      "learning_rate": 4.53842593235125e-06,
      "loss": 0.0046,
      "reward": 1.71875,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 737
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 397.015625,
      "epoch": 0.5904,
      "grad_norm": 0.7521836983481913,
      "kl": 0.77734375,
      "learning_rate": 4.537212688736466e-06,
      "loss": 0.031,
      "reward": 1.578125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 738
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 431.0,
      "epoch": 0.5912,
      "grad_norm": 1.1721982513328153,
      "kl": 0.54296875,
      "learning_rate": 4.535998015327396e-06,
      "loss": 0.0217,
      "reward": 1.546875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 739
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 388.234375,
      "epoch": 0.592,
      "grad_norm": 1.1851939620543772,
      "kl": 0.54296875,
      "learning_rate": 4.534781912976546e-06,
      "loss": 0.0217,
      "reward": 1.53125,
      "reward_std": 0.24251843988895416,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 740
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 385.71875,
      "epoch": 0.5928,
      "grad_norm": 1.1869725036453096,
      "kl": 0.455078125,
      "learning_rate": 4.533564382537421e-06,
      "loss": 0.0181,
      "reward": 1.6875,
      "reward_std": 0.23356689512729645,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 741
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 384.046875,
      "epoch": 0.5936,
      "grad_norm": 0.5718438675844124,
      "kl": 0.12060546875,
      "learning_rate": 4.532345424864533e-06,
      "loss": 0.0048,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 742
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 455.265625,
      "epoch": 0.5944,
      "grad_norm": 5.584850455407113,
      "kl": 0.431640625,
      "learning_rate": 4.531125040813392e-06,
      "loss": 0.0172,
      "reward": 1.921875,
      "reward_std": 0.19044627249240875,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.96875,
      "step": 743
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.5,
      "completion_length": 477.796875,
      "epoch": 0.5952,
      "grad_norm": 51.22100161387199,
      "kl": 30.0,
      "learning_rate": 4.529903231240511e-06,
      "loss": 1.1987,
      "reward": 0.5,
      "reward_std": 0.46577346324920654,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 0.3125,
      "step": 744
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.875,
      "completion_length": 517.625,
      "epoch": 0.596,
      "grad_norm": 39.20600833098527,
      "kl": 29.0,
      "learning_rate": 4.528679997003403e-06,
      "loss": 1.1599,
      "reward": 0.328125,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.078125,
      "rewards/format_reward": 0.25,
      "step": 745
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 529.28125,
      "epoch": 0.5968,
      "grad_norm": 20.64166517450329,
      "kl": 15.25,
      "learning_rate": 4.52745533896058e-06,
      "loss": 0.6117,
      "reward": 0.90625,
      "reward_std": 0.3945944905281067,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.484375,
      "step": 746
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 613.9375,
      "epoch": 0.5976,
      "grad_norm": 12.692577062543846,
      "kl": 11.1875,
      "learning_rate": 4.526229257971556e-06,
      "loss": 0.4475,
      "reward": 0.90625,
      "reward_std": 0.7484728097915649,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.40625,
      "step": 747
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 614.953125,
      "epoch": 0.5984,
      "grad_norm": 10.667956388663862,
      "kl": 4.0625,
      "learning_rate": 4.52500175489684e-06,
      "loss": 0.1625,
      "reward": 1.0,
      "reward_std": 0.6321523189544678,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.484375,
      "step": 748
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 466.03125,
      "epoch": 0.5992,
      "grad_norm": 3.5404744863131983,
      "kl": 0.5859375,
      "learning_rate": 4.523772830597942e-06,
      "loss": 0.0234,
      "reward": 1.1875,
      "reward_std": 0.3434731066226959,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.671875,
      "step": 749
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 378.0,
      "epoch": 0.6,
      "grad_norm": 1.7548650503096686,
      "kl": 0.1298828125,
      "learning_rate": 4.522542485937369e-06,
      "loss": 0.0052,
      "reward": 1.625,
      "reward_std": 0.2619796395301819,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.96875,
      "step": 750
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 468.15625,
      "epoch": 0.6008,
      "grad_norm": 1.6651771330256788,
      "kl": 0.169921875,
      "learning_rate": 4.521310721778622e-06,
      "loss": 0.0068,
      "reward": 1.15625,
      "reward_std": 0.33090677857398987,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.65625,
      "step": 751
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 449.609375,
      "epoch": 0.6016,
      "grad_norm": 1.0959913148107947,
      "kl": 0.142578125,
      "learning_rate": 4.520077538986203e-06,
      "loss": 0.0057,
      "reward": 1.640625,
      "reward_std": 0.23925508558750153,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.796875,
      "step": 752
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 442.046875,
      "epoch": 0.6024,
      "grad_norm": 0.6686004200543701,
      "kl": 0.1259765625,
      "learning_rate": 4.518842938425606e-06,
      "loss": 0.005,
      "reward": 1.78125,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 753
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 396.015625,
      "epoch": 0.6032,
      "grad_norm": 0.5521083003632641,
      "kl": 0.134765625,
      "learning_rate": 4.51760692096332e-06,
      "loss": 0.0054,
      "reward": 1.515625,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 754
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 418.421875,
      "epoch": 0.604,
      "grad_norm": 0.576389331229294,
      "kl": 0.119140625,
      "learning_rate": 4.516369487466832e-06,
      "loss": 0.0048,
      "reward": 1.625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 755
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 397.640625,
      "epoch": 0.6048,
      "grad_norm": 0.9091605258035641,
      "kl": 0.1337890625,
      "learning_rate": 4.5151306388046175e-06,
      "loss": 0.0053,
      "reward": 1.578125,
      "reward_std": 0.2956691384315491,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.921875,
      "step": 756
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 370.828125,
      "epoch": 0.6056,
      "grad_norm": 0.7456635203896337,
      "kl": 0.1318359375,
      "learning_rate": 4.513890375846152e-06,
      "loss": 0.0053,
      "reward": 1.546875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 757
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 374.53125,
      "epoch": 0.6064,
      "grad_norm": 0.4054554163375172,
      "kl": 0.138671875,
      "learning_rate": 4.512648699461897e-06,
      "loss": 0.0055,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 758
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.5,
      "completion_length": 353.203125,
      "epoch": 0.6072,
      "grad_norm": 0.5153817014977968,
      "kl": 0.1416015625,
      "learning_rate": 4.511405610523309e-06,
      "loss": 0.0057,
      "reward": 1.453125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 759
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 348.015625,
      "epoch": 0.608,
      "grad_norm": 0.9645685741953072,
      "kl": 0.13671875,
      "learning_rate": 4.510161109902837e-06,
      "loss": 0.0055,
      "reward": 1.6875,
      "reward_std": 0.23319074511528015,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 760
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 352.625,
      "epoch": 0.6088,
      "grad_norm": 0.6561053834688735,
      "kl": 0.1357421875,
      "learning_rate": 4.508915198473919e-06,
      "loss": 0.0055,
      "reward": 1.71875,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 761
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 371.453125,
      "epoch": 0.6096,
      "grad_norm": 0.4154680865776334,
      "kl": 0.1328125,
      "learning_rate": 4.507667877110982e-06,
      "loss": 0.0053,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 762
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 334.875,
      "epoch": 0.6104,
      "grad_norm": 1.0690327606009025,
      "kl": 0.1357421875,
      "learning_rate": 4.506419146689445e-06,
      "loss": 0.0054,
      "reward": 1.484375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 763
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 357.59375,
      "epoch": 0.6112,
      "grad_norm": 0.09845757431172224,
      "kl": 0.12109375,
      "learning_rate": 4.505169008085717e-06,
      "loss": 0.0049,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 764
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 352.3125,
      "epoch": 0.612,
      "grad_norm": 0.9511197576575076,
      "kl": 0.12890625,
      "learning_rate": 4.503917462177192e-06,
      "loss": 0.0051,
      "reward": 1.6875,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 765
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 354.78125,
      "epoch": 0.6128,
      "grad_norm": 0.3822778374443259,
      "kl": 0.12890625,
      "learning_rate": 4.5026645098422515e-06,
      "loss": 0.0051,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 766
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 368.890625,
      "epoch": 0.6136,
      "grad_norm": 0.8657068830504863,
      "kl": 0.1328125,
      "learning_rate": 4.5014101519602684e-06,
      "loss": 0.0053,
      "reward": 1.515625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 767
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 350.25,
      "epoch": 0.6144,
      "grad_norm": 0.6166122855295121,
      "kl": 0.1328125,
      "learning_rate": 4.500154389411598e-06,
      "loss": 0.0053,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 768
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 360.21875,
      "epoch": 0.6152,
      "grad_norm": 0.8540974873902282,
      "kl": 0.126953125,
      "learning_rate": 4.498897223077582e-06,
      "loss": 0.0051,
      "reward": 1.6875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 769
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 371.90625,
      "epoch": 0.616,
      "grad_norm": 1.2215144813460077,
      "kl": 0.134765625,
      "learning_rate": 4.49763865384055e-06,
      "loss": 0.0054,
      "reward": 1.59375,
      "reward_std": 0.2040124535560608,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.96875,
      "step": 770
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 378.484375,
      "epoch": 0.6168,
      "grad_norm": 0.8932655228788702,
      "kl": 0.12890625,
      "learning_rate": 4.496378682583813e-06,
      "loss": 0.0052,
      "reward": 1.796875,
      "reward_std": 0.22673699259757996,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 771
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 311.90625,
      "epoch": 0.6176,
      "grad_norm": 0.8671336585205175,
      "kl": 0.1416015625,
      "learning_rate": 4.495117310191667e-06,
      "loss": 0.0057,
      "reward": 1.4375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 772
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 321.75,
      "epoch": 0.6184,
      "grad_norm": 0.8212597306559153,
      "kl": 0.140625,
      "learning_rate": 4.493854537549393e-06,
      "loss": 0.0056,
      "reward": 1.546875,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 773
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 301.171875,
      "epoch": 0.6192,
      "grad_norm": 1.5452424345377327,
      "kl": 0.1328125,
      "learning_rate": 4.492590365543253e-06,
      "loss": 0.0053,
      "reward": 1.453125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 774
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 347.703125,
      "epoch": 0.62,
      "grad_norm": 0.8303947701075123,
      "kl": 0.119140625,
      "learning_rate": 4.491324795060491e-06,
      "loss": 0.0048,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 775
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 363.9375,
      "epoch": 0.6208,
      "grad_norm": 0.7069099324776755,
      "kl": 0.11376953125,
      "learning_rate": 4.490057826989333e-06,
      "loss": 0.0045,
      "reward": 1.375,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 776
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 319.453125,
      "epoch": 0.6216,
      "grad_norm": 2.817713219267624,
      "kl": 0.12158203125,
      "learning_rate": 4.488789462218988e-06,
      "loss": 0.0049,
      "reward": 1.734375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 777
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 366.40625,
      "epoch": 0.6224,
      "grad_norm": 0.47685500021164945,
      "kl": 0.1171875,
      "learning_rate": 4.487519701639641e-06,
      "loss": 0.0047,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 778
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 399.75,
      "epoch": 0.6232,
      "grad_norm": 0.566943640083845,
      "kl": 0.1083984375,
      "learning_rate": 4.486248546142459e-06,
      "loss": 0.0043,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 779
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 323.515625,
      "epoch": 0.624,
      "grad_norm": 0.9569549841774113,
      "kl": 0.12353515625,
      "learning_rate": 4.4849759966195885e-06,
      "loss": 0.0049,
      "reward": 1.4375,
      "reward_std": 0.24251842498779297,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.984375,
      "step": 780
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 393.875,
      "epoch": 0.6248,
      "grad_norm": 0.6087250698429332,
      "kl": 0.1279296875,
      "learning_rate": 4.483702053964154e-06,
      "loss": 0.0051,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 781
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 409.75,
      "epoch": 0.6256,
      "grad_norm": 0.7636360924647048,
      "kl": 0.11328125,
      "learning_rate": 4.482426719070258e-06,
      "loss": 0.0045,
      "reward": 1.75,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 782
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 388.59375,
      "epoch": 0.6264,
      "grad_norm": 0.9792797804278635,
      "kl": 0.125,
      "learning_rate": 4.4811499928329775e-06,
      "loss": 0.005,
      "reward": 1.4375,
      "reward_std": 0.2734241485595703,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.96875,
      "step": 783
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 425.796875,
      "epoch": 0.6272,
      "grad_norm": 76.4412477834923,
      "kl": 2.5,
      "learning_rate": 4.479871876148368e-06,
      "loss": 0.0995,
      "reward": 1.78125,
      "reward_std": 0.19506090879440308,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 784
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 420.34375,
      "epoch": 0.628,
      "grad_norm": 0.5444322017275038,
      "kl": 0.125,
      "learning_rate": 4.478592369913464e-06,
      "loss": 0.005,
      "reward": 1.53125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 785
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 458.703125,
      "epoch": 0.6288,
      "grad_norm": 0.08333092753592482,
      "kl": 0.12353515625,
      "learning_rate": 4.477311475026271e-06,
      "loss": 0.0049,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 786
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 441.546875,
      "epoch": 0.6296,
      "grad_norm": 0.32721957273627983,
      "kl": 0.115234375,
      "learning_rate": 4.476029192385769e-06,
      "loss": 0.0046,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 787
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 452.875,
      "epoch": 0.6304,
      "grad_norm": 0.565466382848136,
      "kl": 0.11181640625,
      "learning_rate": 4.474745522891915e-06,
      "loss": 0.0045,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 788
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 377.546875,
      "epoch": 0.6312,
      "grad_norm": 0.8991346765306769,
      "kl": 0.126953125,
      "learning_rate": 4.473460467445637e-06,
      "loss": 0.0051,
      "reward": 1.65625,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.953125,
      "step": 789
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 452.40625,
      "epoch": 0.632,
      "grad_norm": 0.7590185646436524,
      "kl": 0.11669921875,
      "learning_rate": 4.472174026948836e-06,
      "loss": 0.0047,
      "reward": 1.71875,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 790
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 434.125,
      "epoch": 0.6328,
      "grad_norm": 0.6733327069903833,
      "kl": 0.123046875,
      "learning_rate": 4.470886202304385e-06,
      "loss": 0.0049,
      "reward": 1.609375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 791
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 352.796875,
      "epoch": 0.6336,
      "grad_norm": 0.8714933056678619,
      "kl": 0.140625,
      "learning_rate": 4.469596994416131e-06,
      "loss": 0.0056,
      "reward": 1.65625,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 792
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 369.28125,
      "epoch": 0.6344,
      "grad_norm": 2.659537158653402,
      "kl": 0.1396484375,
      "learning_rate": 4.468306404188887e-06,
      "loss": 0.0056,
      "reward": 1.484375,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.984375,
      "step": 793
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 447.296875,
      "epoch": 0.6352,
      "grad_norm": 24.563601751105438,
      "kl": 0.224609375,
      "learning_rate": 4.467014432528441e-06,
      "loss": 0.009,
      "reward": 1.484375,
      "reward_std": 0.2414703369140625,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.96875,
      "step": 794
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 393.484375,
      "epoch": 0.636,
      "grad_norm": 1.4850003247494263,
      "kl": 0.134765625,
      "learning_rate": 4.465721080341547e-06,
      "loss": 0.0054,
      "reward": 1.796875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 795
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 373.96875,
      "epoch": 0.6368,
      "grad_norm": 1.44494639151913,
      "kl": 0.12890625,
      "learning_rate": 4.4644263485359316e-06,
      "loss": 0.0052,
      "reward": 1.515625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 796
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 392.328125,
      "epoch": 0.6376,
      "grad_norm": 0.4537455588808983,
      "kl": 0.134765625,
      "learning_rate": 4.463130238020284e-06,
      "loss": 0.0054,
      "reward": 1.53125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 797
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 439.4375,
      "epoch": 0.6384,
      "grad_norm": 0.4835674647328229,
      "kl": 0.109375,
      "learning_rate": 4.4618327497042676e-06,
      "loss": 0.0044,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 798
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 361.4375,
      "epoch": 0.6392,
      "grad_norm": 1.2639303070692236,
      "kl": 0.134765625,
      "learning_rate": 4.460533884498509e-06,
      "loss": 0.0054,
      "reward": 1.421875,
      "reward_std": 0.32878512144088745,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.96875,
      "step": 799
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 386.453125,
      "epoch": 0.64,
      "grad_norm": 1.0922796218789752,
      "kl": 0.123046875,
      "learning_rate": 4.4592336433146e-06,
      "loss": 0.0049,
      "reward": 1.609375,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 800
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 390.453125,
      "epoch": 0.6408,
      "grad_norm": 1.0156191846007012,
      "kl": 0.1240234375,
      "learning_rate": 4.457932027065102e-06,
      "loss": 0.005,
      "reward": 1.53125,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.96875,
      "step": 801
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 351.078125,
      "epoch": 0.6416,
      "grad_norm": 0.8198981311230296,
      "kl": 0.12255859375,
      "learning_rate": 4.456629036663537e-06,
      "loss": 0.0049,
      "reward": 1.65625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 802
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 468.921875,
      "epoch": 0.6424,
      "grad_norm": 0.5991400033831288,
      "kl": 0.1318359375,
      "learning_rate": 4.455324673024396e-06,
      "loss": 0.0053,
      "reward": 1.9375,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.96875,
      "step": 803
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 422.796875,
      "epoch": 0.6432,
      "grad_norm": 0.6692124112556407,
      "kl": 0.1318359375,
      "learning_rate": 4.4540189370631315e-06,
      "loss": 0.0053,
      "reward": 1.515625,
      "reward_std": 0.24831004440784454,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 804
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 438.140625,
      "epoch": 0.644,
      "grad_norm": 6.832094630975682,
      "kl": 250.0,
      "learning_rate": 4.452711829696158e-06,
      "loss": 10.0082,
      "reward": 1.390625,
      "reward_std": 0.1889965683221817,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.96875,
      "step": 805
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 374.90625,
      "epoch": 0.6448,
      "grad_norm": 0.9717012307975434,
      "kl": 0.1376953125,
      "learning_rate": 4.451403351840855e-06,
      "loss": 0.0055,
      "reward": 1.609375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.953125,
      "step": 806
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 449.40625,
      "epoch": 0.6456,
      "grad_norm": 0.956411793550628,
      "kl": 0.130859375,
      "learning_rate": 4.450093504415562e-06,
      "loss": 0.0052,
      "reward": 1.578125,
      "reward_std": 0.24820663034915924,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.953125,
      "step": 807
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 381.015625,
      "epoch": 0.6464,
      "grad_norm": 0.9250986580827515,
      "kl": 0.125,
      "learning_rate": 4.44878228833958e-06,
      "loss": 0.005,
      "reward": 1.546875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 808
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 407.9375,
      "epoch": 0.6472,
      "grad_norm": 7.617951227193584,
      "kl": 215.0,
      "learning_rate": 4.447469704533172e-06,
      "loss": 8.5714,
      "reward": 1.515625,
      "reward_std": 0.29317614436149597,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.921875,
      "step": 809
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.5,
      "completion_length": 430.40625,
      "epoch": 0.648,
      "grad_norm": 1.5574071141834538,
      "kl": 0.1591796875,
      "learning_rate": 4.446155753917559e-06,
      "loss": 0.0064,
      "reward": 1.328125,
      "reward_std": 0.3298586905002594,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.921875,
      "step": 810
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 395.078125,
      "epoch": 0.6488,
      "grad_norm": 1.160622746543496,
      "kl": 71.5,
      "learning_rate": 4.444840437414923e-06,
      "loss": 2.8578,
      "reward": 1.5,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.953125,
      "step": 811
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 393.78125,
      "epoch": 0.6496,
      "grad_norm": 1.2594757986222074,
      "kl": 0.12890625,
      "learning_rate": 4.443523755948401e-06,
      "loss": 0.0052,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 812
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 460.046875,
      "epoch": 0.6504,
      "grad_norm": 1.1971191096520515,
      "kl": 0.142578125,
      "learning_rate": 4.442205710442095e-06,
      "loss": 0.0057,
      "reward": 1.453125,
      "reward_std": 0.2877861559391022,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.984375,
      "step": 813
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 453.546875,
      "epoch": 0.6512,
      "grad_norm": 0.09472697828466468,
      "kl": 0.1083984375,
      "learning_rate": 4.4408863018210564e-06,
      "loss": 0.0043,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 814
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 432.6875,
      "epoch": 0.652,
      "grad_norm": 2.7013955744300415,
      "kl": 0.2412109375,
      "learning_rate": 4.439565531011299e-06,
      "loss": 0.0096,
      "reward": 1.671875,
      "reward_std": 0.46608567237854004,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.890625,
      "step": 815
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 365.703125,
      "epoch": 0.6528,
      "grad_norm": 1.44901882627503,
      "kl": 0.328125,
      "learning_rate": 4.43824339893979e-06,
      "loss": 0.0131,
      "reward": 1.640625,
      "reward_std": 0.2472364604473114,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.9375,
      "step": 816
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 485.84375,
      "epoch": 0.6536,
      "grad_norm": 3.6030275380795476,
      "kl": 0.431640625,
      "learning_rate": 4.436919906534452e-06,
      "loss": 0.0172,
      "reward": 1.515625,
      "reward_std": 0.4057711362838745,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.875,
      "step": 817
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 451.875,
      "epoch": 0.6544,
      "grad_norm": 3.9299901435845896,
      "kl": 0.72265625,
      "learning_rate": 4.4355950547241645e-06,
      "loss": 0.0289,
      "reward": 1.6875,
      "reward_std": 0.4538106322288513,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.90625,
      "step": 818
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 495.71875,
      "epoch": 0.6552,
      "grad_norm": 4.264532891042201,
      "kl": 169.0,
      "learning_rate": 4.434268844438758e-06,
      "loss": 6.7844,
      "reward": 1.265625,
      "reward_std": 0.4659822881221771,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.8125,
      "step": 819
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 416.625,
      "epoch": 0.656,
      "grad_norm": 10.862042633192596,
      "kl": 316.0,
      "learning_rate": 4.432941276609018e-06,
      "loss": 12.6175,
      "reward": 1.578125,
      "reward_std": 0.5162889957427979,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.890625,
      "step": 820
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 420.71875,
      "epoch": 0.6568,
      "grad_norm": 9.047241686718142,
      "kl": 248.0,
      "learning_rate": 4.431612352166684e-06,
      "loss": 9.9225,
      "reward": 1.4375,
      "reward_std": 0.5500767827033997,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.859375,
      "step": 821
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 518.984375,
      "epoch": 0.6576,
      "grad_norm": 12.760883469416578,
      "kl": 496.0,
      "learning_rate": 4.4302820720444454e-06,
      "loss": 19.7786,
      "reward": 1.21875,
      "reward_std": 0.7390350699424744,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.734375,
      "step": 822
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 437.6875,
      "epoch": 0.6584,
      "grad_norm": 72.80389942581843,
      "kl": 528.0,
      "learning_rate": 4.428950437175944e-06,
      "loss": 21.0454,
      "reward": 1.59375,
      "reward_std": 0.633817195892334,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.859375,
      "step": 823
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 397.421875,
      "epoch": 0.6592,
      "grad_norm": 3.522463920196615,
      "kl": 4.1875,
      "learning_rate": 4.427617448495772e-06,
      "loss": 0.1671,
      "reward": 1.421875,
      "reward_std": 0.589083194732666,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.84375,
      "step": 824
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 336.109375,
      "epoch": 0.66,
      "grad_norm": 1.248801220246186,
      "kl": 1.7109375,
      "learning_rate": 4.426283106939474e-06,
      "loss": 0.0683,
      "reward": 1.78125,
      "reward_std": 0.33225804567337036,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.921875,
      "step": 825
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.375,
      "completion_length": 411.671875,
      "epoch": 0.6608,
      "grad_norm": 5.90413161193254,
      "kl": 4.90625,
      "learning_rate": 4.424947413443539e-06,
      "loss": 0.196,
      "reward": 1.265625,
      "reward_std": 0.46100130677223206,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.890625,
      "step": 826
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 306.03125,
      "epoch": 0.6616,
      "grad_norm": 2.5172821672381973,
      "kl": 1.609375,
      "learning_rate": 4.423610368945411e-06,
      "loss": 0.0646,
      "reward": 1.5625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.9375,
      "step": 827
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.5,
      "completion_length": 382.203125,
      "epoch": 0.6624,
      "grad_norm": 4.785093259770447,
      "kl": 1.109375,
      "learning_rate": 4.422271974383479e-06,
      "loss": 0.0445,
      "reward": 1.265625,
      "reward_std": 0.31761831045150757,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 0.9375,
      "step": 828
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 323.203125,
      "epoch": 0.6632,
      "grad_norm": 2.3084817675871903,
      "kl": 0.9296875,
      "learning_rate": 4.420932230697079e-06,
      "loss": 0.0373,
      "reward": 1.40625,
      "reward_std": 0.2705550193786621,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.96875,
      "step": 829
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 330.15625,
      "epoch": 0.664,
      "grad_norm": 0.4928413062199495,
      "kl": 0.1650390625,
      "learning_rate": 4.419591138826495e-06,
      "loss": 0.0066,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 830
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 311.53125,
      "epoch": 0.6648,
      "grad_norm": 0.923167756488755,
      "kl": 0.1640625,
      "learning_rate": 4.418248699712955e-06,
      "loss": 0.0065,
      "reward": 1.703125,
      "reward_std": 0.23568855226039886,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 831
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 296.578125,
      "epoch": 0.6656,
      "grad_norm": 0.6476239243055839,
      "kl": 0.79296875,
      "learning_rate": 4.416904914298637e-06,
      "loss": 0.0317,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 832
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 322.5625,
      "epoch": 0.6664,
      "grad_norm": 0.5599649961332619,
      "kl": 0.1552734375,
      "learning_rate": 4.415559783526661e-06,
      "loss": 0.0062,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 833
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 297.875,
      "epoch": 0.6672,
      "grad_norm": 0.8868100745745057,
      "kl": 0.1611328125,
      "learning_rate": 4.414213308341092e-06,
      "loss": 0.0064,
      "reward": 1.75,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.96875,
      "step": 834
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 301.125,
      "epoch": 0.668,
      "grad_norm": 1.1584434850079184,
      "kl": 0.16796875,
      "learning_rate": 4.412865489686936e-06,
      "loss": 0.0067,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 835
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 275.0625,
      "epoch": 0.6688,
      "grad_norm": 0.8900494786989672,
      "kl": 0.1513671875,
      "learning_rate": 4.411516328510145e-06,
      "loss": 0.0061,
      "reward": 1.875,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 836
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 300.328125,
      "epoch": 0.6696,
      "grad_norm": 0.8115495134521837,
      "kl": 0.146484375,
      "learning_rate": 4.410165825757613e-06,
      "loss": 0.0059,
      "reward": 1.78125,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 837
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 330.328125,
      "epoch": 0.6704,
      "grad_norm": 0.896029068850484,
      "kl": 0.14453125,
      "learning_rate": 4.408813982377175e-06,
      "loss": 0.0058,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 838
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 314.421875,
      "epoch": 0.6712,
      "grad_norm": 0.5187846213444184,
      "kl": 0.1328125,
      "learning_rate": 4.407460799317605e-06,
      "loss": 0.0053,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 839
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 310.734375,
      "epoch": 0.672,
      "grad_norm": 1.1498330504928742,
      "kl": 0.86328125,
      "learning_rate": 4.40610627752862e-06,
      "loss": 0.0347,
      "reward": 1.53125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 840
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 347.21875,
      "epoch": 0.6728,
      "grad_norm": 0.7601751639149329,
      "kl": 0.1494140625,
      "learning_rate": 4.404750417960876e-06,
      "loss": 0.006,
      "reward": 1.546875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 841
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 351.421875,
      "epoch": 0.6736,
      "grad_norm": 0.734199264480647,
      "kl": 0.1650390625,
      "learning_rate": 4.403393221565966e-06,
      "loss": 0.0066,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 842
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 361.609375,
      "epoch": 0.6744,
      "grad_norm": 1.4623768306458316,
      "kl": 1.1328125,
      "learning_rate": 4.402034689296425e-06,
      "loss": 0.0453,
      "reward": 1.765625,
      "reward_std": 0.20873048901557922,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 843
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.625,
      "completion_length": 384.984375,
      "epoch": 0.6752,
      "grad_norm": 0.49041687670330963,
      "kl": 0.84375,
      "learning_rate": 4.400674822105721e-06,
      "loss": 0.0339,
      "reward": 1.359375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.984375,
      "step": 844
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 447.078125,
      "epoch": 0.676,
      "grad_norm": 3.8286635673292326,
      "kl": 2.78125,
      "learning_rate": 4.399313620948262e-06,
      "loss": 0.1114,
      "reward": 1.421875,
      "reward_std": 0.3187599778175354,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.953125,
      "step": 845
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 402.71875,
      "epoch": 0.6768,
      "grad_norm": 1.39392373494802,
      "kl": 0.5546875,
      "learning_rate": 4.397951086779392e-06,
      "loss": 0.0222,
      "reward": 1.484375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.96875,
      "step": 846
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 350.546875,
      "epoch": 0.6776,
      "grad_norm": 0.8760598270294438,
      "kl": 0.130859375,
      "learning_rate": 4.396587220555389e-06,
      "loss": 0.0052,
      "reward": 1.640625,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 847
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 416.59375,
      "epoch": 0.6784,
      "grad_norm": 0.8117302602451169,
      "kl": 0.6015625,
      "learning_rate": 4.395222023233467e-06,
      "loss": 0.0241,
      "reward": 1.796875,
      "reward_std": 0.22600007057189941,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 848
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 354.140625,
      "epoch": 0.6792,
      "grad_norm": 0.7893827765546036,
      "kl": 0.130859375,
      "learning_rate": 4.393855495771774e-06,
      "loss": 0.0053,
      "reward": 1.65625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 849
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 348.09375,
      "epoch": 0.68,
      "grad_norm": 1.1273192375292231,
      "kl": 0.1435546875,
      "learning_rate": 4.3924876391293915e-06,
      "loss": 0.0057,
      "reward": 1.75,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 850
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 386.171875,
      "epoch": 0.6808,
      "grad_norm": 1.6848210227745468,
      "kl": 0.134765625,
      "learning_rate": 4.391118454266335e-06,
      "loss": 0.0054,
      "reward": 1.65625,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 851
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 425.09375,
      "epoch": 0.6816,
      "grad_norm": 0.5722804310063825,
      "kl": 0.12060546875,
      "learning_rate": 4.389747942143549e-06,
      "loss": 0.0048,
      "reward": 1.71875,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 852
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 427.328125,
      "epoch": 0.6824,
      "grad_norm": 1.7251835603654744,
      "kl": 0.1201171875,
      "learning_rate": 4.388376103722914e-06,
      "loss": 0.0048,
      "reward": 1.671875,
      "reward_std": 0.22673699259757996,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 853
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 406.390625,
      "epoch": 0.6832,
      "grad_norm": 0.4098808078432349,
      "kl": 0.1318359375,
      "learning_rate": 4.387002939967237e-06,
      "loss": 0.0053,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 854
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 462.953125,
      "epoch": 0.684,
      "grad_norm": 1.8043388097299795,
      "kl": 0.353515625,
      "learning_rate": 4.38562845184026e-06,
      "loss": 0.0142,
      "reward": 1.828125,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 855
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 481.984375,
      "epoch": 0.6848,
      "grad_norm": 1.6393645313605443,
      "kl": 0.361328125,
      "learning_rate": 4.384252640306649e-06,
      "loss": 0.0144,
      "reward": 1.65625,
      "reward_std": 0.2041158527135849,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 856
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 461.46875,
      "epoch": 0.6856,
      "grad_norm": 2.221226937529482,
      "kl": 0.3828125,
      "learning_rate": 4.382875506332002e-06,
      "loss": 0.0153,
      "reward": 1.640625,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 857
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 357.734375,
      "epoch": 0.6864,
      "grad_norm": 0.47531247002388066,
      "kl": 0.1298828125,
      "learning_rate": 4.381497050882845e-06,
      "loss": 0.0052,
      "reward": 1.4375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 858
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 431.609375,
      "epoch": 0.6872,
      "grad_norm": 0.9210820085839544,
      "kl": 0.484375,
      "learning_rate": 4.380117274926632e-06,
      "loss": 0.0195,
      "reward": 1.703125,
      "reward_std": 0.17782479524612427,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 859
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 402.34375,
      "epoch": 0.688,
      "grad_norm": 1.3840191855749242,
      "kl": 1.078125,
      "learning_rate": 4.3787361794317405e-06,
      "loss": 0.0433,
      "reward": 1.703125,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 860
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 433.671875,
      "epoch": 0.6888,
      "grad_norm": 0.8237080651822265,
      "kl": 0.1220703125,
      "learning_rate": 4.377353765367479e-06,
      "loss": 0.0049,
      "reward": 1.578125,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 861
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 433.328125,
      "epoch": 0.6896,
      "grad_norm": 0.8167935126873793,
      "kl": 0.1240234375,
      "learning_rate": 4.375970033704078e-06,
      "loss": 0.005,
      "reward": 1.625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 862
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 430.328125,
      "epoch": 0.6904,
      "grad_norm": 0.6034458812927239,
      "kl": 0.11865234375,
      "learning_rate": 4.374584985412692e-06,
      "loss": 0.0048,
      "reward": 1.703125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 863
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 444.0625,
      "epoch": 0.6912,
      "grad_norm": 2.037211644377725,
      "kl": 2.96875,
      "learning_rate": 4.373198621465405e-06,
      "loss": 0.1192,
      "reward": 1.453125,
      "reward_std": 0.43403828144073486,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.9375,
      "step": 864
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 404.515625,
      "epoch": 0.692,
      "grad_norm": 0.942355756417357,
      "kl": 0.1162109375,
      "learning_rate": 4.3718109428352155e-06,
      "loss": 0.0046,
      "reward": 1.546875,
      "reward_std": 0.21778544783592224,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 865
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 480.65625,
      "epoch": 0.6928,
      "grad_norm": 0.819499283019901,
      "kl": 0.1455078125,
      "learning_rate": 4.370421950496055e-06,
      "loss": 0.0058,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 866
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 399.265625,
      "epoch": 0.6936,
      "grad_norm": 0.8418814238510208,
      "kl": 0.119140625,
      "learning_rate": 4.369031645422768e-06,
      "loss": 0.0048,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 867
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 429.265625,
      "epoch": 0.6944,
      "grad_norm": 0.6174620957043189,
      "kl": 0.7421875,
      "learning_rate": 4.367640028591126e-06,
      "loss": 0.0297,
      "reward": 1.546875,
      "reward_std": 0.09300297498703003,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 868
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 465.25,
      "epoch": 0.6952,
      "grad_norm": 0.5750532827732322,
      "kl": 0.1201171875,
      "learning_rate": 4.366247100977818e-06,
      "loss": 0.0048,
      "reward": 1.53125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 869
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 464.40625,
      "epoch": 0.696,
      "grad_norm": 1.7061165580059114,
      "kl": 2.65625,
      "learning_rate": 4.364852863560456e-06,
      "loss": 0.106,
      "reward": 1.796875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.953125,
      "step": 870
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 442.515625,
      "epoch": 0.6968,
      "grad_norm": 0.985137653664329,
      "kl": 1.71875,
      "learning_rate": 4.363457317317568e-06,
      "loss": 0.0687,
      "reward": 1.5625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 871
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 423.40625,
      "epoch": 0.6976,
      "grad_norm": 0.8717810235148079,
      "kl": 0.75,
      "learning_rate": 4.362060463228603e-06,
      "loss": 0.03,
      "reward": 1.78125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 872
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 397.8125,
      "epoch": 0.6984,
      "grad_norm": 1.8094845589703843,
      "kl": 1.296875,
      "learning_rate": 4.360662302273926e-06,
      "loss": 0.0517,
      "reward": 1.640625,
      "reward_std": 0.30296874046325684,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.953125,
      "step": 873
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 420.5625,
      "epoch": 0.6992,
      "grad_norm": 0.6926124145060691,
      "kl": 0.115234375,
      "learning_rate": 4.35926283543482e-06,
      "loss": 0.0046,
      "reward": 1.671875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 874
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 479.53125,
      "epoch": 0.7,
      "grad_norm": 3.0898708500563794,
      "kl": 0.94140625,
      "learning_rate": 4.357862063693486e-06,
      "loss": 0.0376,
      "reward": 1.875,
      "reward_std": 0.24359199404716492,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 875
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 431.890625,
      "epoch": 0.7008,
      "grad_norm": 0.6974488775760038,
      "kl": 0.1162109375,
      "learning_rate": 4.356459988033039e-06,
      "loss": 0.0046,
      "reward": 1.484375,
      "reward_std": 0.19408093392848969,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 876
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 444.078125,
      "epoch": 0.7016,
      "grad_norm": 3.037156735334447,
      "kl": 2.078125,
      "learning_rate": 4.355056609437509e-06,
      "loss": 0.0829,
      "reward": 1.5625,
      "reward_std": 0.21968847513198853,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.9375,
      "step": 877
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 455.84375,
      "epoch": 0.7024,
      "grad_norm": 0.40572337908482486,
      "kl": 0.119140625,
      "learning_rate": 4.353651928891842e-06,
      "loss": 0.0047,
      "reward": 1.9375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 878
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 442.0,
      "epoch": 0.7032,
      "grad_norm": 1.4903271791393706,
      "kl": 0.83984375,
      "learning_rate": 4.352245947381897e-06,
      "loss": 0.0336,
      "reward": 1.65625,
      "reward_std": 0.19918900728225708,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.96875,
      "step": 879
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 448.578125,
      "epoch": 0.704,
      "grad_norm": 2.0545542691137295,
      "kl": 0.37890625,
      "learning_rate": 4.3508386658944455e-06,
      "loss": 0.0152,
      "reward": 1.6875,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 880
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 447.203125,
      "epoch": 0.7048,
      "grad_norm": 1.0909881391793068,
      "kl": 0.3671875,
      "learning_rate": 4.349430085417171e-06,
      "loss": 0.0148,
      "reward": 1.796875,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 881
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 403.828125,
      "epoch": 0.7056,
      "grad_norm": 1.5553388649827689,
      "kl": 0.1064453125,
      "learning_rate": 4.348020206938672e-06,
      "loss": 0.0043,
      "reward": 1.8125,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 882
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 395.734375,
      "epoch": 0.7064,
      "grad_norm": 0.4552842879173502,
      "kl": 0.1142578125,
      "learning_rate": 4.3466090314484526e-06,
      "loss": 0.0046,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 883
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 417.0625,
      "epoch": 0.7072,
      "grad_norm": 0.3362833858620528,
      "kl": 0.1064453125,
      "learning_rate": 4.345196559936931e-06,
      "loss": 0.0043,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 884
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 396.265625,
      "epoch": 0.708,
      "grad_norm": 1.6585334236065075,
      "kl": 0.1875,
      "learning_rate": 4.343782793395435e-06,
      "loss": 0.0075,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.984375,
      "step": 885
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 490.6875,
      "epoch": 0.7088,
      "grad_norm": 0.5715755935163027,
      "kl": 0.1064453125,
      "learning_rate": 4.3423677328162e-06,
      "loss": 0.0043,
      "reward": 1.546875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 886
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 450.1875,
      "epoch": 0.7096,
      "grad_norm": 0.5989269164215718,
      "kl": 0.10693359375,
      "learning_rate": 4.340951379192369e-06,
      "loss": 0.0043,
      "reward": 1.71875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 887
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 447.09375,
      "epoch": 0.7104,
      "grad_norm": 0.340813035742099,
      "kl": 0.10205078125,
      "learning_rate": 4.3395337335179945e-06,
      "loss": 0.0041,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 888
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 420.78125,
      "epoch": 0.7112,
      "grad_norm": 0.4025425134963079,
      "kl": 0.1162109375,
      "learning_rate": 4.338114796788035e-06,
      "loss": 0.0047,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 889
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 371.484375,
      "epoch": 0.712,
      "grad_norm": 1.4984277973866036,
      "kl": 0.12109375,
      "learning_rate": 4.336694569998354e-06,
      "loss": 0.0048,
      "reward": 1.796875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 890
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 479.71875,
      "epoch": 0.7128,
      "grad_norm": 1.5903143900453378,
      "kl": 0.4765625,
      "learning_rate": 4.3352730541457215e-06,
      "loss": 0.0191,
      "reward": 1.921875,
      "reward_std": 0.18139132857322693,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.96875,
      "step": 891
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 487.484375,
      "epoch": 0.7136,
      "grad_norm": 1.040021665372247,
      "kl": 0.72265625,
      "learning_rate": 4.333850250227814e-06,
      "loss": 0.0289,
      "reward": 1.859375,
      "reward_std": 0.2414703369140625,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.96875,
      "step": 892
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 356.828125,
      "epoch": 0.7144,
      "grad_norm": 0.8145869982087229,
      "kl": 0.94140625,
      "learning_rate": 4.332426159243206e-06,
      "loss": 0.0377,
      "reward": 1.65625,
      "reward_std": 0.2041158527135849,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 893
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 485.140625,
      "epoch": 0.7152,
      "grad_norm": 1.097727906062679,
      "kl": 2.0,
      "learning_rate": 4.331000782191384e-06,
      "loss": 0.0801,
      "reward": 1.609375,
      "reward_std": 0.22673699259757996,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.953125,
      "step": 894
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 515.203125,
      "epoch": 0.716,
      "grad_norm": 3.025969830890034,
      "kl": 3.578125,
      "learning_rate": 4.329574120072728e-06,
      "loss": 0.1432,
      "reward": 1.6875,
      "reward_std": 0.3413130044937134,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.921875,
      "step": 895
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 549.359375,
      "epoch": 0.7168,
      "grad_norm": 0.7902604584398955,
      "kl": 2.96875,
      "learning_rate": 4.328146173888528e-06,
      "loss": 0.1187,
      "reward": 1.484375,
      "reward_std": 0.33669838309288025,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.921875,
      "step": 896
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 498.9375,
      "epoch": 0.7176,
      "grad_norm": 1.3940284999410812,
      "kl": 1.546875,
      "learning_rate": 4.32671694464097e-06,
      "loss": 0.062,
      "reward": 1.484375,
      "reward_std": 0.36355799436569214,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.921875,
      "step": 897
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 516.40625,
      "epoch": 0.7184,
      "grad_norm": 1.9310527726746134,
      "kl": 3.59375,
      "learning_rate": 4.3252864333331424e-06,
      "loss": 0.1439,
      "reward": 1.328125,
      "reward_std": 0.41361188888549805,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.875,
      "step": 898
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 480.234375,
      "epoch": 0.7192,
      "grad_norm": 4.6985717880758395,
      "kl": 2.84375,
      "learning_rate": 4.323854640969033e-06,
      "loss": 0.1137,
      "reward": 1.703125,
      "reward_std": 0.45561131834983826,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.921875,
      "step": 899
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 362.921875,
      "epoch": 0.72,
      "grad_norm": 0.7749778063374342,
      "kl": 0.76171875,
      "learning_rate": 4.322421568553529e-06,
      "loss": 0.0304,
      "reward": 1.5,
      "reward_std": 0.23356689512729645,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.984375,
      "step": 900
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 536.71875,
      "epoch": 0.7208,
      "grad_norm": 1.801671379190005,
      "kl": 3.0625,
      "learning_rate": 4.320987217092416e-06,
      "loss": 0.1229,
      "reward": 1.59375,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.9375,
      "step": 901
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 540.734375,
      "epoch": 0.7216,
      "grad_norm": 2.3304463660704235,
      "kl": 3.578125,
      "learning_rate": 4.319551587592377e-06,
      "loss": 0.1429,
      "reward": 1.515625,
      "reward_std": 0.40822190046310425,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.921875,
      "step": 902
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 470.0625,
      "epoch": 0.7224,
      "grad_norm": 1.7944162093678502,
      "kl": 4.46875,
      "learning_rate": 4.318114681060989e-06,
      "loss": 0.1795,
      "reward": 1.703125,
      "reward_std": 0.42508673667907715,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.90625,
      "step": 903
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 587.203125,
      "epoch": 0.7232,
      "grad_norm": 1.6222665192833228,
      "kl": 1.9375,
      "learning_rate": 4.316676498506735e-06,
      "loss": 0.0775,
      "reward": 1.53125,
      "reward_std": 0.40749478340148926,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.921875,
      "step": 904
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 505.015625,
      "epoch": 0.724,
      "grad_norm": 1.1143695214080347,
      "kl": 1.8671875,
      "learning_rate": 4.3152370409389795e-06,
      "loss": 0.0749,
      "reward": 1.53125,
      "reward_std": 0.3014557659626007,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.953125,
      "step": 905
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 489.875,
      "epoch": 0.7248,
      "grad_norm": 1.5889331546013732,
      "kl": 2.03125,
      "learning_rate": 4.3137963093679945e-06,
      "loss": 0.0814,
      "reward": 1.71875,
      "reward_std": 0.23689788579940796,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.9375,
      "step": 906
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 564.921875,
      "epoch": 0.7256,
      "grad_norm": 1.0491259327278086,
      "kl": 2.015625,
      "learning_rate": 4.3123543048049395e-06,
      "loss": 0.0804,
      "reward": 1.65625,
      "reward_std": 0.3010796010494232,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.921875,
      "step": 907
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 486.375,
      "epoch": 0.7264,
      "grad_norm": 1.1311641700877069,
      "kl": 1.2421875,
      "learning_rate": 4.310911028261867e-06,
      "loss": 0.0496,
      "reward": 1.8125,
      "reward_std": 0.24608495831489563,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.96875,
      "step": 908
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 495.1875,
      "epoch": 0.7272,
      "grad_norm": 1.0562690996200423,
      "kl": 0.5703125,
      "learning_rate": 4.309466480751726e-06,
      "loss": 0.0228,
      "reward": 1.78125,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 909
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 455.0625,
      "epoch": 0.728,
      "grad_norm": 1.5715659468664727,
      "kl": 1.7265625,
      "learning_rate": 4.308020663288356e-06,
      "loss": 0.0691,
      "reward": 1.578125,
      "reward_std": 0.24039676785469055,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.9375,
      "step": 910
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 532.40625,
      "epoch": 0.7288,
      "grad_norm": 0.6750795964090044,
      "kl": 0.10888671875,
      "learning_rate": 4.306573576886485e-06,
      "loss": 0.0044,
      "reward": 1.640625,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 911
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 434.671875,
      "epoch": 0.7296,
      "grad_norm": 0.6186247071856164,
      "kl": 0.11865234375,
      "learning_rate": 4.305125222561736e-06,
      "loss": 0.0047,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 912
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 469.046875,
      "epoch": 0.7304,
      "grad_norm": 1.228159699553349,
      "kl": 0.427734375,
      "learning_rate": 4.303675601330618e-06,
      "loss": 0.0171,
      "reward": 1.640625,
      "reward_std": 0.31983357667922974,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 913
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 470.859375,
      "epoch": 0.7312,
      "grad_norm": 1.3058863651653687,
      "kl": 0.63671875,
      "learning_rate": 4.302224714210532e-06,
      "loss": 0.0255,
      "reward": 1.671875,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 914
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 424.21875,
      "epoch": 0.732,
      "grad_norm": 0.6512093483515646,
      "kl": 0.1162109375,
      "learning_rate": 4.3007725622197675e-06,
      "loss": 0.0047,
      "reward": 1.84375,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 915
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 512.078125,
      "epoch": 0.7328,
      "grad_norm": 1.229335381648818,
      "kl": 2.65625,
      "learning_rate": 4.2993191463775e-06,
      "loss": 0.1065,
      "reward": 1.71875,
      "reward_std": 0.3868919014930725,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.921875,
      "step": 916
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 380.0625,
      "epoch": 0.7336,
      "grad_norm": 1.3564010145791259,
      "kl": 2.296875,
      "learning_rate": 4.29786446770379e-06,
      "loss": 0.092,
      "reward": 1.40625,
      "reward_std": 0.46144071221351624,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.921875,
      "step": 917
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 473.984375,
      "epoch": 0.7344,
      "grad_norm": 0.8627534279098009,
      "kl": 2.140625,
      "learning_rate": 4.296408527219592e-06,
      "loss": 0.0855,
      "reward": 1.5,
      "reward_std": 0.287577360868454,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.9375,
      "step": 918
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 469.734375,
      "epoch": 0.7352,
      "grad_norm": 0.7801019710985146,
      "kl": 2.46875,
      "learning_rate": 4.294951325946737e-06,
      "loss": 0.0988,
      "reward": 1.34375,
      "reward_std": 0.40489843487739563,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.9375,
      "step": 919
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 403.359375,
      "epoch": 0.736,
      "grad_norm": 1.5718654200701025,
      "kl": 0.828125,
      "learning_rate": 4.293492864907947e-06,
      "loss": 0.0332,
      "reward": 1.609375,
      "reward_std": 0.34564992785453796,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 920
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 534.484375,
      "epoch": 0.7368,
      "grad_norm": 2.161536986347826,
      "kl": 2.546875,
      "learning_rate": 4.2920331451268246e-06,
      "loss": 0.102,
      "reward": 1.515625,
      "reward_std": 0.5058395266532898,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.890625,
      "step": 921
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 512.09375,
      "epoch": 0.7376,
      "grad_norm": 1.7295118963451739,
      "kl": 3.421875,
      "learning_rate": 4.2905721676278585e-06,
      "loss": 0.1369,
      "reward": 1.3125,
      "reward_std": 0.3742702603340149,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.90625,
      "step": 922
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 519.265625,
      "epoch": 0.7384,
      "grad_norm": 2.1540118335096063,
      "kl": 4.3125,
      "learning_rate": 4.28910993343642e-06,
      "loss": 0.1727,
      "reward": 1.375,
      "reward_std": 0.5602303147315979,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.875,
      "step": 923
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 450.78125,
      "epoch": 0.7392,
      "grad_norm": 0.9790998139016089,
      "kl": 2.421875,
      "learning_rate": 4.2876464435787576e-06,
      "loss": 0.0965,
      "reward": 1.53125,
      "reward_std": 0.3380831182003021,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.9375,
      "step": 924
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.5,
      "completion_length": 494.484375,
      "epoch": 0.74,
      "grad_norm": 2.399007793907817,
      "kl": 2.25,
      "learning_rate": 4.286181699082008e-06,
      "loss": 0.0899,
      "reward": 1.359375,
      "reward_std": 0.26196980476379395,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.921875,
      "step": 925
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 385.15625,
      "epoch": 0.7408,
      "grad_norm": 1.1416933071007234,
      "kl": 0.93359375,
      "learning_rate": 4.284715700974186e-06,
      "loss": 0.0373,
      "reward": 1.578125,
      "reward_std": 0.2971188426017761,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.96875,
      "step": 926
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 521.203125,
      "epoch": 0.7416,
      "grad_norm": 9.194586955494502,
      "kl": 260.0,
      "learning_rate": 4.283248450284182e-06,
      "loss": 10.4072,
      "reward": 1.390625,
      "reward_std": 0.5807557106018066,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.796875,
      "step": 927
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 553.21875,
      "epoch": 0.7424,
      "grad_norm": 4.26876911233488,
      "kl": 7.28125,
      "learning_rate": 4.281779948041772e-06,
      "loss": 0.2909,
      "reward": 1.265625,
      "reward_std": 0.6378883123397827,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.765625,
      "step": 928
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 538.5,
      "epoch": 0.7432,
      "grad_norm": 2.353768064399416,
      "kl": 5.875,
      "learning_rate": 4.280310195277606e-06,
      "loss": 0.2358,
      "reward": 1.265625,
      "reward_std": 0.48670461773872375,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.765625,
      "step": 929
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 400.03125,
      "epoch": 0.744,
      "grad_norm": 1.5584445115662269,
      "kl": 0.66796875,
      "learning_rate": 4.278839193023214e-06,
      "loss": 0.0267,
      "reward": 1.5625,
      "reward_std": 0.29176726937294006,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.96875,
      "step": 930
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 397.390625,
      "epoch": 0.7448,
      "grad_norm": 0.6210216856962318,
      "kl": 0.1171875,
      "learning_rate": 4.277366942311001e-06,
      "loss": 0.0047,
      "reward": 1.59375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 931
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 366.84375,
      "epoch": 0.7456,
      "grad_norm": 0.30790529190005606,
      "kl": 0.11474609375,
      "learning_rate": 4.2758934441742494e-06,
      "loss": 0.0046,
      "reward": 1.9375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 932
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 412.296875,
      "epoch": 0.7464,
      "grad_norm": 0.6533715565725364,
      "kl": 0.1171875,
      "learning_rate": 4.274418699647117e-06,
      "loss": 0.0047,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 933
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.625,
      "completion_length": 347.0,
      "epoch": 0.7472,
      "grad_norm": 0.6429134637471882,
      "kl": 0.1220703125,
      "learning_rate": 4.272942709764638e-06,
      "loss": 0.0049,
      "reward": 1.140625,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.140625,
      "rewards/format_reward": 1.0,
      "step": 934
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 409.3125,
      "epoch": 0.748,
      "grad_norm": 0.7219361422305827,
      "kl": 0.150390625,
      "learning_rate": 4.271465475562716e-06,
      "loss": 0.006,
      "reward": 1.765625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.96875,
      "step": 935
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 385.421875,
      "epoch": 0.7488,
      "grad_norm": 0.9652043265766445,
      "kl": 0.12060546875,
      "learning_rate": 4.269986998078132e-06,
      "loss": 0.0048,
      "reward": 1.6875,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 936
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 439.1875,
      "epoch": 0.7496,
      "grad_norm": 1.0733356267908236,
      "kl": 0.1171875,
      "learning_rate": 4.268507278348539e-06,
      "loss": 0.0047,
      "reward": 1.71875,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 937
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 416.90625,
      "epoch": 0.7504,
      "grad_norm": 0.37362274771795734,
      "kl": 0.10595703125,
      "learning_rate": 4.2670263174124615e-06,
      "loss": 0.0042,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 938
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 404.390625,
      "epoch": 0.7512,
      "grad_norm": 0.09190962754036436,
      "kl": 0.11376953125,
      "learning_rate": 4.265544116309294e-06,
      "loss": 0.0045,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 939
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 410.171875,
      "epoch": 0.752,
      "grad_norm": 0.9576930951379631,
      "kl": 0.1259765625,
      "learning_rate": 4.264060676079302e-06,
      "loss": 0.0051,
      "reward": 1.53125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 940
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 423.234375,
      "epoch": 0.7528,
      "grad_norm": 0.3486492469895735,
      "kl": 0.11474609375,
      "learning_rate": 4.262575997763622e-06,
      "loss": 0.0046,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 941
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 411.390625,
      "epoch": 0.7536,
      "grad_norm": 0.42797705325285396,
      "kl": 0.11669921875,
      "learning_rate": 4.2610900824042575e-06,
      "loss": 0.0047,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 942
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 411.640625,
      "epoch": 0.7544,
      "grad_norm": 0.5371471332126149,
      "kl": 0.10791015625,
      "learning_rate": 4.2596029310440826e-06,
      "loss": 0.0043,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 943
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 386.515625,
      "epoch": 0.7552,
      "grad_norm": 0.516151719265387,
      "kl": 0.11328125,
      "learning_rate": 4.258114544726835e-06,
      "loss": 0.0045,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 944
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 396.53125,
      "epoch": 0.756,
      "grad_norm": 1.6891573571293064,
      "kl": 0.123046875,
      "learning_rate": 4.256624924497124e-06,
      "loss": 0.0049,
      "reward": 1.640625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 945
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 392.703125,
      "epoch": 0.7568,
      "grad_norm": 1.0182792334186193,
      "kl": 0.12109375,
      "learning_rate": 4.25513407140042e-06,
      "loss": 0.0048,
      "reward": 1.71875,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 946
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 400.703125,
      "epoch": 0.7576,
      "grad_norm": 0.420747136044565,
      "kl": 0.11669921875,
      "learning_rate": 4.253641986483063e-06,
      "loss": 0.0047,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 947
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 386.0,
      "epoch": 0.7584,
      "grad_norm": 0.9703799860470895,
      "kl": 0.1572265625,
      "learning_rate": 4.2521486707922545e-06,
      "loss": 0.0063,
      "reward": 1.5625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 948
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 414.265625,
      "epoch": 0.7592,
      "grad_norm": 0.5790431184943604,
      "kl": 0.1171875,
      "learning_rate": 4.250654125376062e-06,
      "loss": 0.0047,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 949
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 356.109375,
      "epoch": 0.76,
      "grad_norm": 0.784741344714855,
      "kl": 0.130859375,
      "learning_rate": 4.249158351283414e-06,
      "loss": 0.0052,
      "reward": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 950
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 365.96875,
      "epoch": 0.7608,
      "grad_norm": 0.3682297319192504,
      "kl": 0.12060546875,
      "learning_rate": 4.247661349564103e-06,
      "loss": 0.0048,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 951
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 425.4375,
      "epoch": 0.7616,
      "grad_norm": 0.542369647139902,
      "kl": 0.1181640625,
      "learning_rate": 4.246163121268782e-06,
      "loss": 0.0047,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 952
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 356.765625,
      "epoch": 0.7624,
      "grad_norm": 0.6686418492448905,
      "kl": 0.126953125,
      "learning_rate": 4.244663667448965e-06,
      "loss": 0.0051,
      "reward": 1.796875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 953
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 465.984375,
      "epoch": 0.7632,
      "grad_norm": 0.5241841620615234,
      "kl": 0.1259765625,
      "learning_rate": 4.243162989157027e-06,
      "loss": 0.005,
      "reward": 1.484375,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.984375,
      "step": 954
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 414.03125,
      "epoch": 0.764,
      "grad_norm": 0.6609924834357154,
      "kl": 0.138671875,
      "learning_rate": 4.241661087446202e-06,
      "loss": 0.0055,
      "reward": 1.765625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 955
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.375,
      "completion_length": 418.1875,
      "epoch": 0.7648,
      "grad_norm": 2.635363041629432,
      "kl": 0.2216796875,
      "learning_rate": 4.240157963370583e-06,
      "loss": 0.0089,
      "reward": 1.296875,
      "reward_std": 0.4558286964893341,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.921875,
      "step": 956
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 379.171875,
      "epoch": 0.7656,
      "grad_norm": 1.4448050764201095,
      "kl": 0.1611328125,
      "learning_rate": 4.2386536179851175e-06,
      "loss": 0.0064,
      "reward": 1.5625,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 957
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 446.421875,
      "epoch": 0.7664,
      "grad_norm": 1.000212044079561,
      "kl": 0.12451171875,
      "learning_rate": 4.2371480523456156e-06,
      "loss": 0.005,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 958
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 448.609375,
      "epoch": 0.7672,
      "grad_norm": 0.8981210410115046,
      "kl": 0.1298828125,
      "learning_rate": 4.235641267508741e-06,
      "loss": 0.0052,
      "reward": 1.78125,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 959
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 380.765625,
      "epoch": 0.768,
      "grad_norm": 0.8232337759805564,
      "kl": 0.12890625,
      "learning_rate": 4.234133264532012e-06,
      "loss": 0.0052,
      "reward": 1.484375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 960
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 417.203125,
      "epoch": 0.7688,
      "grad_norm": 5.143963317662924,
      "kl": 0.29296875,
      "learning_rate": 4.232624044473805e-06,
      "loss": 0.0117,
      "reward": 1.71875,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.953125,
      "step": 961
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 389.5,
      "epoch": 0.7696,
      "grad_norm": 1.1019180106834146,
      "kl": 0.2197265625,
      "learning_rate": 4.231113608393348e-06,
      "loss": 0.0088,
      "reward": 1.515625,
      "reward_std": 0.23925507068634033,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 962
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 409.78125,
      "epoch": 0.7704,
      "grad_norm": 1.8150596559083683,
      "kl": 0.353515625,
      "learning_rate": 4.229601957350722e-06,
      "loss": 0.0142,
      "reward": 1.703125,
      "reward_std": 0.20189079642295837,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 963
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 429.234375,
      "epoch": 0.7712,
      "grad_norm": 0.5328641427292242,
      "kl": 0.1513671875,
      "learning_rate": 4.228089092406863e-06,
      "loss": 0.0061,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 964
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 475.234375,
      "epoch": 0.772,
      "grad_norm": 0.7284892417847438,
      "kl": 0.61328125,
      "learning_rate": 4.226575014623557e-06,
      "loss": 0.0245,
      "reward": 1.671875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.953125,
      "step": 965
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 409.828125,
      "epoch": 0.7728,
      "grad_norm": 2.672494169486163,
      "kl": 0.83984375,
      "learning_rate": 4.225059725063444e-06,
      "loss": 0.0336,
      "reward": 1.65625,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.953125,
      "step": 966
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 425.15625,
      "epoch": 0.7736,
      "grad_norm": 0.8381645589499661,
      "kl": 0.138671875,
      "learning_rate": 4.22354322479001e-06,
      "loss": 0.0056,
      "reward": 1.59375,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 967
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 388.28125,
      "epoch": 0.7744,
      "grad_norm": 1.5610358286513286,
      "kl": 0.2060546875,
      "learning_rate": 4.222025514867596e-06,
      "loss": 0.0083,
      "reward": 1.671875,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 968
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 349.265625,
      "epoch": 0.7752,
      "grad_norm": 1.3621580253801306,
      "kl": 0.16015625,
      "learning_rate": 4.220506596361387e-06,
      "loss": 0.0064,
      "reward": 1.453125,
      "reward_std": 0.35612428188323975,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 969
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 420.953125,
      "epoch": 0.776,
      "grad_norm": 0.7268537039641286,
      "kl": 0.130859375,
      "learning_rate": 4.218986470337419e-06,
      "loss": 0.0052,
      "reward": 1.828125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 970
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 419.234375,
      "epoch": 0.7768,
      "grad_norm": 0.6318043018805877,
      "kl": 0.1298828125,
      "learning_rate": 4.217465137862575e-06,
      "loss": 0.0052,
      "reward": 1.703125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 971
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 413.625,
      "epoch": 0.7776,
      "grad_norm": 0.5299685435492659,
      "kl": 0.12158203125,
      "learning_rate": 4.215942600004586e-06,
      "loss": 0.0049,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 972
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 396.90625,
      "epoch": 0.7784,
      "grad_norm": 0.1458544807785085,
      "kl": 0.140625,
      "learning_rate": 4.214418857832025e-06,
      "loss": 0.0056,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 973
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 355.703125,
      "epoch": 0.7792,
      "grad_norm": 0.9539917998314951,
      "kl": 0.13671875,
      "learning_rate": 4.212893912414316e-06,
      "loss": 0.0055,
      "reward": 1.6875,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 974
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.625,
      "completion_length": 399.453125,
      "epoch": 0.78,
      "grad_norm": 1.7882583607879647,
      "kl": 0.158203125,
      "learning_rate": 4.211367764821722e-06,
      "loss": 0.0063,
      "reward": 1.15625,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.171875,
      "rewards/format_reward": 0.984375,
      "step": 975
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 420.9375,
      "epoch": 0.7808,
      "grad_norm": 1.8457448576517015,
      "kl": 1.578125,
      "learning_rate": 4.209840416125353e-06,
      "loss": 0.0633,
      "reward": 1.515625,
      "reward_std": 0.3847702443599701,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.90625,
      "step": 976
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 450.15625,
      "epoch": 0.7816,
      "grad_norm": 0.8375536084250883,
      "kl": 1.859375,
      "learning_rate": 4.208311867397162e-06,
      "loss": 0.0743,
      "reward": 1.65625,
      "reward_std": 0.3535533845424652,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.9375,
      "step": 977
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 496.421875,
      "epoch": 0.7824,
      "grad_norm": 2.695559553833614,
      "kl": 5.46875,
      "learning_rate": 4.206782119709942e-06,
      "loss": 0.2185,
      "reward": 1.46875,
      "reward_std": 0.4887067973613739,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.78125,
      "step": 978
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 399.75,
      "epoch": 0.7832,
      "grad_norm": 0.6598807469508163,
      "kl": 0.8828125,
      "learning_rate": 4.205251174137329e-06,
      "loss": 0.0354,
      "reward": 1.484375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.96875,
      "step": 979
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 420.53125,
      "epoch": 0.784,
      "grad_norm": 0.9743811025961808,
      "kl": 1.2734375,
      "learning_rate": 4.2037190317538e-06,
      "loss": 0.0508,
      "reward": 1.65625,
      "reward_std": 0.43191659450531006,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.921875,
      "step": 980
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 413.984375,
      "epoch": 0.7848,
      "grad_norm": 0.842145948104087,
      "kl": 0.296875,
      "learning_rate": 4.202185693634671e-06,
      "loss": 0.0119,
      "reward": 1.71875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 981
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.5,
      "completion_length": 419.078125,
      "epoch": 0.7856,
      "grad_norm": 1.3977982611011994,
      "kl": 1.5546875,
      "learning_rate": 4.200651160856099e-06,
      "loss": 0.062,
      "reward": 1.296875,
      "reward_std": 0.33669838309288025,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.921875,
      "step": 982
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 473.46875,
      "epoch": 0.7864,
      "grad_norm": 0.28050828036992015,
      "kl": 0.5859375,
      "learning_rate": 4.1991154344950755e-06,
      "loss": 0.0234,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 983
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 403.359375,
      "epoch": 0.7872,
      "grad_norm": 0.9636221712471047,
      "kl": 0.390625,
      "learning_rate": 4.197578515629435e-06,
      "loss": 0.0156,
      "reward": 1.4375,
      "reward_std": 0.2925041913986206,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.984375,
      "step": 984
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 460.03125,
      "epoch": 0.788,
      "grad_norm": 0.342473944432868,
      "kl": 0.271484375,
      "learning_rate": 4.196040405337846e-06,
      "loss": 0.0109,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 985
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 364.203125,
      "epoch": 0.7888,
      "grad_norm": 1.7808786921240929,
      "kl": 0.9609375,
      "learning_rate": 4.194501104699813e-06,
      "loss": 0.0385,
      "reward": 1.59375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.96875,
      "step": 986
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 420.75,
      "epoch": 0.7896,
      "grad_norm": 0.8194480679343348,
      "kl": 0.71875,
      "learning_rate": 4.192960614795676e-06,
      "loss": 0.0287,
      "reward": 1.640625,
      "reward_std": 0.2773849070072174,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.96875,
      "step": 987
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 361.515625,
      "epoch": 0.7904,
      "grad_norm": 0.9579962256217175,
      "kl": 0.1337890625,
      "learning_rate": 4.19141893670661e-06,
      "loss": 0.0053,
      "reward": 1.453125,
      "reward_std": 0.23568853735923767,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 988
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 387.6875,
      "epoch": 0.7912,
      "grad_norm": 0.9229492984968124,
      "kl": 0.9765625,
      "learning_rate": 4.189876071514624e-06,
      "loss": 0.0392,
      "reward": 1.640625,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 989
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 396.078125,
      "epoch": 0.792,
      "grad_norm": 0.9196193490960514,
      "kl": 0.126953125,
      "learning_rate": 4.188332020302561e-06,
      "loss": 0.0051,
      "reward": 1.734375,
      "reward_std": 0.3051002323627472,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 990
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 373.609375,
      "epoch": 0.7928,
      "grad_norm": 0.4291672668439854,
      "kl": 0.123046875,
      "learning_rate": 4.186786784154096e-06,
      "loss": 0.0049,
      "reward": 1.46875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 991
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 401.03125,
      "epoch": 0.7936,
      "grad_norm": 0.9173605611044626,
      "kl": 0.58203125,
      "learning_rate": 4.1852403641537344e-06,
      "loss": 0.0233,
      "reward": 1.578125,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.96875,
      "step": 992
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 421.625,
      "epoch": 0.7944,
      "grad_norm": 0.8235714182946214,
      "kl": 0.48828125,
      "learning_rate": 4.183692761386813e-06,
      "loss": 0.0196,
      "reward": 1.5625,
      "reward_std": 0.26409146189689636,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 993
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.5,
      "completion_length": 401.5625,
      "epoch": 0.7952,
      "grad_norm": 0.5304761774438076,
      "kl": 0.12890625,
      "learning_rate": 4.1821439769395e-06,
      "loss": 0.0052,
      "reward": 1.390625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 994
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 387.140625,
      "epoch": 0.796,
      "grad_norm": 0.6290673111443011,
      "kl": 0.42578125,
      "learning_rate": 4.180594011898791e-06,
      "loss": 0.0171,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 995
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 351.84375,
      "epoch": 0.7968,
      "grad_norm": 0.6991693797366741,
      "kl": 0.1318359375,
      "learning_rate": 4.1790428673525104e-06,
      "loss": 0.0053,
      "reward": 1.8125,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 996
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 460.28125,
      "epoch": 0.7976,
      "grad_norm": 2.638023572486133,
      "kl": 0.8046875,
      "learning_rate": 4.177490544389313e-06,
      "loss": 0.0322,
      "reward": 1.578125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.96875,
      "step": 997
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 407.90625,
      "epoch": 0.7984,
      "grad_norm": 1.7128764247694277,
      "kl": 0.11962890625,
      "learning_rate": 4.175937044098678e-06,
      "loss": 0.0048,
      "reward": 1.59375,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 998
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 327.734375,
      "epoch": 0.7992,
      "grad_norm": 1.0275671648896874,
      "kl": 0.1484375,
      "learning_rate": 4.1743823675709115e-06,
      "loss": 0.0059,
      "reward": 1.46875,
      "reward_std": 0.3119301199913025,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 999
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 361.578125,
      "epoch": 0.8,
      "grad_norm": 0.9927971728000622,
      "kl": 0.89453125,
      "learning_rate": 4.172826515897146e-06,
      "loss": 0.0358,
      "reward": 1.4375,
      "reward_std": 0.26409146189689636,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.96875,
      "step": 1000
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 432.828125,
      "epoch": 0.8008,
      "grad_norm": 0.3048605528477739,
      "kl": 0.12451171875,
      "learning_rate": 4.171269490169337e-06,
      "loss": 0.005,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1001
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.5,
      "completion_length": 361.703125,
      "epoch": 0.8016,
      "grad_norm": 0.09732700080514368,
      "kl": 0.1240234375,
      "learning_rate": 4.1697112914802665e-06,
      "loss": 0.005,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1002
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 383.140625,
      "epoch": 0.8024,
      "grad_norm": 0.6311599518634863,
      "kl": 0.376953125,
      "learning_rate": 4.168151920923536e-06,
      "loss": 0.015,
      "reward": 1.609375,
      "reward_std": 0.17782479524612427,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1003
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 447.625,
      "epoch": 0.8032,
      "grad_norm": 0.7021377784094579,
      "kl": 1.5078125,
      "learning_rate": 4.1665913795935755e-06,
      "loss": 0.0604,
      "reward": 1.5,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 1004
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 393.5,
      "epoch": 0.804,
      "grad_norm": 1.171767142404209,
      "kl": 0.74609375,
      "learning_rate": 4.16502966858563e-06,
      "loss": 0.0298,
      "reward": 1.59375,
      "reward_std": 0.3093337416648865,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 1005
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 417.84375,
      "epoch": 0.8048,
      "grad_norm": 0.32791847154584824,
      "kl": 0.119140625,
      "learning_rate": 4.163466788995768e-06,
      "loss": 0.0048,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1006
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 482.203125,
      "epoch": 0.8056,
      "grad_norm": 0.8809647797227202,
      "kl": 1.1640625,
      "learning_rate": 4.161902741920881e-06,
      "loss": 0.0465,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.96875,
      "step": 1007
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 408.09375,
      "epoch": 0.8064,
      "grad_norm": 0.9255991702128258,
      "kl": 1.375,
      "learning_rate": 4.160337528458676e-06,
      "loss": 0.055,
      "reward": 1.78125,
      "reward_std": 0.3014557659626007,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.953125,
      "step": 1008
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 463.078125,
      "epoch": 0.8072,
      "grad_norm": 0.8992639035654059,
      "kl": 1.609375,
      "learning_rate": 4.15877114970768e-06,
      "loss": 0.0642,
      "reward": 1.78125,
      "reward_std": 0.28566449880599976,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 1009
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 450.265625,
      "epoch": 0.808,
      "grad_norm": 1.119342046783512,
      "kl": 1.875,
      "learning_rate": 4.1572036067672386e-06,
      "loss": 0.0749,
      "reward": 1.75,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.96875,
      "step": 1010
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 434.8125,
      "epoch": 0.8088,
      "grad_norm": 0.9183180237329227,
      "kl": 0.130859375,
      "learning_rate": 4.155634900737513e-06,
      "loss": 0.0052,
      "reward": 1.796875,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1011
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 396.03125,
      "epoch": 0.8096,
      "grad_norm": 1.332965526658357,
      "kl": 1.265625,
      "learning_rate": 4.154065032719482e-06,
      "loss": 0.0508,
      "reward": 1.421875,
      "reward_std": 0.30617380142211914,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.96875,
      "step": 1012
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 421.046875,
      "epoch": 0.8104,
      "grad_norm": 1.1436703759975864,
      "kl": 1.2578125,
      "learning_rate": 4.152494003814939e-06,
      "loss": 0.0503,
      "reward": 1.53125,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.953125,
      "step": 1013
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 416.796875,
      "epoch": 0.8112,
      "grad_norm": 6.194941679937528,
      "kl": 0.1201171875,
      "learning_rate": 4.150921815126493e-06,
      "loss": 0.0048,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 1014
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 394.640625,
      "epoch": 0.812,
      "grad_norm": 0.736162386203766,
      "kl": 0.224609375,
      "learning_rate": 4.149348467757566e-06,
      "loss": 0.009,
      "reward": 1.828125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1015
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 414.03125,
      "epoch": 0.8128,
      "grad_norm": 1.2415159824198114,
      "kl": 0.2001953125,
      "learning_rate": 4.147773962812393e-06,
      "loss": 0.008,
      "reward": 1.59375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 1016
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 405.625,
      "epoch": 0.8136,
      "grad_norm": 1.160255873423496,
      "kl": 0.5390625,
      "learning_rate": 4.146198301396025e-06,
      "loss": 0.0217,
      "reward": 1.40625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.984375,
      "step": 1017
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 427.015625,
      "epoch": 0.8144,
      "grad_norm": 0.8885955478672967,
      "kl": 1.7734375,
      "learning_rate": 4.14462148461432e-06,
      "loss": 0.0709,
      "reward": 1.828125,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.953125,
      "step": 1018
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 389.96875,
      "epoch": 0.8152,
      "grad_norm": 0.9742868988911085,
      "kl": 0.3046875,
      "learning_rate": 4.143043513573949e-06,
      "loss": 0.0122,
      "reward": 1.875,
      "reward_std": 0.21556037664413452,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.984375,
      "step": 1019
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 439.859375,
      "epoch": 0.816,
      "grad_norm": 0.9084803679903584,
      "kl": 1.484375,
      "learning_rate": 4.141464389382392e-06,
      "loss": 0.0593,
      "reward": 1.46875,
      "reward_std": 0.213067427277565,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 1020
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 423.1875,
      "epoch": 0.8168,
      "grad_norm": 1.095865842968603,
      "kl": 0.82421875,
      "learning_rate": 4.13988411314794e-06,
      "loss": 0.0329,
      "reward": 1.421875,
      "reward_std": 0.3298586905002594,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.96875,
      "step": 1021
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 550.21875,
      "epoch": 0.8176,
      "grad_norm": 0.9089747180881477,
      "kl": 0.984375,
      "learning_rate": 4.13830268597969e-06,
      "loss": 0.0391,
      "reward": 1.515625,
      "reward_std": 0.2773849070072174,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.96875,
      "step": 1022
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 416.953125,
      "epoch": 0.8184,
      "grad_norm": 0.8960698067226799,
      "kl": 0.34375,
      "learning_rate": 4.136720108987552e-06,
      "loss": 0.0138,
      "reward": 1.625,
      "reward_std": 0.3014557659626007,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 1023
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 450.75,
      "epoch": 0.8192,
      "grad_norm": 1.0249080707929057,
      "kl": 0.5,
      "learning_rate": 4.1351363832822364e-06,
      "loss": 0.0201,
      "reward": 1.5,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 1024
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 372.90625,
      "epoch": 0.82,
      "grad_norm": 0.851722960815373,
      "kl": 0.7265625,
      "learning_rate": 4.133551509975264e-06,
      "loss": 0.0291,
      "reward": 1.546875,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 1025
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 523.890625,
      "epoch": 0.8208,
      "grad_norm": 1.5004102169250988,
      "kl": 2.90625,
      "learning_rate": 4.13196549017896e-06,
      "loss": 0.1165,
      "reward": 1.75,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.9375,
      "step": 1026
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 437.0,
      "epoch": 0.8216,
      "grad_norm": 1.7522783445835257,
      "kl": 1.4296875,
      "learning_rate": 4.130378325006453e-06,
      "loss": 0.0573,
      "reward": 1.4375,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.96875,
      "step": 1027
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 464.09375,
      "epoch": 0.8224,
      "grad_norm": 1.2636399601831858,
      "kl": 1.109375,
      "learning_rate": 4.128790015571679e-06,
      "loss": 0.0445,
      "reward": 1.46875,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 1028
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 447.28125,
      "epoch": 0.8232,
      "grad_norm": 0.665300889424532,
      "kl": 1.5390625,
      "learning_rate": 4.127200562989372e-06,
      "loss": 0.0618,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.953125,
      "step": 1029
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 413.09375,
      "epoch": 0.824,
      "grad_norm": 0.49619710604110157,
      "kl": 0.78515625,
      "learning_rate": 4.125609968375073e-06,
      "loss": 0.0314,
      "reward": 1.734375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1030
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 392.21875,
      "epoch": 0.8248,
      "grad_norm": 0.8067248471828645,
      "kl": 0.671875,
      "learning_rate": 4.12401823284512e-06,
      "loss": 0.0267,
      "reward": 1.578125,
      "reward_std": 0.2867125868797302,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 1031
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 434.671875,
      "epoch": 0.8256,
      "grad_norm": 0.07466081667756266,
      "kl": 0.10302734375,
      "learning_rate": 4.122425357516658e-06,
      "loss": 0.0041,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1032
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 406.875,
      "epoch": 0.8264,
      "grad_norm": 0.8089643320442237,
      "kl": 1.1171875,
      "learning_rate": 4.1208313435076255e-06,
      "loss": 0.0445,
      "reward": 1.40625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.96875,
      "step": 1033
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 449.1875,
      "epoch": 0.8272,
      "grad_norm": 0.42636839515317615,
      "kl": 0.64453125,
      "learning_rate": 4.119236191936764e-06,
      "loss": 0.0258,
      "reward": 1.453125,
      "reward_std": 0.09300297498703003,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.984375,
      "step": 1034
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 400.84375,
      "epoch": 0.828,
      "grad_norm": 0.6555441663508101,
      "kl": 0.251953125,
      "learning_rate": 4.117639903923611e-06,
      "loss": 0.0101,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1035
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 412.21875,
      "epoch": 0.8288,
      "grad_norm": 3.028610298530577,
      "kl": 0.2578125,
      "learning_rate": 4.116042480588505e-06,
      "loss": 0.0103,
      "reward": 1.578125,
      "reward_std": 0.2414703369140625,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 1036
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 451.34375,
      "epoch": 0.8296,
      "grad_norm": 0.7112208658020002,
      "kl": 1.1875,
      "learning_rate": 4.114443923052577e-06,
      "loss": 0.0475,
      "reward": 1.484375,
      "reward_std": 0.18133097887039185,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.96875,
      "step": 1037
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 456.03125,
      "epoch": 0.8304,
      "grad_norm": 0.9702026427863657,
      "kl": 0.10986328125,
      "learning_rate": 4.112844232437757e-06,
      "loss": 0.0044,
      "reward": 1.84375,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1038
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 420.0,
      "epoch": 0.8312,
      "grad_norm": 1.0720073881012364,
      "kl": 1.0703125,
      "learning_rate": 4.11124340986677e-06,
      "loss": 0.0427,
      "reward": 1.578125,
      "reward_std": 0.4240131676197052,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.96875,
      "step": 1039
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 417.703125,
      "epoch": 0.832,
      "grad_norm": 1.2995545112867135,
      "kl": 0.58203125,
      "learning_rate": 4.109641456463135e-06,
      "loss": 0.0233,
      "reward": 1.625,
      "reward_std": 0.31290027499198914,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 1040
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 442.75,
      "epoch": 0.8328,
      "grad_norm": 0.6877738385337426,
      "kl": 0.2080078125,
      "learning_rate": 4.108038373351163e-06,
      "loss": 0.0083,
      "reward": 1.4375,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 1041
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 448.5625,
      "epoch": 0.8336,
      "grad_norm": 0.7300017276955302,
      "kl": 0.24609375,
      "learning_rate": 4.106434161655962e-06,
      "loss": 0.0099,
      "reward": 1.421875,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 1042
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 460.828125,
      "epoch": 0.8344,
      "grad_norm": 0.6772855946508758,
      "kl": 1.609375,
      "learning_rate": 4.104828822503427e-06,
      "loss": 0.0648,
      "reward": 1.484375,
      "reward_std": 0.2723158895969391,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.953125,
      "step": 1043
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 444.421875,
      "epoch": 0.8352,
      "grad_norm": 0.7554695451244722,
      "kl": 0.1240234375,
      "learning_rate": 4.103222357020248e-06,
      "loss": 0.005,
      "reward": 1.484375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 1044
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 375.703125,
      "epoch": 0.836,
      "grad_norm": 0.8594837473654137,
      "kl": 0.1875,
      "learning_rate": 4.101614766333904e-06,
      "loss": 0.0075,
      "reward": 1.65625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1045
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 492.578125,
      "epoch": 0.8368,
      "grad_norm": 1.5062552579782205,
      "kl": 2.125,
      "learning_rate": 4.100006051572664e-06,
      "loss": 0.0848,
      "reward": 1.421875,
      "reward_std": 0.2790825366973877,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.9375,
      "step": 1046
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 451.21875,
      "epoch": 0.8376,
      "grad_norm": 0.655316319652123,
      "kl": 0.66796875,
      "learning_rate": 4.098396213865587e-06,
      "loss": 0.0268,
      "reward": 1.515625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 1047
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 394.546875,
      "epoch": 0.8384,
      "grad_norm": 1.2043472474942574,
      "kl": 0.12255859375,
      "learning_rate": 4.096785254342518e-06,
      "loss": 0.0049,
      "reward": 1.703125,
      "reward_std": 0.28460076451301575,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1048
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 427.515625,
      "epoch": 0.8392,
      "grad_norm": 0.9783731884645066,
      "kl": 2.234375,
      "learning_rate": 4.095173174134091e-06,
      "loss": 0.0897,
      "reward": 1.59375,
      "reward_std": 0.3618125021457672,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.921875,
      "step": 1049
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 459.9375,
      "epoch": 0.84,
      "grad_norm": 0.3746688535860715,
      "kl": 0.095703125,
      "learning_rate": 4.093559974371725e-06,
      "loss": 0.0038,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1050
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 482.0,
      "epoch": 0.8408,
      "grad_norm": 2.6628157148567038,
      "kl": 2.78125,
      "learning_rate": 4.091945656187626e-06,
      "loss": 0.1114,
      "reward": 1.78125,
      "reward_std": 0.42674916982650757,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.921875,
      "step": 1051
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.375,
      "completion_length": 430.953125,
      "epoch": 0.8416,
      "grad_norm": 5.829316587830681,
      "kl": 1.140625,
      "learning_rate": 4.090330220714785e-06,
      "loss": 0.0456,
      "reward": 1.34375,
      "reward_std": 0.34352827072143555,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 0.953125,
      "step": 1052
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 413.296875,
      "epoch": 0.8424,
      "grad_norm": 0.9181553115126243,
      "kl": 2.953125,
      "learning_rate": 4.0887136690869774e-06,
      "loss": 0.1181,
      "reward": 1.6875,
      "reward_std": 0.3714011013507843,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.90625,
      "step": 1053
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 444.453125,
      "epoch": 0.8432,
      "grad_norm": 0.7858814990477181,
      "kl": 1.890625,
      "learning_rate": 4.08709600243876e-06,
      "loss": 0.0758,
      "reward": 1.515625,
      "reward_std": 0.26977968215942383,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.9375,
      "step": 1054
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 437.28125,
      "epoch": 0.844,
      "grad_norm": 1.7486694303914967,
      "kl": 2.078125,
      "learning_rate": 4.0854772219054735e-06,
      "loss": 0.0834,
      "reward": 1.6875,
      "reward_std": 0.3807891607284546,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.921875,
      "step": 1055
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 533.75,
      "epoch": 0.8448,
      "grad_norm": 0.8953239164155159,
      "kl": 3.671875,
      "learning_rate": 4.083857328623243e-06,
      "loss": 0.1468,
      "reward": 1.640625,
      "reward_std": 0.4467737078666687,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.875,
      "step": 1056
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 422.921875,
      "epoch": 0.8456,
      "grad_norm": 1.298852110723827,
      "kl": 2.53125,
      "learning_rate": 4.082236323728969e-06,
      "loss": 0.1011,
      "reward": 1.71875,
      "reward_std": 0.3654222786426544,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.90625,
      "step": 1057
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 510.84375,
      "epoch": 0.8464,
      "grad_norm": 4.666457472223463,
      "kl": 7.84375,
      "learning_rate": 4.0806142083603365e-06,
      "loss": 0.3131,
      "reward": 1.4375,
      "reward_std": 0.6150861978530884,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.765625,
      "step": 1058
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 470.5625,
      "epoch": 0.8472,
      "grad_norm": 1.1720509995194253,
      "kl": 4.15625,
      "learning_rate": 4.078990983655807e-06,
      "loss": 0.1665,
      "reward": 1.546875,
      "reward_std": 0.4772983193397522,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.859375,
      "step": 1059
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 389.375,
      "epoch": 0.848,
      "grad_norm": 1.1323759184229216,
      "kl": 1.6328125,
      "learning_rate": 4.077366650754624e-06,
      "loss": 0.0653,
      "reward": 1.765625,
      "reward_std": 0.3095765709877014,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.9375,
      "step": 1060
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 494.640625,
      "epoch": 0.8488,
      "grad_norm": 0.8141453768595988,
      "kl": 1.4453125,
      "learning_rate": 4.075741210796806e-06,
      "loss": 0.0578,
      "reward": 1.671875,
      "reward_std": 0.2834876775741577,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.9375,
      "step": 1061
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 391.921875,
      "epoch": 0.8496,
      "grad_norm": 1.537521701564151,
      "kl": 2.5,
      "learning_rate": 4.07411466492315e-06,
      "loss": 0.1001,
      "reward": 1.515625,
      "reward_std": 0.46376699209213257,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.90625,
      "step": 1062
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 383.0625,
      "epoch": 0.8504,
      "grad_norm": 2.096686728399835,
      "kl": 1.3203125,
      "learning_rate": 4.072487014275228e-06,
      "loss": 0.0527,
      "reward": 1.765625,
      "reward_std": 0.4027767479419708,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.953125,
      "step": 1063
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 445.1875,
      "epoch": 0.8512,
      "grad_norm": 0.8970506248741964,
      "kl": 0.62109375,
      "learning_rate": 4.070858259995388e-06,
      "loss": 0.0248,
      "reward": 1.515625,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.96875,
      "step": 1064
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 407.015625,
      "epoch": 0.852,
      "grad_norm": 1.889448913986524,
      "kl": 0.9296875,
      "learning_rate": 4.069228403226751e-06,
      "loss": 0.0374,
      "reward": 1.546875,
      "reward_std": 0.3776973485946655,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.953125,
      "step": 1065
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 417.859375,
      "epoch": 0.8528,
      "grad_norm": 1.8209643815758436,
      "kl": 1.03125,
      "learning_rate": 4.067597445113216e-06,
      "loss": 0.0413,
      "reward": 1.859375,
      "reward_std": 0.25726157426834106,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.96875,
      "step": 1066
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 422.9375,
      "epoch": 0.8536,
      "grad_norm": 1.138793212784693,
      "kl": 0.326171875,
      "learning_rate": 4.06596538679945e-06,
      "loss": 0.013,
      "reward": 1.546875,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 1067
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 351.546875,
      "epoch": 0.8544,
      "grad_norm": 1.319715729705474,
      "kl": 0.86328125,
      "learning_rate": 4.064332229430895e-06,
      "loss": 0.0344,
      "reward": 1.5,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 1068
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 428.375,
      "epoch": 0.8552,
      "grad_norm": 1.0128713148904005,
      "kl": 1.2421875,
      "learning_rate": 4.062697974153764e-06,
      "loss": 0.0499,
      "reward": 1.640625,
      "reward_std": 0.24831002950668335,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.953125,
      "step": 1069
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 403.71875,
      "epoch": 0.856,
      "grad_norm": 1.2929267924245809,
      "kl": 0.7578125,
      "learning_rate": 4.06106262211504e-06,
      "loss": 0.0302,
      "reward": 1.6875,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 1070
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 405.484375,
      "epoch": 0.8568,
      "grad_norm": 1.6960882081379258,
      "kl": 0.21484375,
      "learning_rate": 4.059426174462476e-06,
      "loss": 0.0086,
      "reward": 1.546875,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 1071
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 412.40625,
      "epoch": 0.8576,
      "grad_norm": 0.9100252003673931,
      "kl": 0.62890625,
      "learning_rate": 4.057788632344594e-06,
      "loss": 0.0252,
      "reward": 1.546875,
      "reward_std": 0.2993341088294983,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 1072
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 427.671875,
      "epoch": 0.8584,
      "grad_norm": 1.0317845091800215,
      "kl": 1.28125,
      "learning_rate": 4.056149996910683e-06,
      "loss": 0.0513,
      "reward": 1.6875,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 1073
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 398.21875,
      "epoch": 0.8592,
      "grad_norm": 1.9072878508109186,
      "kl": 2.296875,
      "learning_rate": 4.054510269310803e-06,
      "loss": 0.0921,
      "reward": 1.53125,
      "reward_std": 0.3230288028717041,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.921875,
      "step": 1074
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 387.25,
      "epoch": 0.86,
      "grad_norm": 1.0221055564406814,
      "kl": 0.126953125,
      "learning_rate": 4.052869450695776e-06,
      "loss": 0.0051,
      "reward": 1.453125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 1075
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 425.875,
      "epoch": 0.8608,
      "grad_norm": 0.8024075102370594,
      "kl": 1.21875,
      "learning_rate": 4.051227542217192e-06,
      "loss": 0.0487,
      "reward": 1.421875,
      "reward_std": 0.28460076451301575,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.96875,
      "step": 1076
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 397.484375,
      "epoch": 0.8616,
      "grad_norm": 1.0949966164463376,
      "kl": 0.251953125,
      "learning_rate": 4.049584545027406e-06,
      "loss": 0.0101,
      "reward": 1.625,
      "reward_std": 0.2756394147872925,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1077
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 460.3125,
      "epoch": 0.8624,
      "grad_norm": 1.4031116868993465,
      "kl": 1.4921875,
      "learning_rate": 4.047940460279537e-06,
      "loss": 0.0597,
      "reward": 1.671875,
      "reward_std": 0.22673699259757996,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.9375,
      "step": 1078
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 429.421875,
      "epoch": 0.8632,
      "grad_norm": 1.05338284339803,
      "kl": 0.82421875,
      "learning_rate": 4.046295289127466e-06,
      "loss": 0.033,
      "reward": 1.640625,
      "reward_std": 0.29826053977012634,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 1079
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 378.84375,
      "epoch": 0.864,
      "grad_norm": 0.5837210666821087,
      "kl": 0.4296875,
      "learning_rate": 4.044649032725836e-06,
      "loss": 0.0172,
      "reward": 1.421875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.984375,
      "step": 1080
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 399.9375,
      "epoch": 0.8648,
      "grad_norm": 1.2920251508082814,
      "kl": 0.890625,
      "learning_rate": 4.0430016922300566e-06,
      "loss": 0.0356,
      "reward": 1.828125,
      "reward_std": 0.31983357667922974,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 1081
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 455.78125,
      "epoch": 0.8656,
      "grad_norm": 2.0959285204173743,
      "kl": 1.8359375,
      "learning_rate": 4.0413532687962926e-06,
      "loss": 0.0733,
      "reward": 1.5625,
      "reward_std": 0.16130642592906952,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 1082
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 381.296875,
      "epoch": 0.8664,
      "grad_norm": 0.6957391888221854,
      "kl": 0.61328125,
      "learning_rate": 4.039703763581472e-06,
      "loss": 0.0245,
      "reward": 1.53125,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 1083
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 375.390625,
      "epoch": 0.8672,
      "grad_norm": 0.9464861405251479,
      "kl": 1.21875,
      "learning_rate": 4.038053177743279e-06,
      "loss": 0.0489,
      "reward": 1.453125,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.96875,
      "step": 1084
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 484.1875,
      "epoch": 0.868,
      "grad_norm": 0.622831183756262,
      "kl": 1.2265625,
      "learning_rate": 4.036401512440161e-06,
      "loss": 0.0493,
      "reward": 1.78125,
      "reward_std": 0.1735912710428238,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 1085
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 418.25,
      "epoch": 0.8688,
      "grad_norm": 0.8347975194065455,
      "kl": 2.21875,
      "learning_rate": 4.034748768831319e-06,
      "loss": 0.0889,
      "reward": 1.640625,
      "reward_std": 0.26977968215942383,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.9375,
      "step": 1086
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 421.375,
      "epoch": 0.8696,
      "grad_norm": 0.7875842265559289,
      "kl": 0.1220703125,
      "learning_rate": 4.033094948076713e-06,
      "loss": 0.0049,
      "reward": 1.671875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1087
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 379.453125,
      "epoch": 0.8704,
      "grad_norm": 0.6386806171987647,
      "kl": 0.126953125,
      "learning_rate": 4.031440051337056e-06,
      "loss": 0.0051,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1088
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 414.65625,
      "epoch": 0.8712,
      "grad_norm": 2.838155575700541,
      "kl": 1.78125,
      "learning_rate": 4.02978407977382e-06,
      "loss": 0.0713,
      "reward": 1.5625,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 1089
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 409.578125,
      "epoch": 0.872,
      "grad_norm": 1.508732730579972,
      "kl": 1.7578125,
      "learning_rate": 4.02812703454923e-06,
      "loss": 0.0701,
      "reward": 1.5625,
      "reward_std": 0.1828794628381729,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 1090
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 472.84375,
      "epoch": 0.8728,
      "grad_norm": 1.1816446654777986,
      "kl": 0.12109375,
      "learning_rate": 4.026468916826262e-06,
      "loss": 0.0048,
      "reward": 1.359375,
      "reward_std": 0.28930896520614624,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 1091
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 454.3125,
      "epoch": 0.8736,
      "grad_norm": 0.9523527525690174,
      "kl": 1.0390625,
      "learning_rate": 4.024809727768648e-06,
      "loss": 0.0414,
      "reward": 1.78125,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 1092
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 358.671875,
      "epoch": 0.8744,
      "grad_norm": 1.3600026041030495,
      "kl": 0.140625,
      "learning_rate": 4.023149468540871e-06,
      "loss": 0.0056,
      "reward": 1.625,
      "reward_std": 0.3104073107242584,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1093
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 376.578125,
      "epoch": 0.8752,
      "grad_norm": 0.76887038816068,
      "kl": 0.57421875,
      "learning_rate": 4.021488140308165e-06,
      "loss": 0.0229,
      "reward": 1.703125,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 1094
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 448.9375,
      "epoch": 0.876,
      "grad_norm": 0.8105513072622308,
      "kl": 1.0703125,
      "learning_rate": 4.019825744236514e-06,
      "loss": 0.0429,
      "reward": 1.765625,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.96875,
      "step": 1095
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 440.015625,
      "epoch": 0.8768,
      "grad_norm": 0.7716018449654013,
      "kl": 0.91796875,
      "learning_rate": 4.018162281492651e-06,
      "loss": 0.0367,
      "reward": 1.84375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 1096
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 435.078125,
      "epoch": 0.8776,
      "grad_norm": 0.6501186507228375,
      "kl": 1.390625,
      "learning_rate": 4.016497753244058e-06,
      "loss": 0.0557,
      "reward": 1.65625,
      "reward_std": 0.26658445596694946,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.953125,
      "step": 1097
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 528.0,
      "epoch": 0.8784,
      "grad_norm": 0.5794337582028911,
      "kl": 0.6875,
      "learning_rate": 4.014832160658966e-06,
      "loss": 0.0276,
      "reward": 1.65625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 1098
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 501.40625,
      "epoch": 0.8792,
      "grad_norm": 0.8259051532628278,
      "kl": 0.58984375,
      "learning_rate": 4.013165504906352e-06,
      "loss": 0.0235,
      "reward": 1.640625,
      "reward_std": 0.2414703369140625,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 1099
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 470.9375,
      "epoch": 0.88,
      "grad_norm": 1.1016484370480908,
      "kl": 0.7578125,
      "learning_rate": 4.011497787155938e-06,
      "loss": 0.0303,
      "reward": 1.765625,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.96875,
      "step": 1100
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 442.0,
      "epoch": 0.8808,
      "grad_norm": 0.5794692144971305,
      "kl": 0.123046875,
      "learning_rate": 4.009829008578192e-06,
      "loss": 0.0049,
      "reward": 1.84375,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1101
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 467.859375,
      "epoch": 0.8816,
      "grad_norm": 0.3343264060635989,
      "kl": 0.12060546875,
      "learning_rate": 4.00815917034433e-06,
      "loss": 0.0048,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1102
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 493.53125,
      "epoch": 0.8824,
      "grad_norm": 0.6032848408012575,
      "kl": 1.4921875,
      "learning_rate": 4.006488273626307e-06,
      "loss": 0.0596,
      "reward": 1.734375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1103
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 447.625,
      "epoch": 0.8832,
      "grad_norm": 2.1882872072718604,
      "kl": 0.91796875,
      "learning_rate": 4.004816319596822e-06,
      "loss": 0.0368,
      "reward": 1.703125,
      "reward_std": 0.27554580569267273,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 1104
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 440.578125,
      "epoch": 0.884,
      "grad_norm": 2.0923907225823064,
      "kl": 2.28125,
      "learning_rate": 4.003143309429317e-06,
      "loss": 0.0909,
      "reward": 1.515625,
      "reward_std": 0.48076385259628296,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.890625,
      "step": 1105
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 458.203125,
      "epoch": 0.8848,
      "grad_norm": 1.978736290913115,
      "kl": 3.390625,
      "learning_rate": 4.0014692442979756e-06,
      "loss": 0.1358,
      "reward": 1.3125,
      "reward_std": 0.436609148979187,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.90625,
      "step": 1106
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 461.1875,
      "epoch": 0.8856,
      "grad_norm": 6.48888169484247,
      "kl": 1.234375,
      "learning_rate": 3.999794125377721e-06,
      "loss": 0.0494,
      "reward": 1.640625,
      "reward_std": 0.3481428921222687,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.96875,
      "step": 1107
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 484.0625,
      "epoch": 0.8864,
      "grad_norm": 2.906234421867136,
      "kl": 4.71875,
      "learning_rate": 3.998117953844215e-06,
      "loss": 0.1884,
      "reward": 1.34375,
      "reward_std": 0.4506639540195465,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.84375,
      "step": 1108
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 436.0625,
      "epoch": 0.8872,
      "grad_norm": 1.4255548194518288,
      "kl": 1.65625,
      "learning_rate": 3.996440730873861e-06,
      "loss": 0.0664,
      "reward": 1.65625,
      "reward_std": 0.3335031270980835,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.953125,
      "step": 1109
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 444.609375,
      "epoch": 0.888,
      "grad_norm": 1.234066141392723,
      "kl": 0.76953125,
      "learning_rate": 3.9947624576437975e-06,
      "loss": 0.0306,
      "reward": 1.875,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1110
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 432.40625,
      "epoch": 0.8888,
      "grad_norm": 1.4277767028325805,
      "kl": 0.90234375,
      "learning_rate": 3.9930831353319025e-06,
      "loss": 0.0361,
      "reward": 1.59375,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.953125,
      "step": 1111
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 450.796875,
      "epoch": 0.8896,
      "grad_norm": 1.2248055853199278,
      "kl": 0.369140625,
      "learning_rate": 3.9914027651167866e-06,
      "loss": 0.0148,
      "reward": 1.71875,
      "reward_std": 0.391263484954834,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 1112
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 488.015625,
      "epoch": 0.8904,
      "grad_norm": 1.109925095383779,
      "kl": 1.1875,
      "learning_rate": 3.989721348177801e-06,
      "loss": 0.0477,
      "reward": 1.765625,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.953125,
      "step": 1113
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 391.234375,
      "epoch": 0.8912,
      "grad_norm": 1.0836512915420775,
      "kl": 0.138671875,
      "learning_rate": 3.988038885695028e-06,
      "loss": 0.0055,
      "reward": 1.765625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1114
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 476.578125,
      "epoch": 0.892,
      "grad_norm": 1.9829381449184804,
      "kl": 1.1015625,
      "learning_rate": 3.986355378849284e-06,
      "loss": 0.0439,
      "reward": 1.671875,
      "reward_std": 0.27883464097976685,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.953125,
      "step": 1115
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 400.734375,
      "epoch": 0.8928,
      "grad_norm": 0.9301515106093913,
      "kl": 0.1201171875,
      "learning_rate": 3.984670828822118e-06,
      "loss": 0.0048,
      "reward": 1.65625,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1116
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 474.0625,
      "epoch": 0.8936,
      "grad_norm": 2.258156989865185,
      "kl": 1.796875,
      "learning_rate": 3.982985236795815e-06,
      "loss": 0.0721,
      "reward": 1.546875,
      "reward_std": 0.3276434540748596,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.921875,
      "step": 1117
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 431.703125,
      "epoch": 0.8944,
      "grad_norm": 0.6029594087123232,
      "kl": 0.130859375,
      "learning_rate": 3.981298603953385e-06,
      "loss": 0.0052,
      "reward": 1.703125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1118
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 464.78125,
      "epoch": 0.8952,
      "grad_norm": 1.0329750184692497,
      "kl": 1.6015625,
      "learning_rate": 3.979610931478574e-06,
      "loss": 0.0642,
      "reward": 1.703125,
      "reward_std": 0.30607038736343384,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.9375,
      "step": 1119
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 446.453125,
      "epoch": 0.896,
      "grad_norm": 0.8947904659942725,
      "kl": 0.484375,
      "learning_rate": 3.977922220555855e-06,
      "loss": 0.0194,
      "reward": 1.765625,
      "reward_std": 0.3025038540363312,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 1120
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 356.34375,
      "epoch": 0.8968,
      "grad_norm": 0.8427329686967138,
      "kl": 1.5546875,
      "learning_rate": 3.976232472370431e-06,
      "loss": 0.0622,
      "reward": 1.46875,
      "reward_std": 0.25292468070983887,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.953125,
      "step": 1121
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 447.9375,
      "epoch": 0.8976,
      "grad_norm": 2.9962251111735,
      "kl": 1.3046875,
      "learning_rate": 3.97454168810823e-06,
      "loss": 0.0521,
      "reward": 1.609375,
      "reward_std": 0.36722296476364136,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.9375,
      "step": 1122
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 447.40625,
      "epoch": 0.8984,
      "grad_norm": 1.6896760435640983,
      "kl": 2.203125,
      "learning_rate": 3.972849868955913e-06,
      "loss": 0.0882,
      "reward": 1.8125,
      "reward_std": 0.3945523500442505,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.921875,
      "step": 1123
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 494.34375,
      "epoch": 0.8992,
      "grad_norm": 4.102207391039708,
      "kl": 7.21875,
      "learning_rate": 3.97115701610086e-06,
      "loss": 0.2888,
      "reward": 1.21875,
      "reward_std": 0.5415065288543701,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.78125,
      "step": 1124
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 435.03125,
      "epoch": 0.9,
      "grad_norm": 0.7286601798437264,
      "kl": 0.98828125,
      "learning_rate": 3.969463130731183e-06,
      "loss": 0.0396,
      "reward": 1.5625,
      "reward_std": 0.26409146189689636,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.96875,
      "step": 1125
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 512.15625,
      "epoch": 0.9008,
      "grad_norm": 1.7493609069325222,
      "kl": 3.3125,
      "learning_rate": 3.967768214035716e-06,
      "loss": 0.1325,
      "reward": 1.421875,
      "reward_std": 0.4455862045288086,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.84375,
      "step": 1126
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 483.546875,
      "epoch": 0.9016,
      "grad_norm": 0.9227670533354237,
      "kl": 2.46875,
      "learning_rate": 3.966072267204014e-06,
      "loss": 0.099,
      "reward": 1.703125,
      "reward_std": 0.4069768190383911,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.921875,
      "step": 1127
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 504.484375,
      "epoch": 0.9024,
      "grad_norm": 1.7716691686347417,
      "kl": 4.40625,
      "learning_rate": 3.964375291426361e-06,
      "loss": 0.1756,
      "reward": 1.53125,
      "reward_std": 0.6183469295501709,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.84375,
      "step": 1128
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 435.25,
      "epoch": 0.9032,
      "grad_norm": 2.445301463312812,
      "kl": 5.28125,
      "learning_rate": 3.962677287893758e-06,
      "loss": 0.2111,
      "reward": 1.28125,
      "reward_std": 0.6274019479751587,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.796875,
      "step": 1129
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 548.484375,
      "epoch": 0.904,
      "grad_norm": 5.123589632294507,
      "kl": 8.4375,
      "learning_rate": 3.9609782577979305e-06,
      "loss": 0.3384,
      "reward": 1.328125,
      "reward_std": 0.5813445448875427,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.734375,
      "step": 1130
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 517.125,
      "epoch": 0.9048,
      "grad_norm": 0.9773982278643543,
      "kl": 2.421875,
      "learning_rate": 3.959278202331323e-06,
      "loss": 0.0966,
      "reward": 1.53125,
      "reward_std": 0.4171832799911499,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.90625,
      "step": 1131
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 606.265625,
      "epoch": 0.9056,
      "grad_norm": 1.8726346922242276,
      "kl": 2.734375,
      "learning_rate": 3.9575771226870986e-06,
      "loss": 0.1091,
      "reward": 1.484375,
      "reward_std": 0.37261295318603516,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.890625,
      "step": 1132
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 461.75,
      "epoch": 0.9064,
      "grad_norm": 1.1884237859088511,
      "kl": 1.203125,
      "learning_rate": 3.955875020059141e-06,
      "loss": 0.0481,
      "reward": 1.734375,
      "reward_std": 0.22673699259757996,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.953125,
      "step": 1133
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 536.765625,
      "epoch": 0.9072,
      "grad_norm": 5.028142425015897,
      "kl": 2.0625,
      "learning_rate": 3.954171895642052e-06,
      "loss": 0.0822,
      "reward": 1.546875,
      "reward_std": 0.5079944133758545,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.875,
      "step": 1134
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 481.765625,
      "epoch": 0.908,
      "grad_norm": 2.6416702453276626,
      "kl": 1.671875,
      "learning_rate": 3.9524677506311505e-06,
      "loss": 0.067,
      "reward": 1.59375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.96875,
      "step": 1135
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 490.15625,
      "epoch": 0.9088,
      "grad_norm": 0.9754178084529734,
      "kl": 1.3046875,
      "learning_rate": 3.950762586222469e-06,
      "loss": 0.0521,
      "reward": 1.5625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 1136
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 478.46875,
      "epoch": 0.9096,
      "grad_norm": 2.0716244414818745,
      "kl": 0.6484375,
      "learning_rate": 3.949056403612758e-06,
      "loss": 0.026,
      "reward": 1.390625,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.984375,
      "step": 1137
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.5,
      "completion_length": 415.828125,
      "epoch": 0.9104,
      "grad_norm": 1.4351468584193703,
      "kl": 1.078125,
      "learning_rate": 3.947349203999485e-06,
      "loss": 0.0431,
      "reward": 1.21875,
      "reward_std": 0.2536902129650116,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 0.96875,
      "step": 1138
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 393.84375,
      "epoch": 0.9112,
      "grad_norm": 1.0885993983624644,
      "kl": 0.25,
      "learning_rate": 3.945640988580824e-06,
      "loss": 0.01,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1139
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 497.140625,
      "epoch": 0.912,
      "grad_norm": 1.481975895271482,
      "kl": 0.734375,
      "learning_rate": 3.943931758555669e-06,
      "loss": 0.0293,
      "reward": 1.734375,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.96875,
      "step": 1140
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 488.171875,
      "epoch": 0.9128,
      "grad_norm": 1.831247403573637,
      "kl": 1.484375,
      "learning_rate": 3.942221515123624e-06,
      "loss": 0.0595,
      "reward": 1.625,
      "reward_std": 0.42873120307922363,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.90625,
      "step": 1141
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 466.890625,
      "epoch": 0.9136,
      "grad_norm": 3.348503762208204,
      "kl": 0.60546875,
      "learning_rate": 3.940510259485002e-06,
      "loss": 0.0242,
      "reward": 1.5,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 1142
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 375.828125,
      "epoch": 0.9144,
      "grad_norm": 0.6862518615346115,
      "kl": 0.82421875,
      "learning_rate": 3.938797992840828e-06,
      "loss": 0.0328,
      "reward": 1.53125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.953125,
      "step": 1143
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 428.375,
      "epoch": 0.9152,
      "grad_norm": 0.8253381131921054,
      "kl": 0.58984375,
      "learning_rate": 3.937084716392839e-06,
      "loss": 0.0236,
      "reward": 1.5625,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 1144
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 416.40625,
      "epoch": 0.916,
      "grad_norm": 1.4288078021535953,
      "kl": 1.5234375,
      "learning_rate": 3.935370431343475e-06,
      "loss": 0.0611,
      "reward": 1.59375,
      "reward_std": 0.28247910737991333,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.953125,
      "step": 1145
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.5,
      "completion_length": 426.421875,
      "epoch": 0.9168,
      "grad_norm": 1.3095918335689882,
      "kl": 1.296875,
      "learning_rate": 3.933655138895889e-06,
      "loss": 0.0518,
      "reward": 1.1875,
      "reward_std": 0.3014557361602783,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 0.9375,
      "step": 1146
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 449.078125,
      "epoch": 0.9176,
      "grad_norm": 0.5376006623531796,
      "kl": 0.107421875,
      "learning_rate": 3.9319388402539395e-06,
      "loss": 0.0043,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1147
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 406.859375,
      "epoch": 0.9184,
      "grad_norm": 0.9998521561750789,
      "kl": 0.91796875,
      "learning_rate": 3.930221536622192e-06,
      "loss": 0.0366,
      "reward": 1.625,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.96875,
      "step": 1148
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 411.171875,
      "epoch": 0.9192,
      "grad_norm": 1.0514708714206056,
      "kl": 1.484375,
      "learning_rate": 3.928503229205913e-06,
      "loss": 0.0595,
      "reward": 1.796875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.9375,
      "step": 1149
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 384.34375,
      "epoch": 0.92,
      "grad_norm": 1.0901882090105197,
      "kl": 0.79296875,
      "learning_rate": 3.92678391921108e-06,
      "loss": 0.0318,
      "reward": 1.390625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.984375,
      "step": 1150
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 388.921875,
      "epoch": 0.9208,
      "grad_norm": 0.7060251409446954,
      "kl": 0.9765625,
      "learning_rate": 3.92506360784437e-06,
      "loss": 0.0392,
      "reward": 1.671875,
      "reward_std": 0.19044627249240875,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.96875,
      "step": 1151
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 343.203125,
      "epoch": 0.9216,
      "grad_norm": 1.3515674063755896,
      "kl": 1.1484375,
      "learning_rate": 3.923342296313162e-06,
      "loss": 0.0459,
      "reward": 1.515625,
      "reward_std": 0.24039676785469055,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 1152
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 373.1875,
      "epoch": 0.9224,
      "grad_norm": 1.4739617557166975,
      "kl": 2.0625,
      "learning_rate": 3.92161998582554e-06,
      "loss": 0.0827,
      "reward": 1.6875,
      "reward_std": 0.2470981925725937,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.9375,
      "step": 1153
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 378.109375,
      "epoch": 0.9232,
      "grad_norm": 1.0225942609340664,
      "kl": 1.671875,
      "learning_rate": 3.919896677590289e-06,
      "loss": 0.067,
      "reward": 1.5625,
      "reward_std": 0.2925041913986206,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.9375,
      "step": 1154
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 489.203125,
      "epoch": 0.924,
      "grad_norm": 1.2150982355587674,
      "kl": 1.8359375,
      "learning_rate": 3.918172372816892e-06,
      "loss": 0.0734,
      "reward": 1.59375,
      "reward_std": 0.4230673909187317,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.953125,
      "step": 1155
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 416.546875,
      "epoch": 0.9248,
      "grad_norm": 0.8533076568013679,
      "kl": 1.234375,
      "learning_rate": 3.916447072715531e-06,
      "loss": 0.0493,
      "reward": 1.84375,
      "reward_std": 0.24608497321605682,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.953125,
      "step": 1156
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 366.921875,
      "epoch": 0.9256,
      "grad_norm": 0.6901681181296133,
      "kl": 1.4453125,
      "learning_rate": 3.914720778497091e-06,
      "loss": 0.0578,
      "reward": 1.625,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.953125,
      "step": 1157
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 446.59375,
      "epoch": 0.9264,
      "grad_norm": 0.9450905375866617,
      "kl": 1.484375,
      "learning_rate": 3.91299349137315e-06,
      "loss": 0.059,
      "reward": 1.546875,
      "reward_std": 0.22707363963127136,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 1158
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 421.390625,
      "epoch": 0.9272,
      "grad_norm": 0.8093811017020308,
      "kl": 3.0625,
      "learning_rate": 3.9112652125559845e-06,
      "loss": 0.122,
      "reward": 1.390625,
      "reward_std": 0.329900860786438,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.890625,
      "step": 1159
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 406.15625,
      "epoch": 0.928,
      "grad_norm": 1.2667773493866263,
      "kl": 1.234375,
      "learning_rate": 3.909535943258567e-06,
      "loss": 0.0494,
      "reward": 1.515625,
      "reward_std": 0.2777610719203949,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.96875,
      "step": 1160
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 374.65625,
      "epoch": 0.9288,
      "grad_norm": 2.229480230522335,
      "kl": 2.25,
      "learning_rate": 3.907805684694567e-06,
      "loss": 0.0902,
      "reward": 1.390625,
      "reward_std": 0.33374619483947754,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.921875,
      "step": 1161
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 420.296875,
      "epoch": 0.9296,
      "grad_norm": 0.9551605600836488,
      "kl": 3.046875,
      "learning_rate": 3.906074438078343e-06,
      "loss": 0.1216,
      "reward": 1.625,
      "reward_std": 0.3139738440513611,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.921875,
      "step": 1162
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 416.234375,
      "epoch": 0.9304,
      "grad_norm": 2.1055650505222054,
      "kl": 1.7578125,
      "learning_rate": 3.904342204624955e-06,
      "loss": 0.07,
      "reward": 1.875,
      "reward_std": 0.28566452860832214,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.9375,
      "step": 1163
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 458.703125,
      "epoch": 0.9312,
      "grad_norm": 0.5495339289985234,
      "kl": 0.9453125,
      "learning_rate": 3.9026089855501475e-06,
      "loss": 0.0377,
      "reward": 1.78125,
      "reward_std": 0.15870985388755798,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.9375,
      "step": 1164
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 389.625,
      "epoch": 0.932,
      "grad_norm": 0.3289399541161126,
      "kl": 0.12353515625,
      "learning_rate": 3.900874782070362e-06,
      "loss": 0.0049,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1165
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 403.015625,
      "epoch": 0.9328,
      "grad_norm": 0.9725077327143433,
      "kl": 0.68359375,
      "learning_rate": 3.899139595402729e-06,
      "loss": 0.0273,
      "reward": 1.453125,
      "reward_std": 0.24831002950668335,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.984375,
      "step": 1166
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 397.109375,
      "epoch": 0.9336,
      "grad_norm": 0.5044647084513043,
      "kl": 0.259765625,
      "learning_rate": 3.8974034267650695e-06,
      "loss": 0.0103,
      "reward": 1.75,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1167
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 390.6875,
      "epoch": 0.9344,
      "grad_norm": 0.6203756574718026,
      "kl": 0.306640625,
      "learning_rate": 3.895666277375892e-06,
      "loss": 0.0122,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 1168
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 383.328125,
      "epoch": 0.9352,
      "grad_norm": 0.8361423511027497,
      "kl": 0.61328125,
      "learning_rate": 3.893928148454398e-06,
      "loss": 0.0245,
      "reward": 1.59375,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1169
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 354.515625,
      "epoch": 0.936,
      "grad_norm": 0.39279756261166693,
      "kl": 0.16796875,
      "learning_rate": 3.89218904122047e-06,
      "loss": 0.0067,
      "reward": 1.4375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 1170
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 363.59375,
      "epoch": 0.9368,
      "grad_norm": 1.205513888683326,
      "kl": 1.578125,
      "learning_rate": 3.890448956894682e-06,
      "loss": 0.063,
      "reward": 1.453125,
      "reward_std": 0.22636085748672485,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.921875,
      "step": 1171
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 342.65625,
      "epoch": 0.9376,
      "grad_norm": 1.6243501485485625,
      "kl": 0.87890625,
      "learning_rate": 3.888707896698293e-06,
      "loss": 0.0352,
      "reward": 1.640625,
      "reward_std": 0.3359614610671997,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.953125,
      "step": 1172
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 387.859375,
      "epoch": 0.9384,
      "grad_norm": 0.8566248245185764,
      "kl": 0.89453125,
      "learning_rate": 3.886965861853243e-06,
      "loss": 0.0357,
      "reward": 1.5625,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 1173
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 377.796875,
      "epoch": 0.9392,
      "grad_norm": 0.7413774914060944,
      "kl": 0.30859375,
      "learning_rate": 3.885222853582163e-06,
      "loss": 0.0123,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1174
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 429.1875,
      "epoch": 0.94,
      "grad_norm": 0.8688677070068936,
      "kl": 1.1796875,
      "learning_rate": 3.88347887310836e-06,
      "loss": 0.0472,
      "reward": 1.734375,
      "reward_std": 0.2902791500091553,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.96875,
      "step": 1175
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 402.859375,
      "epoch": 0.9408,
      "grad_norm": 1.9388823393178904,
      "kl": 1.1015625,
      "learning_rate": 3.881733921655829e-06,
      "loss": 0.0444,
      "reward": 1.5,
      "reward_std": 0.2435920089483261,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.953125,
      "step": 1176
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 433.453125,
      "epoch": 0.9416,
      "grad_norm": 1.1395263747176578,
      "kl": 0.546875,
      "learning_rate": 3.879988000449243e-06,
      "loss": 0.0218,
      "reward": 1.546875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 1177
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 399.03125,
      "epoch": 0.9424,
      "grad_norm": 1.772426608201281,
      "kl": 3.0,
      "learning_rate": 3.878241110713957e-06,
      "loss": 0.1196,
      "reward": 1.53125,
      "reward_std": 0.34023943543434143,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.90625,
      "step": 1178
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 426.734375,
      "epoch": 0.9432,
      "grad_norm": 0.8069962534616799,
      "kl": 1.015625,
      "learning_rate": 3.876493253676004e-06,
      "loss": 0.0408,
      "reward": 1.734375,
      "reward_std": 0.28778618574142456,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.96875,
      "step": 1179
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 394.890625,
      "epoch": 0.944,
      "grad_norm": 0.717350065603133,
      "kl": 0.89453125,
      "learning_rate": 3.8747444305621e-06,
      "loss": 0.0358,
      "reward": 1.8125,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.953125,
      "step": 1180
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 394.625,
      "epoch": 0.9448,
      "grad_norm": 0.6645661864927441,
      "kl": 0.92578125,
      "learning_rate": 3.872994642599635e-06,
      "loss": 0.0369,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.96875,
      "step": 1181
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 444.84375,
      "epoch": 0.9456,
      "grad_norm": 0.7144934656162855,
      "kl": 1.171875,
      "learning_rate": 3.871243891016676e-06,
      "loss": 0.0467,
      "reward": 1.703125,
      "reward_std": 0.24831002950668335,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 1182
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 382.265625,
      "epoch": 0.9464,
      "grad_norm": 1.5216401243548279,
      "kl": 2.484375,
      "learning_rate": 3.869492177041971e-06,
      "loss": 0.0997,
      "reward": 1.71875,
      "reward_std": 0.2924008071422577,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.9375,
      "step": 1183
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 383.28125,
      "epoch": 0.9472,
      "grad_norm": 0.873052809745924,
      "kl": 1.0,
      "learning_rate": 3.867739501904938e-06,
      "loss": 0.0399,
      "reward": 1.734375,
      "reward_std": 0.2902791500091553,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.96875,
      "step": 1184
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 384.640625,
      "epoch": 0.948,
      "grad_norm": 0.5752901880466262,
      "kl": 1.140625,
      "learning_rate": 3.8659858668356735e-06,
      "loss": 0.0458,
      "reward": 1.578125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.96875,
      "step": 1185
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 408.046875,
      "epoch": 0.9488,
      "grad_norm": 1.9259470342314413,
      "kl": 2.5,
      "learning_rate": 3.864231273064944e-06,
      "loss": 0.0997,
      "reward": 1.484375,
      "reward_std": 0.25726157426834106,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.9375,
      "step": 1186
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 392.40625,
      "epoch": 0.9496,
      "grad_norm": 0.7736666345822933,
      "kl": 0.1123046875,
      "learning_rate": 3.862475721824193e-06,
      "loss": 0.0045,
      "reward": 1.921875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1187
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 365.296875,
      "epoch": 0.9504,
      "grad_norm": 0.4368484960802902,
      "kl": 0.1650390625,
      "learning_rate": 3.8607192143455325e-06,
      "loss": 0.0066,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1188
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 403.5625,
      "epoch": 0.9512,
      "grad_norm": 1.2212503251152664,
      "kl": 0.5859375,
      "learning_rate": 3.858961751861748e-06,
      "loss": 0.0236,
      "reward": 1.53125,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 1189
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 431.703125,
      "epoch": 0.952,
      "grad_norm": 0.8227885381713197,
      "kl": 0.52734375,
      "learning_rate": 3.857203335606294e-06,
      "loss": 0.0211,
      "reward": 1.640625,
      "reward_std": 0.32407689094543457,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.96875,
      "step": 1190
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 448.828125,
      "epoch": 0.9528,
      "grad_norm": 0.3710378790969126,
      "kl": 0.796875,
      "learning_rate": 3.855443966813295e-06,
      "loss": 0.0319,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 1191
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 373.046875,
      "epoch": 0.9536,
      "grad_norm": 0.8855073880304769,
      "kl": 0.412109375,
      "learning_rate": 3.853683646717543e-06,
      "loss": 0.0165,
      "reward": 1.703125,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1192
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 400.296875,
      "epoch": 0.9544,
      "grad_norm": 0.7794909961979359,
      "kl": 1.2890625,
      "learning_rate": 3.8519223765544985e-06,
      "loss": 0.0517,
      "reward": 1.625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.96875,
      "step": 1193
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 395.9375,
      "epoch": 0.9552,
      "grad_norm": 0.6807966086632076,
      "kl": 1.25,
      "learning_rate": 3.85016015756029e-06,
      "loss": 0.0499,
      "reward": 1.4375,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.96875,
      "step": 1194
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 369.875,
      "epoch": 0.956,
      "grad_norm": 0.8135046009324578,
      "kl": 0.62109375,
      "learning_rate": 3.848396990971709e-06,
      "loss": 0.0249,
      "reward": 1.453125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.984375,
      "step": 1195
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 397.5,
      "epoch": 0.9568,
      "grad_norm": 4.325337691463792,
      "kl": 0.8125,
      "learning_rate": 3.846632878026214e-06,
      "loss": 0.0326,
      "reward": 1.453125,
      "reward_std": 0.1889965683221817,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.96875,
      "step": 1196
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 404.125,
      "epoch": 0.9576,
      "grad_norm": 1.0605527119395879,
      "kl": 1.8125,
      "learning_rate": 3.844867819961928e-06,
      "loss": 0.0726,
      "reward": 1.671875,
      "reward_std": 0.24831002950668335,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.953125,
      "step": 1197
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 356.5625,
      "epoch": 0.9584,
      "grad_norm": 2.5066496244472956,
      "kl": 0.11865234375,
      "learning_rate": 3.843101818017637e-06,
      "loss": 0.0047,
      "reward": 1.703125,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1198
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 372.375,
      "epoch": 0.9592,
      "grad_norm": 0.3707204148434657,
      "kl": 0.11376953125,
      "learning_rate": 3.841334873432789e-06,
      "loss": 0.0046,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1199
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 416.296875,
      "epoch": 0.96,
      "grad_norm": 1.3171464037408456,
      "kl": 1.359375,
      "learning_rate": 3.839566987447492e-06,
      "loss": 0.0543,
      "reward": 1.609375,
      "reward_std": 0.28778618574142456,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 1200
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 396.109375,
      "epoch": 0.9608,
      "grad_norm": 0.7324569740101239,
      "kl": 0.1220703125,
      "learning_rate": 3.837798161302518e-06,
      "loss": 0.0049,
      "reward": 1.421875,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 1201
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 443.875,
      "epoch": 0.9616,
      "grad_norm": 0.4050516138647568,
      "kl": 0.7578125,
      "learning_rate": 3.836028396239297e-06,
      "loss": 0.0303,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 1202
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 340.453125,
      "epoch": 0.9624,
      "grad_norm": 2.007072847142048,
      "kl": 0.453125,
      "learning_rate": 3.8342576934999184e-06,
      "loss": 0.0182,
      "reward": 1.390625,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 1203
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 410.234375,
      "epoch": 0.9632,
      "grad_norm": 0.6148437659824656,
      "kl": 0.30859375,
      "learning_rate": 3.832486054327131e-06,
      "loss": 0.0124,
      "reward": 1.609375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 1204
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 410.921875,
      "epoch": 0.964,
      "grad_norm": 1.613091575087136,
      "kl": 0.91796875,
      "learning_rate": 3.830713479964335e-06,
      "loss": 0.0367,
      "reward": 1.796875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 1205
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 390.65625,
      "epoch": 0.9648,
      "grad_norm": 1.0370096843423675,
      "kl": 1.890625,
      "learning_rate": 3.828939971655595e-06,
      "loss": 0.0757,
      "reward": 1.609375,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.953125,
      "step": 1206
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 424.71875,
      "epoch": 0.9656,
      "grad_norm": 0.7326482459614472,
      "kl": 1.9375,
      "learning_rate": 3.827165530645627e-06,
      "loss": 0.0773,
      "reward": 1.59375,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.953125,
      "step": 1207
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 397.375,
      "epoch": 0.9664,
      "grad_norm": 0.5215922072444026,
      "kl": 0.384765625,
      "learning_rate": 3.825390158179802e-06,
      "loss": 0.0153,
      "reward": 1.515625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 1208
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 391.109375,
      "epoch": 0.9672,
      "grad_norm": 0.9092113219446152,
      "kl": 0.6171875,
      "learning_rate": 3.823613855504144e-06,
      "loss": 0.0247,
      "reward": 1.59375,
      "reward_std": 0.2041158527135849,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 1209
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 400.453125,
      "epoch": 0.968,
      "grad_norm": 0.10844602723473387,
      "kl": 0.1103515625,
      "learning_rate": 3.82183662386533e-06,
      "loss": 0.0044,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1210
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 365.171875,
      "epoch": 0.9688,
      "grad_norm": 0.5927214106596028,
      "kl": 0.1943359375,
      "learning_rate": 3.82005846451069e-06,
      "loss": 0.0078,
      "reward": 1.5625,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1211
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 398.21875,
      "epoch": 0.9696,
      "grad_norm": 4.644003560394865,
      "kl": 0.29296875,
      "learning_rate": 3.8182793786882065e-06,
      "loss": 0.0117,
      "reward": 1.65625,
      "reward_std": 0.25513994693756104,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 1212
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 369.546875,
      "epoch": 0.9704,
      "grad_norm": 0.7984385845301895,
      "kl": 0.11767578125,
      "learning_rate": 3.816499367646508e-06,
      "loss": 0.0047,
      "reward": 1.75,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1213
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 409.078125,
      "epoch": 0.9712,
      "grad_norm": 0.5037814885514016,
      "kl": 0.18359375,
      "learning_rate": 3.814718432634877e-06,
      "loss": 0.0074,
      "reward": 1.578125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 1214
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 386.921875,
      "epoch": 0.972,
      "grad_norm": 0.5793986664724105,
      "kl": 0.10400390625,
      "learning_rate": 3.8129365749032398e-06,
      "loss": 0.0042,
      "reward": 1.5625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1215
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 420.125,
      "epoch": 0.9728,
      "grad_norm": 1.0724754037429123,
      "kl": 1.2421875,
      "learning_rate": 3.8111537957021736e-06,
      "loss": 0.0499,
      "reward": 1.625,
      "reward_std": 0.37722259759902954,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.96875,
      "step": 1216
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 450.171875,
      "epoch": 0.9736,
      "grad_norm": 2.094546656882904,
      "kl": 1.375,
      "learning_rate": 3.809370096282903e-06,
      "loss": 0.0551,
      "reward": 1.703125,
      "reward_std": 0.23925508558750153,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 1217
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 444.296875,
      "epoch": 0.9744,
      "grad_norm": 0.8850136039652126,
      "kl": 0.5546875,
      "learning_rate": 3.807585477897296e-06,
      "loss": 0.0222,
      "reward": 1.640625,
      "reward_std": 0.22636085748672485,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 1218
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 411.0,
      "epoch": 0.9752,
      "grad_norm": 1.0741412183704828,
      "kl": 0.1201171875,
      "learning_rate": 3.8057999417978654e-06,
      "loss": 0.0048,
      "reward": 1.59375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1219
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 322.921875,
      "epoch": 0.976,
      "grad_norm": 0.7900524943733731,
      "kl": 0.1328125,
      "learning_rate": 3.8040134892377702e-06,
      "loss": 0.0053,
      "reward": 1.546875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 1220
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 387.71875,
      "epoch": 0.9768,
      "grad_norm": 0.082133830199952,
      "kl": 0.1162109375,
      "learning_rate": 3.802226121470811e-06,
      "loss": 0.0046,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1221
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 386.6875,
      "epoch": 0.9776,
      "grad_norm": 0.7853953238612952,
      "kl": 0.11572265625,
      "learning_rate": 3.800437839751432e-06,
      "loss": 0.0046,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1222
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 419.125,
      "epoch": 0.9784,
      "grad_norm": 0.7374526921125558,
      "kl": 0.1201171875,
      "learning_rate": 3.7986486453347183e-06,
      "loss": 0.0048,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1223
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 425.6875,
      "epoch": 0.9792,
      "grad_norm": 1.971344763280152,
      "kl": 0.140625,
      "learning_rate": 3.796858539476394e-06,
      "loss": 0.0056,
      "reward": 1.65625,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 1224
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 356.96875,
      "epoch": 0.98,
      "grad_norm": 0.43392110846331944,
      "kl": 0.1240234375,
      "learning_rate": 3.795067523432826e-06,
      "loss": 0.005,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1225
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 373.59375,
      "epoch": 0.9808,
      "grad_norm": 2.595408737762125,
      "kl": 0.228515625,
      "learning_rate": 3.793275598461017e-06,
      "loss": 0.0091,
      "reward": 1.40625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 1226
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 354.1875,
      "epoch": 0.9816,
      "grad_norm": 0.6044362340878098,
      "kl": 0.12255859375,
      "learning_rate": 3.7914827658186104e-06,
      "loss": 0.0049,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1227
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 392.46875,
      "epoch": 0.9824,
      "grad_norm": 0.9164506072852291,
      "kl": 0.1123046875,
      "learning_rate": 3.7896890267638832e-06,
      "loss": 0.0045,
      "reward": 1.65625,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1228
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 464.078125,
      "epoch": 0.9832,
      "grad_norm": 1.060333204050679,
      "kl": 0.8046875,
      "learning_rate": 3.787894382555752e-06,
      "loss": 0.0322,
      "reward": 1.671875,
      "reward_std": 0.17782479524612427,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 1229
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 402.6875,
      "epoch": 0.984,
      "grad_norm": 0.8099401116907572,
      "kl": 0.31640625,
      "learning_rate": 3.7860988344537664e-06,
      "loss": 0.0126,
      "reward": 1.625,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 1230
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 402.390625,
      "epoch": 0.9848,
      "grad_norm": 0.48940393479073535,
      "kl": 0.1220703125,
      "learning_rate": 3.7843023837181126e-06,
      "loss": 0.0049,
      "reward": 1.515625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 1231
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.5,
      "completion_length": 376.8125,
      "epoch": 0.9856,
      "grad_norm": 0.5796471326337517,
      "kl": 0.68359375,
      "learning_rate": 3.782505031609607e-06,
      "loss": 0.0273,
      "reward": 1.421875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.984375,
      "step": 1232
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 378.828125,
      "epoch": 0.9864,
      "grad_norm": 1.1469455017093382,
      "kl": 0.466796875,
      "learning_rate": 3.7807067793897006e-06,
      "loss": 0.0188,
      "reward": 1.765625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 1233
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 407.640625,
      "epoch": 0.9872,
      "grad_norm": 0.8698725511170176,
      "kl": 0.33984375,
      "learning_rate": 3.778907628320477e-06,
      "loss": 0.0136,
      "reward": 1.4375,
      "reward_std": 0.1911182403564453,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.984375,
      "step": 1234
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 370.078125,
      "epoch": 0.988,
      "grad_norm": 0.7152895003564234,
      "kl": 0.1396484375,
      "learning_rate": 3.77710757966465e-06,
      "loss": 0.0056,
      "reward": 1.796875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1235
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 374.84375,
      "epoch": 0.9888,
      "grad_norm": 0.09151468916312608,
      "kl": 0.12060546875,
      "learning_rate": 3.775306634685562e-06,
      "loss": 0.0048,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1236
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 389.5625,
      "epoch": 0.9896,
      "grad_norm": 0.8930270892171244,
      "kl": 0.2119140625,
      "learning_rate": 3.773504794647187e-06,
      "loss": 0.0085,
      "reward": 1.546875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 1237
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 390.0625,
      "epoch": 0.9904,
      "grad_norm": 0.4402466302262891,
      "kl": 0.12890625,
      "learning_rate": 3.771702060814123e-06,
      "loss": 0.0052,
      "reward": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 1238
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 374.390625,
      "epoch": 0.9912,
      "grad_norm": 0.8015938392825657,
      "kl": 0.126953125,
      "learning_rate": 3.7698984344516e-06,
      "loss": 0.0051,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1239
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 317.71875,
      "epoch": 0.992,
      "grad_norm": 2.6897329924487776,
      "kl": 0.1396484375,
      "learning_rate": 3.7680939168254733e-06,
      "loss": 0.0056,
      "reward": 1.609375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1240
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 381.171875,
      "epoch": 0.9928,
      "grad_norm": 0.8108734422973586,
      "kl": 0.1259765625,
      "learning_rate": 3.7662885092022206e-06,
      "loss": 0.005,
      "reward": 1.765625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 1241
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 423.1875,
      "epoch": 0.9936,
      "grad_norm": 0.7538924663679275,
      "kl": 1.5078125,
      "learning_rate": 3.7644822128489476e-06,
      "loss": 0.0603,
      "reward": 1.5625,
      "reward_std": 0.2619796097278595,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.953125,
      "step": 1242
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 380.328125,
      "epoch": 0.9944,
      "grad_norm": 0.4142326854937394,
      "kl": 0.123046875,
      "learning_rate": 3.7626750290333824e-06,
      "loss": 0.0049,
      "reward": 1.40625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 1243
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 431.875,
      "epoch": 0.9952,
      "grad_norm": 1.9432831031618978,
      "kl": 0.44921875,
      "learning_rate": 3.7608669590238765e-06,
      "loss": 0.0179,
      "reward": 1.6875,
      "reward_std": 0.31078842282295227,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 1244
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 398.875,
      "epoch": 0.996,
      "grad_norm": 1.7965152421438884,
      "kl": 0.78515625,
      "learning_rate": 3.7590580040894025e-06,
      "loss": 0.0315,
      "reward": 1.5625,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.96875,
      "step": 1245
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 355.953125,
      "epoch": 0.9968,
      "grad_norm": 0.8301550724750443,
      "kl": 0.6015625,
      "learning_rate": 3.7572481654995554e-06,
      "loss": 0.0241,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 1246
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 394.671875,
      "epoch": 0.9976,
      "grad_norm": 0.9261270003987451,
      "kl": 0.1435546875,
      "learning_rate": 3.755437444524548e-06,
      "loss": 0.0057,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1247
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.5,
      "completion_length": 415.3125,
      "epoch": 0.9984,
      "grad_norm": 0.678962993990223,
      "kl": 0.5,
      "learning_rate": 3.7536258424352164e-06,
      "loss": 0.0201,
      "reward": 1.390625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.984375,
      "step": 1248
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 369.421875,
      "epoch": 0.9992,
      "grad_norm": 1.117264262366207,
      "kl": 0.8828125,
      "learning_rate": 3.75181336050301e-06,
      "loss": 0.0353,
      "reward": 1.828125,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 1249
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 417.53125,
      "epoch": 1.0,
      "grad_norm": 0.3650120709042497,
      "kl": 0.1396484375,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.0056,
      "reward": 1.53125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1250
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 410.21875,
      "epoch": 1.0008,
      "grad_norm": 0.5572048293663555,
      "kl": 0.1396484375,
      "learning_rate": 3.7481857621988734e-06,
      "loss": 0.0056,
      "reward": 1.71875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1251
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 383.171875,
      "epoch": 1.0016,
      "grad_norm": 0.9908764499175217,
      "kl": 0.1884765625,
      "learning_rate": 3.74637064837293e-06,
      "loss": 0.0075,
      "reward": 1.4375,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 1252
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 388.265625,
      "epoch": 1.0024,
      "grad_norm": 1.239940799707091,
      "kl": 1.1640625,
      "learning_rate": 3.7445546597960882e-06,
      "loss": 0.0466,
      "reward": 1.78125,
      "reward_std": 0.3413130044937134,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 1253
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 392.25,
      "epoch": 1.0032,
      "grad_norm": 1.2157593225154961,
      "kl": 1.484375,
      "learning_rate": 3.742737797742878e-06,
      "loss": 0.0591,
      "reward": 1.40625,
      "reward_std": 0.5019016861915588,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.953125,
      "step": 1254
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 407.578125,
      "epoch": 1.004,
      "grad_norm": 0.9881956568631687,
      "kl": 1.1953125,
      "learning_rate": 3.7409200634884425e-06,
      "loss": 0.0482,
      "reward": 1.671875,
      "reward_std": 0.3107786178588867,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.96875,
      "step": 1255
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 437.140625,
      "epoch": 1.0048,
      "grad_norm": 2.8231668758499793,
      "kl": 4.0,
      "learning_rate": 3.7391014583085384e-06,
      "loss": 0.1607,
      "reward": 1.734375,
      "reward_std": 0.39445874094963074,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.921875,
      "step": 1256
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 459.890625,
      "epoch": 1.0056,
      "grad_norm": 4.6616832371486545,
      "kl": 3.921875,
      "learning_rate": 3.737281983479534e-06,
      "loss": 0.1573,
      "reward": 1.65625,
      "reward_std": 0.35262539982795715,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.921875,
      "step": 1257
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 427.953125,
      "epoch": 1.0064,
      "grad_norm": 0.2940974042507545,
      "kl": 0.2392578125,
      "learning_rate": 3.735461640278404e-06,
      "loss": 0.0096,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1258
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 462.625,
      "epoch": 1.0072,
      "grad_norm": 0.587074373271045,
      "kl": 0.1572265625,
      "learning_rate": 3.733640429982738e-06,
      "loss": 0.0063,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1259
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 430.328125,
      "epoch": 1.008,
      "grad_norm": 0.7864857509468988,
      "kl": 1.640625,
      "learning_rate": 3.731818353870729e-06,
      "loss": 0.0656,
      "reward": 1.84375,
      "reward_std": 0.2709311544895172,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.953125,
      "step": 1260
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 463.234375,
      "epoch": 1.0088,
      "grad_norm": 0.9842190799681781,
      "kl": 0.123046875,
      "learning_rate": 3.729995413221183e-06,
      "loss": 0.0049,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1261
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 421.328125,
      "epoch": 1.0096,
      "grad_norm": 7.878004543869069,
      "kl": 0.271484375,
      "learning_rate": 3.7281716093135068e-06,
      "loss": 0.0109,
      "reward": 1.734375,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1262
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 416.84375,
      "epoch": 1.0104,
      "grad_norm": 4.849516653718158,
      "kl": 1.046875,
      "learning_rate": 3.726346943427719e-06,
      "loss": 0.0418,
      "reward": 1.703125,
      "reward_std": 0.26621314883232117,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.953125,
      "step": 1263
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.375,
      "completion_length": 559.78125,
      "epoch": 1.0112,
      "grad_norm": 33.10205014775728,
      "kl": 27.0,
      "learning_rate": 3.7245214168444388e-06,
      "loss": 1.081,
      "reward": 0.6875,
      "reward_std": 0.5918154716491699,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.125,
      "step": 1264
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.375,
      "completion_length": 499.515625,
      "epoch": 1.012,
      "grad_norm": 22.709782412189966,
      "kl": 17.875,
      "learning_rate": 3.722695030844891e-06,
      "loss": 0.7162,
      "reward": 1.046875,
      "reward_std": 0.6453423500061035,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.296875,
      "step": 1265
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 480.21875,
      "epoch": 1.0128,
      "grad_norm": 14.901755317161799,
      "kl": 10.875,
      "learning_rate": 3.7208677867109042e-06,
      "loss": 0.4345,
      "reward": 0.953125,
      "reward_std": 0.527733564376831,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.453125,
      "step": 1266
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.0,
      "completion_length": 460.578125,
      "epoch": 1.0136,
      "grad_norm": 6.68621373718699,
      "kl": 3.0,
      "learning_rate": 3.7190396857249087e-06,
      "loss": 0.1202,
      "reward": 1.4375,
      "reward_std": 0.5098880529403687,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.484375,
      "step": 1267
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.25,
      "completion_length": 430.765625,
      "epoch": 1.0144,
      "grad_norm": 1.1857603865634236,
      "kl": 0.16015625,
      "learning_rate": 3.7172107291699356e-06,
      "loss": 0.0064,
      "reward": 1.28125,
      "reward_std": 0.4535999596118927,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.5,
      "step": 1268
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 524.328125,
      "epoch": 1.0152,
      "grad_norm": 1.9815516032068903,
      "kl": 0.134765625,
      "learning_rate": 3.7153809183296174e-06,
      "loss": 0.0054,
      "reward": 1.609375,
      "reward_std": 0.434487521648407,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.625,
      "step": 1269
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 378.78125,
      "epoch": 1.016,
      "grad_norm": 1.1517316386147365,
      "kl": 0.1376953125,
      "learning_rate": 3.713550254488185e-06,
      "loss": 0.0055,
      "reward": 1.59375,
      "reward_std": 0.328794926404953,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.859375,
      "step": 1270
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 402.640625,
      "epoch": 1.0168,
      "grad_norm": 0.725989870887737,
      "kl": 0.126953125,
      "learning_rate": 3.7117187389304703e-06,
      "loss": 0.0051,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.9375,
      "step": 1271
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 471.40625,
      "epoch": 1.0176,
      "grad_norm": 0.3711125652334141,
      "kl": 0.1162109375,
      "learning_rate": 3.7098863729418997e-06,
      "loss": 0.0046,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1272
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 436.203125,
      "epoch": 1.0184,
      "grad_norm": 0.6912563396133344,
      "kl": 0.1328125,
      "learning_rate": 3.7080531578085e-06,
      "loss": 0.0053,
      "reward": 1.859375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1273
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 436.625,
      "epoch": 1.0192,
      "grad_norm": 0.4707630040511406,
      "kl": 0.1357421875,
      "learning_rate": 3.7062190948168906e-06,
      "loss": 0.0054,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 1274
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 423.828125,
      "epoch": 1.02,
      "grad_norm": 0.630313746945571,
      "kl": 0.1357421875,
      "learning_rate": 3.7043841852542884e-06,
      "loss": 0.0054,
      "reward": 1.796875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1275
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 433.421875,
      "epoch": 1.0208,
      "grad_norm": 0.5010734063741127,
      "kl": 0.1357421875,
      "learning_rate": 3.7025484304085035e-06,
      "loss": 0.0054,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1276
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 433.359375,
      "epoch": 1.0216,
      "grad_norm": 1.054618400071382,
      "kl": 0.1337890625,
      "learning_rate": 3.7007118315679384e-06,
      "loss": 0.0054,
      "reward": 1.875,
      "reward_std": 0.1243029236793518,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 1277
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 405.234375,
      "epoch": 1.0224,
      "grad_norm": 0.573816283179805,
      "kl": 0.14453125,
      "learning_rate": 3.6988743900215895e-06,
      "loss": 0.0058,
      "reward": 1.625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1278
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 390.015625,
      "epoch": 1.0232,
      "grad_norm": 16.83587370066281,
      "kl": 203.0,
      "learning_rate": 3.6970361070590443e-06,
      "loss": 8.1376,
      "reward": 1.640625,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 1279
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 470.953125,
      "epoch": 1.024,
      "grad_norm": 0.7086540648129875,
      "kl": 0.1181640625,
      "learning_rate": 3.695196983970481e-06,
      "loss": 0.0047,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1280
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 473.640625,
      "epoch": 1.0248,
      "grad_norm": 0.5542911690599307,
      "kl": 0.1455078125,
      "learning_rate": 3.6933570220466654e-06,
      "loss": 0.0058,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1281
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 362.046875,
      "epoch": 1.0256,
      "grad_norm": 0.389045624925544,
      "kl": 0.13671875,
      "learning_rate": 3.6915162225789546e-06,
      "loss": 0.0055,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1282
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 422.015625,
      "epoch": 1.0264,
      "grad_norm": 1.3551429041345482,
      "kl": 0.138671875,
      "learning_rate": 3.6896745868592924e-06,
      "loss": 0.0056,
      "reward": 1.703125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1283
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 435.8125,
      "epoch": 1.0272,
      "grad_norm": 0.18925265515795672,
      "kl": 0.11328125,
      "learning_rate": 3.6878321161802106e-06,
      "loss": 0.0045,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1284
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 428.203125,
      "epoch": 1.028,
      "grad_norm": 0.9443154827084691,
      "kl": 0.1259765625,
      "learning_rate": 3.685988811834823e-06,
      "loss": 0.0051,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 1285
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 450.109375,
      "epoch": 1.0288,
      "grad_norm": 0.6444994814202532,
      "kl": 0.1328125,
      "learning_rate": 3.684144675116836e-06,
      "loss": 0.0053,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1286
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 389.171875,
      "epoch": 1.0296,
      "grad_norm": 0.4733455270190008,
      "kl": 0.1259765625,
      "learning_rate": 3.682299707320532e-06,
      "loss": 0.005,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 1287
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 374.421875,
      "epoch": 1.0304,
      "grad_norm": 0.9509859105214624,
      "kl": 0.1455078125,
      "learning_rate": 3.680453909740782e-06,
      "loss": 0.0058,
      "reward": 1.578125,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 1288
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 382.890625,
      "epoch": 1.0312,
      "grad_norm": 0.7783995301911126,
      "kl": 0.1376953125,
      "learning_rate": 3.6786072836730376e-06,
      "loss": 0.0055,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1289
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 346.53125,
      "epoch": 1.032,
      "grad_norm": 0.6724494774107679,
      "kl": 0.1494140625,
      "learning_rate": 3.6767598304133325e-06,
      "loss": 0.006,
      "reward": 1.5625,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 1290
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 406.1875,
      "epoch": 1.0328,
      "grad_norm": 0.11063345904636407,
      "kl": 0.1259765625,
      "learning_rate": 3.674911551258279e-06,
      "loss": 0.005,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1291
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 401.203125,
      "epoch": 1.0336,
      "grad_norm": 0.12315637894265835,
      "kl": 0.1240234375,
      "learning_rate": 3.673062447505072e-06,
      "loss": 0.005,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1292
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 358.796875,
      "epoch": 1.0344,
      "grad_norm": 1.0541213420282047,
      "kl": 0.12890625,
      "learning_rate": 3.6712125204514836e-06,
      "loss": 0.0051,
      "reward": 1.78125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1293
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 403.5625,
      "epoch": 1.0352,
      "grad_norm": 1.1660069524803196,
      "kl": 0.1318359375,
      "learning_rate": 3.6693617713958633e-06,
      "loss": 0.0053,
      "reward": 1.59375,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1294
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 389.65625,
      "epoch": 1.036,
      "grad_norm": 0.5864727282579134,
      "kl": 0.1337890625,
      "learning_rate": 3.6675102016371387e-06,
      "loss": 0.0053,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1295
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 379.40625,
      "epoch": 1.0368,
      "grad_norm": 0.43355442362640784,
      "kl": 0.12109375,
      "learning_rate": 3.665657812474812e-06,
      "loss": 0.0049,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1296
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 386.125,
      "epoch": 1.0376,
      "grad_norm": 0.0828289450261741,
      "kl": 0.1201171875,
      "learning_rate": 3.6638046052089614e-06,
      "loss": 0.0048,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1297
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 393.5625,
      "epoch": 1.0384,
      "grad_norm": 0.08981516615883395,
      "kl": 0.1298828125,
      "learning_rate": 3.661950581140239e-06,
      "loss": 0.0052,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1298
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 376.078125,
      "epoch": 1.0392,
      "grad_norm": 1.1288553696182786,
      "kl": 0.130859375,
      "learning_rate": 3.660095741569871e-06,
      "loss": 0.0053,
      "reward": 1.703125,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1299
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 372.71875,
      "epoch": 1.04,
      "grad_norm": 1.117937867759022,
      "kl": 0.1474609375,
      "learning_rate": 3.658240087799655e-06,
      "loss": 0.0059,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1300
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 382.265625,
      "epoch": 1.0408,
      "grad_norm": 0.3479631908733468,
      "kl": 0.1318359375,
      "learning_rate": 3.6563836211319593e-06,
      "loss": 0.0053,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1301
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 381.65625,
      "epoch": 1.0416,
      "grad_norm": 0.4921254026277044,
      "kl": 0.12060546875,
      "learning_rate": 3.654526342869724e-06,
      "loss": 0.0048,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1302
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 382.046875,
      "epoch": 1.0424,
      "grad_norm": 0.08843345511752333,
      "kl": 0.12060546875,
      "learning_rate": 3.65266825431646e-06,
      "loss": 0.0048,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1303
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 412.859375,
      "epoch": 1.0432,
      "grad_norm": 0.41148349370278325,
      "kl": 0.1220703125,
      "learning_rate": 3.6508093567762425e-06,
      "loss": 0.0049,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1304
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 433.359375,
      "epoch": 1.044,
      "grad_norm": 0.7053228415897961,
      "kl": 0.123046875,
      "learning_rate": 3.6489496515537204e-06,
      "loss": 0.0049,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1305
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 369.796875,
      "epoch": 1.0448,
      "grad_norm": 2.9345924019689282,
      "kl": 0.12255859375,
      "learning_rate": 3.647089139954104e-06,
      "loss": 0.0049,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1306
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 406.328125,
      "epoch": 1.0456,
      "grad_norm": 0.46572528117117695,
      "kl": 0.12451171875,
      "learning_rate": 3.6452278232831734e-06,
      "loss": 0.005,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1307
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 392.1875,
      "epoch": 1.0464,
      "grad_norm": 0.5515119208835106,
      "kl": 0.12255859375,
      "learning_rate": 3.643365702847272e-06,
      "loss": 0.0049,
      "reward": 1.875,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1308
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 416.828125,
      "epoch": 1.0472,
      "grad_norm": 0.11326186521321348,
      "kl": 0.130859375,
      "learning_rate": 3.641502779953307e-06,
      "loss": 0.0053,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1309
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 374.5,
      "epoch": 1.048,
      "grad_norm": 0.4224649270374836,
      "kl": 0.1318359375,
      "learning_rate": 3.639639055908751e-06,
      "loss": 0.0053,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1310
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 399.78125,
      "epoch": 1.0488,
      "grad_norm": 0.31213689183525883,
      "kl": 0.12158203125,
      "learning_rate": 3.6377745320216346e-06,
      "loss": 0.0049,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1311
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 346.671875,
      "epoch": 1.0496,
      "grad_norm": 0.8880636818688568,
      "kl": 0.1279296875,
      "learning_rate": 3.635909209600555e-06,
      "loss": 0.0051,
      "reward": 1.59375,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1312
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 388.640625,
      "epoch": 1.0504,
      "grad_norm": 0.8555766336251143,
      "kl": 0.1396484375,
      "learning_rate": 3.6340430899546656e-06,
      "loss": 0.0056,
      "reward": 1.859375,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1313
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 393.46875,
      "epoch": 1.0512,
      "grad_norm": 0.6914240310267927,
      "kl": 0.11962890625,
      "learning_rate": 3.632176174393682e-06,
      "loss": 0.0048,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1314
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 408.9375,
      "epoch": 1.052,
      "grad_norm": 0.08793480129649127,
      "kl": 0.1328125,
      "learning_rate": 3.630308464227877e-06,
      "loss": 0.0053,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1315
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 354.28125,
      "epoch": 1.0528,
      "grad_norm": 0.11135048815683783,
      "kl": 0.13671875,
      "learning_rate": 3.628439960768082e-06,
      "loss": 0.0055,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1316
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 339.1875,
      "epoch": 1.0536,
      "grad_norm": 0.12706180835891803,
      "kl": 0.1357421875,
      "learning_rate": 3.6265706653256837e-06,
      "loss": 0.0054,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1317
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 434.4375,
      "epoch": 1.0544,
      "grad_norm": 0.4226628009029332,
      "kl": 0.12255859375,
      "learning_rate": 3.624700579212626e-06,
      "loss": 0.0049,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1318
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 378.875,
      "epoch": 1.0552,
      "grad_norm": 0.6815461611812511,
      "kl": 0.11328125,
      "learning_rate": 3.6228297037414077e-06,
      "loss": 0.0045,
      "reward": 1.640625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1319
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 446.6875,
      "epoch": 1.056,
      "grad_norm": 0.5380691915171303,
      "kl": 0.11328125,
      "learning_rate": 3.6209580402250816e-06,
      "loss": 0.0045,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1320
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 374.359375,
      "epoch": 1.0568,
      "grad_norm": 0.1148843685975855,
      "kl": 0.12255859375,
      "learning_rate": 3.619085589977251e-06,
      "loss": 0.0049,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1321
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 377.40625,
      "epoch": 1.0576,
      "grad_norm": 0.4208076770387228,
      "kl": 0.1259765625,
      "learning_rate": 3.617212354312076e-06,
      "loss": 0.005,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1322
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 387.296875,
      "epoch": 1.0584,
      "grad_norm": 0.07878249078730427,
      "kl": 0.11962890625,
      "learning_rate": 3.615338334544265e-06,
      "loss": 0.0048,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1323
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 357.015625,
      "epoch": 1.0592,
      "grad_norm": 0.8057658584326876,
      "kl": 0.1318359375,
      "learning_rate": 3.6134635319890763e-06,
      "loss": 0.0053,
      "reward": 1.71875,
      "reward_std": 0.17570313811302185,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1324
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 395.6875,
      "epoch": 1.06,
      "grad_norm": 0.4294003205075124,
      "kl": 0.1162109375,
      "learning_rate": 3.611587947962319e-06,
      "loss": 0.0047,
      "reward": 1.390625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 1325
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 404.578125,
      "epoch": 1.0608,
      "grad_norm": 0.7194137607156134,
      "kl": 0.1259765625,
      "learning_rate": 3.6097115837803504e-06,
      "loss": 0.005,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1326
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 419.15625,
      "epoch": 1.0616,
      "grad_norm": 0.08999590606098172,
      "kl": 0.11962890625,
      "learning_rate": 3.6078344407600744e-06,
      "loss": 0.0048,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1327
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 390.28125,
      "epoch": 1.0624,
      "grad_norm": 0.08026476914716063,
      "kl": 0.1162109375,
      "learning_rate": 3.6059565202189433e-06,
      "loss": 0.0047,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1328
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 416.28125,
      "epoch": 1.0632,
      "grad_norm": 0.44695944586322334,
      "kl": 0.119140625,
      "learning_rate": 3.604077823474954e-06,
      "loss": 0.0048,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1329
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 393.71875,
      "epoch": 1.064,
      "grad_norm": 1.7134577087326561,
      "kl": 0.111328125,
      "learning_rate": 3.6021983518466468e-06,
      "loss": 0.0044,
      "reward": 1.53125,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1330
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 376.140625,
      "epoch": 1.0648,
      "grad_norm": 0.7882508740354264,
      "kl": 0.125,
      "learning_rate": 3.600318106653108e-06,
      "loss": 0.005,
      "reward": 1.765625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1331
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 399.390625,
      "epoch": 1.0656,
      "grad_norm": 0.4124598459493163,
      "kl": 0.11474609375,
      "learning_rate": 3.5984370892139663e-06,
      "loss": 0.0046,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1332
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 430.125,
      "epoch": 1.0664,
      "grad_norm": 0.8139864545899247,
      "kl": 0.12255859375,
      "learning_rate": 3.5965553008493924e-06,
      "loss": 0.0049,
      "reward": 1.859375,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1333
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 411.359375,
      "epoch": 1.0672,
      "grad_norm": 0.5002200010531586,
      "kl": 0.1162109375,
      "learning_rate": 3.594672742880097e-06,
      "loss": 0.0046,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1334
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 440.90625,
      "epoch": 1.068,
      "grad_norm": 0.07014226607248494,
      "kl": 0.1015625,
      "learning_rate": 3.5927894166273324e-06,
      "loss": 0.0041,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1335
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 425.640625,
      "epoch": 1.0688,
      "grad_norm": 0.07627824876575508,
      "kl": 0.109375,
      "learning_rate": 3.5909053234128893e-06,
      "loss": 0.0044,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1336
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 416.28125,
      "epoch": 1.0695999999999999,
      "grad_norm": 0.4377301448094491,
      "kl": 0.11865234375,
      "learning_rate": 3.5890204645590964e-06,
      "loss": 0.0047,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1337
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 381.34375,
      "epoch": 1.0704,
      "grad_norm": 0.07781407006259818,
      "kl": 0.1181640625,
      "learning_rate": 3.5871348413888207e-06,
      "loss": 0.0047,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1338
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 392.0,
      "epoch": 1.0712,
      "grad_norm": 0.47270631237798827,
      "kl": 0.11865234375,
      "learning_rate": 3.585248455225466e-06,
      "loss": 0.0047,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1339
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 369.28125,
      "epoch": 1.072,
      "grad_norm": 0.0864109007904575,
      "kl": 0.1279296875,
      "learning_rate": 3.5833613073929684e-06,
      "loss": 0.0051,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1340
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 398.125,
      "epoch": 1.0728,
      "grad_norm": 0.9026686274543017,
      "kl": 0.11767578125,
      "learning_rate": 3.5814733992158025e-06,
      "loss": 0.0047,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1341
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 427.515625,
      "epoch": 1.0735999999999999,
      "grad_norm": 0.0923061944502762,
      "kl": 0.109375,
      "learning_rate": 3.579584732018975e-06,
      "loss": 0.0044,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1342
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 411.5,
      "epoch": 1.0744,
      "grad_norm": 0.08281407405240734,
      "kl": 0.11279296875,
      "learning_rate": 3.577695307128024e-06,
      "loss": 0.0045,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1343
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 405.953125,
      "epoch": 1.0752,
      "grad_norm": 0.4274854944873935,
      "kl": 0.119140625,
      "learning_rate": 3.5758051258690223e-06,
      "loss": 0.0048,
      "reward": 1.53125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1344
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 405.296875,
      "epoch": 1.076,
      "grad_norm": 0.43568786694628714,
      "kl": 0.12109375,
      "learning_rate": 3.5739141895685708e-06,
      "loss": 0.0048,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1345
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 427.625,
      "epoch": 1.0768,
      "grad_norm": 0.5244828704127512,
      "kl": 0.12060546875,
      "learning_rate": 3.5720224995538023e-06,
      "loss": 0.0048,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1346
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 393.859375,
      "epoch": 1.0776,
      "grad_norm": 1.0476679067026615,
      "kl": 0.1162109375,
      "learning_rate": 3.5701300571523757e-06,
      "loss": 0.0047,
      "reward": 1.46875,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 1347
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 428.71875,
      "epoch": 1.0784,
      "grad_norm": 0.39801606252683097,
      "kl": 0.1171875,
      "learning_rate": 3.5682368636924825e-06,
      "loss": 0.0047,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1348
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 405.28125,
      "epoch": 1.0792,
      "grad_norm": 0.9423569674835038,
      "kl": 0.12109375,
      "learning_rate": 3.566342920502837e-06,
      "loss": 0.0048,
      "reward": 1.640625,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 1349
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 410.421875,
      "epoch": 1.08,
      "grad_norm": 0.8476625272807433,
      "kl": 0.1064453125,
      "learning_rate": 3.564448228912682e-06,
      "loss": 0.0043,
      "reward": 1.640625,
      "reward_std": 0.2198973149061203,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1350
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 407.5,
      "epoch": 1.0808,
      "grad_norm": 1.0587495022677087,
      "kl": 0.126953125,
      "learning_rate": 3.562552790251785e-06,
      "loss": 0.0051,
      "reward": 1.875,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1351
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 392.328125,
      "epoch": 1.0816,
      "grad_norm": 0.9746242455558621,
      "kl": 0.11865234375,
      "learning_rate": 3.5606566058504377e-06,
      "loss": 0.0048,
      "reward": 1.75,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1352
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 411.40625,
      "epoch": 1.0824,
      "grad_norm": 0.8637433949095339,
      "kl": 0.12255859375,
      "learning_rate": 3.558759677039455e-06,
      "loss": 0.0049,
      "reward": 1.90625,
      "reward_std": 0.13719715178012848,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.984375,
      "step": 1353
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 460.40625,
      "epoch": 1.0832,
      "grad_norm": 0.06551415921125266,
      "kl": 0.10595703125,
      "learning_rate": 3.5568620051501755e-06,
      "loss": 0.0042,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1354
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 412.796875,
      "epoch": 1.084,
      "grad_norm": 0.7756525458796553,
      "kl": 0.1201171875,
      "learning_rate": 3.5549635915144578e-06,
      "loss": 0.0048,
      "reward": 1.890625,
      "reward_std": 0.16887325048446655,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1355
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 458.6875,
      "epoch": 1.0848,
      "grad_norm": 0.08179437760795527,
      "kl": 0.11279296875,
      "learning_rate": 3.553064437464682e-06,
      "loss": 0.0045,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1356
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 434.265625,
      "epoch": 1.0856,
      "grad_norm": 2.280415542766174,
      "kl": 0.12158203125,
      "learning_rate": 3.551164544333745e-06,
      "loss": 0.0049,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1357
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 417.25,
      "epoch": 1.0864,
      "grad_norm": 0.4581094884937273,
      "kl": 0.11083984375,
      "learning_rate": 3.549263913455069e-06,
      "loss": 0.0044,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1358
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 468.765625,
      "epoch": 1.0872,
      "grad_norm": 0.08229665237670332,
      "kl": 0.11279296875,
      "learning_rate": 3.5473625461625884e-06,
      "loss": 0.0045,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1359
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 452.375,
      "epoch": 1.088,
      "grad_norm": 0.9424553271395688,
      "kl": 0.1171875,
      "learning_rate": 3.5454604437907535e-06,
      "loss": 0.0047,
      "reward": 1.609375,
      "reward_std": 0.2198973149061203,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1360
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 441.703125,
      "epoch": 1.0888,
      "grad_norm": 0.7114250380195982,
      "kl": 0.125,
      "learning_rate": 3.543557607674537e-06,
      "loss": 0.005,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1361
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 451.84375,
      "epoch": 1.0896,
      "grad_norm": 0.7450571641255354,
      "kl": 0.103515625,
      "learning_rate": 3.54165403914942e-06,
      "loss": 0.0041,
      "reward": 1.828125,
      "reward_std": 0.16887325048446655,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 1362
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 450.84375,
      "epoch": 1.0904,
      "grad_norm": 0.4169709765198773,
      "kl": 0.1201171875,
      "learning_rate": 3.539749739551401e-06,
      "loss": 0.0048,
      "reward": 1.59375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1363
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 406.125,
      "epoch": 1.0912,
      "grad_norm": 1.069548025484488,
      "kl": 0.12255859375,
      "learning_rate": 3.53784471021699e-06,
      "loss": 0.0049,
      "reward": 1.84375,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1364
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 435.34375,
      "epoch": 1.092,
      "grad_norm": 0.10591100316080641,
      "kl": 0.10546875,
      "learning_rate": 3.535938952483211e-06,
      "loss": 0.0042,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1365
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 389.515625,
      "epoch": 1.0928,
      "grad_norm": 0.6580784303551637,
      "kl": 0.12255859375,
      "learning_rate": 3.534032467687597e-06,
      "loss": 0.0049,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1366
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 469.171875,
      "epoch": 1.0936,
      "grad_norm": 0.45277713493051125,
      "kl": 0.107421875,
      "learning_rate": 3.532125257168193e-06,
      "loss": 0.0043,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1367
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 377.40625,
      "epoch": 1.0944,
      "grad_norm": 0.26519651006330924,
      "kl": 0.1298828125,
      "learning_rate": 3.5302173222635526e-06,
      "loss": 0.0052,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1368
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 411.484375,
      "epoch": 1.0952,
      "grad_norm": 0.09805385952145897,
      "kl": 0.1123046875,
      "learning_rate": 3.5283086643127396e-06,
      "loss": 0.0045,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1369
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 439.40625,
      "epoch": 1.096,
      "grad_norm": 0.5890017494184737,
      "kl": 0.103515625,
      "learning_rate": 3.5263992846553203e-06,
      "loss": 0.0041,
      "reward": 1.796875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1370
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 447.328125,
      "epoch": 1.0968,
      "grad_norm": 0.07065941566935242,
      "kl": 0.10791015625,
      "learning_rate": 3.5244891846313733e-06,
      "loss": 0.0043,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1371
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 425.46875,
      "epoch": 1.0976,
      "grad_norm": 0.9668118573558747,
      "kl": 0.1318359375,
      "learning_rate": 3.5225783655814798e-06,
      "loss": 0.0053,
      "reward": 1.71875,
      "reward_std": 0.2709311842918396,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 1372
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 388.96875,
      "epoch": 1.0984,
      "grad_norm": 0.07215970710955445,
      "kl": 0.11474609375,
      "learning_rate": 3.520666828846726e-06,
      "loss": 0.0046,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1373
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 495.265625,
      "epoch": 1.0992,
      "grad_norm": 0.3594606902019938,
      "kl": 0.1015625,
      "learning_rate": 3.518754575768702e-06,
      "loss": 0.0041,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1374
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 440.828125,
      "epoch": 1.1,
      "grad_norm": 0.8495536664673914,
      "kl": 0.11572265625,
      "learning_rate": 3.516841607689501e-06,
      "loss": 0.0046,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1375
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 422.453125,
      "epoch": 1.1008,
      "grad_norm": 0.4668886660518382,
      "kl": 0.1142578125,
      "learning_rate": 3.5149279259517165e-06,
      "loss": 0.0046,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1376
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 418.109375,
      "epoch": 1.1016,
      "grad_norm": 0.43961505825250097,
      "kl": 0.11767578125,
      "learning_rate": 3.5130135318984454e-06,
      "loss": 0.0047,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1377
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 481.90625,
      "epoch": 1.1024,
      "grad_norm": 0.43290932955061817,
      "kl": 0.1025390625,
      "learning_rate": 3.5110984268732827e-06,
      "loss": 0.0041,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1378
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 422.890625,
      "epoch": 1.1032,
      "grad_norm": 0.07043072502096129,
      "kl": 0.1123046875,
      "learning_rate": 3.509182612220322e-06,
      "loss": 0.0045,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1379
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 466.40625,
      "epoch": 1.104,
      "grad_norm": 0.3747882985557012,
      "kl": 0.10498046875,
      "learning_rate": 3.507266089284157e-06,
      "loss": 0.0042,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1380
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 405.9375,
      "epoch": 1.1048,
      "grad_norm": 1.8793652736629933,
      "kl": 0.10888671875,
      "learning_rate": 3.5053488594098763e-06,
      "loss": 0.0044,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1381
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 391.84375,
      "epoch": 1.1056,
      "grad_norm": 0.4966571203948814,
      "kl": 0.11279296875,
      "learning_rate": 3.5034309239430664e-06,
      "loss": 0.0045,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1382
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 396.25,
      "epoch": 1.1064,
      "grad_norm": 0.12919829421090323,
      "kl": 0.1220703125,
      "learning_rate": 3.501512284229807e-06,
      "loss": 0.0049,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1383
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 364.4375,
      "epoch": 1.1072,
      "grad_norm": 0.1075597809162098,
      "kl": 0.1171875,
      "learning_rate": 3.4995929416166756e-06,
      "loss": 0.0047,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1384
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 328.953125,
      "epoch": 1.108,
      "grad_norm": 1.0073464095529048,
      "kl": 0.1259765625,
      "learning_rate": 3.4976728974507387e-06,
      "loss": 0.005,
      "reward": 1.640625,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1385
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 431.09375,
      "epoch": 1.1088,
      "grad_norm": 0.4899062847323215,
      "kl": 0.1103515625,
      "learning_rate": 3.4957521530795576e-06,
      "loss": 0.0044,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1386
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 428.1875,
      "epoch": 1.1096,
      "grad_norm": 0.8881361855371956,
      "kl": 0.10400390625,
      "learning_rate": 3.493830709851185e-06,
      "loss": 0.0042,
      "reward": 1.71875,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 1387
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 376.96875,
      "epoch": 1.1104,
      "grad_norm": 0.5312134517477242,
      "kl": 0.119140625,
      "learning_rate": 3.4919085691141636e-06,
      "loss": 0.0048,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 1388
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 418.03125,
      "epoch": 1.1112,
      "grad_norm": 1.2335427817309588,
      "kl": 0.1337890625,
      "learning_rate": 3.4899857322175252e-06,
      "loss": 0.0053,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 1389
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 447.53125,
      "epoch": 1.112,
      "grad_norm": 0.12412149321676529,
      "kl": 0.099609375,
      "learning_rate": 3.4880622005107916e-06,
      "loss": 0.004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1390
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 456.34375,
      "epoch": 1.1128,
      "grad_norm": 0.11572666283984473,
      "kl": 0.10498046875,
      "learning_rate": 3.486137975343971e-06,
      "loss": 0.0042,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1391
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 409.578125,
      "epoch": 1.1136,
      "grad_norm": 0.10714842129617189,
      "kl": 0.09619140625,
      "learning_rate": 3.484213058067559e-06,
      "loss": 0.0038,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1392
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 406.3125,
      "epoch": 1.1144,
      "grad_norm": 0.5554838829199905,
      "kl": 0.10595703125,
      "learning_rate": 3.482287450032536e-06,
      "loss": 0.0042,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1393
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 426.3125,
      "epoch": 1.1152,
      "grad_norm": 0.0853444982378338,
      "kl": 0.1044921875,
      "learning_rate": 3.4803611525903687e-06,
      "loss": 0.0042,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1394
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 425.78125,
      "epoch": 1.116,
      "grad_norm": 0.5599055800840694,
      "kl": 0.0966796875,
      "learning_rate": 3.4784341670930067e-06,
      "loss": 0.0039,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1395
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.625,
      "completion_length": 378.671875,
      "epoch": 1.1168,
      "grad_norm": 0.07257007906770074,
      "kl": 0.10400390625,
      "learning_rate": 3.4765064948928813e-06,
      "loss": 0.0042,
      "reward": 1.375,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 1396
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 413.953125,
      "epoch": 1.1176,
      "grad_norm": 0.48566788345707423,
      "kl": 0.1201171875,
      "learning_rate": 3.474578137342909e-06,
      "loss": 0.0048,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1397
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 453.265625,
      "epoch": 1.1184,
      "grad_norm": 1.0924516188881157,
      "kl": 0.1025390625,
      "learning_rate": 3.4726490957964836e-06,
      "loss": 0.0041,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1398
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 448.96875,
      "epoch": 1.1192,
      "grad_norm": 0.777686275790424,
      "kl": 0.09423828125,
      "learning_rate": 3.4707193716074816e-06,
      "loss": 0.0038,
      "reward": 1.734375,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1399
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 410.0625,
      "epoch": 1.12,
      "grad_norm": 0.10795816109103065,
      "kl": 0.1025390625,
      "learning_rate": 3.4687889661302577e-06,
      "loss": 0.0041,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1400
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 450.703125,
      "epoch": 1.1208,
      "grad_norm": 0.5445734373198614,
      "kl": 0.10888671875,
      "learning_rate": 3.466857880719645e-06,
      "loss": 0.0044,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1401
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 411.140625,
      "epoch": 1.1216,
      "grad_norm": 0.6848000310402026,
      "kl": 0.1044921875,
      "learning_rate": 3.464926116730953e-06,
      "loss": 0.0042,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 1402
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 408.875,
      "epoch": 1.1224,
      "grad_norm": 0.4923579855658618,
      "kl": 0.11328125,
      "learning_rate": 3.462993675519968e-06,
      "loss": 0.0045,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1403
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 450.0625,
      "epoch": 1.1232,
      "grad_norm": 0.6043566172446442,
      "kl": 0.10986328125,
      "learning_rate": 3.4610605584429526e-06,
      "loss": 0.0044,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1404
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 464.875,
      "epoch": 1.124,
      "grad_norm": 0.06675601359060121,
      "kl": 0.09619140625,
      "learning_rate": 3.4591267668566412e-06,
      "loss": 0.0038,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1405
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 402.359375,
      "epoch": 1.1248,
      "grad_norm": 1.0616775591146432,
      "kl": 0.12255859375,
      "learning_rate": 3.457192302118244e-06,
      "loss": 0.0049,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1406
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 453.921875,
      "epoch": 1.1256,
      "grad_norm": 1.1647138985440137,
      "kl": 0.08935546875,
      "learning_rate": 3.455257165585444e-06,
      "loss": 0.0036,
      "reward": 1.859375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1407
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 449.5625,
      "epoch": 1.1264,
      "grad_norm": 0.7541875292982689,
      "kl": 0.1015625,
      "learning_rate": 3.453321358616393e-06,
      "loss": 0.0041,
      "reward": 1.75,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1408
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 470.8125,
      "epoch": 1.1272,
      "grad_norm": 0.08380687851384189,
      "kl": 0.10205078125,
      "learning_rate": 3.4513848825697145e-06,
      "loss": 0.0041,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1409
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 474.671875,
      "epoch": 1.1280000000000001,
      "grad_norm": 0.5145916486333977,
      "kl": 0.10107421875,
      "learning_rate": 3.4494477388045035e-06,
      "loss": 0.0041,
      "reward": 1.78125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1410
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 406.109375,
      "epoch": 1.1288,
      "grad_norm": 0.543150234735062,
      "kl": 0.1162109375,
      "learning_rate": 3.4475099286803204e-06,
      "loss": 0.0047,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1411
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 430.96875,
      "epoch": 1.1296,
      "grad_norm": 0.7635360148609007,
      "kl": 0.11669921875,
      "learning_rate": 3.445571453557196e-06,
      "loss": 0.0047,
      "reward": 1.859375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1412
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 433.828125,
      "epoch": 1.1304,
      "grad_norm": 0.07728577054889625,
      "kl": 0.1044921875,
      "learning_rate": 3.443632314795627e-06,
      "loss": 0.0042,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1413
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 473.34375,
      "epoch": 1.1312,
      "grad_norm": 0.06768809458452375,
      "kl": 0.10400390625,
      "learning_rate": 3.4416925137565756e-06,
      "loss": 0.0042,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1414
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 439.59375,
      "epoch": 1.1320000000000001,
      "grad_norm": 0.5746438413962025,
      "kl": 0.0986328125,
      "learning_rate": 3.439752051801467e-06,
      "loss": 0.004,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1415
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 468.4375,
      "epoch": 1.1328,
      "grad_norm": 0.433038922604871,
      "kl": 0.10400390625,
      "learning_rate": 3.4378109302921946e-06,
      "loss": 0.0042,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1416
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 461.546875,
      "epoch": 1.1336,
      "grad_norm": 0.6659817414420092,
      "kl": 0.11181640625,
      "learning_rate": 3.4358691505911105e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1417
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 472.59375,
      "epoch": 1.1344,
      "grad_norm": 0.6911444735566226,
      "kl": 0.09814453125,
      "learning_rate": 3.4339267140610317e-06,
      "loss": 0.0039,
      "reward": 1.671875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1418
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 476.625,
      "epoch": 1.1352,
      "grad_norm": 0.09875310122596062,
      "kl": 0.11474609375,
      "learning_rate": 3.4319836220652334e-06,
      "loss": 0.0046,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1419
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 402.859375,
      "epoch": 1.1360000000000001,
      "grad_norm": 0.8264825473284627,
      "kl": 0.12109375,
      "learning_rate": 3.430039875967454e-06,
      "loss": 0.0048,
      "reward": 1.75,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1420
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 415.609375,
      "epoch": 1.1368,
      "grad_norm": 0.6031482260506018,
      "kl": 0.12158203125,
      "learning_rate": 3.428095477131888e-06,
      "loss": 0.0049,
      "reward": 1.515625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 1421
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 393.78125,
      "epoch": 1.1376,
      "grad_norm": 0.4291622496601647,
      "kl": 0.09912109375,
      "learning_rate": 3.4261504269231904e-06,
      "loss": 0.004,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1422
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 416.359375,
      "epoch": 1.1384,
      "grad_norm": 0.6574559335779393,
      "kl": 0.125,
      "learning_rate": 3.4242047267064714e-06,
      "loss": 0.005,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1423
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 459.078125,
      "epoch": 1.1392,
      "grad_norm": 0.7045531699568626,
      "kl": 0.1142578125,
      "learning_rate": 3.4222583778472997e-06,
      "loss": 0.0046,
      "reward": 1.8125,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 1424
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 468.078125,
      "epoch": 1.1400000000000001,
      "grad_norm": 0.10368832579911602,
      "kl": 0.111328125,
      "learning_rate": 3.4203113817116955e-06,
      "loss": 0.0044,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1425
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 463.609375,
      "epoch": 1.1408,
      "grad_norm": 0.7714420431453896,
      "kl": 0.11669921875,
      "learning_rate": 3.4183637396661372e-06,
      "loss": 0.0047,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1426
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 459.265625,
      "epoch": 1.1416,
      "grad_norm": 21.463436447860914,
      "kl": 304.0,
      "learning_rate": 3.4164154530775552e-06,
      "loss": 12.1845,
      "reward": 1.4375,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.984375,
      "step": 1427
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 430.328125,
      "epoch": 1.1424,
      "grad_norm": 0.463088686013149,
      "kl": 0.107421875,
      "learning_rate": 3.4144665233133318e-06,
      "loss": 0.0043,
      "reward": 1.4375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 1428
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 426.59375,
      "epoch": 1.1432,
      "grad_norm": 0.5761050088039148,
      "kl": 0.109375,
      "learning_rate": 3.4125169517413005e-06,
      "loss": 0.0044,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 1429
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.625,
      "completion_length": 386.96875,
      "epoch": 1.144,
      "grad_norm": 0.08700823346346177,
      "kl": 0.119140625,
      "learning_rate": 3.410566739729746e-06,
      "loss": 0.0048,
      "reward": 1.375,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 1430
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 402.375,
      "epoch": 1.1448,
      "grad_norm": 0.644533417538252,
      "kl": 0.1103515625,
      "learning_rate": 3.408615888647402e-06,
      "loss": 0.0044,
      "reward": 1.734375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1431
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 444.53125,
      "epoch": 1.1456,
      "grad_norm": 0.5729802767468924,
      "kl": 0.10693359375,
      "learning_rate": 3.4066643998634506e-06,
      "loss": 0.0043,
      "reward": 1.84375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1432
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 464.359375,
      "epoch": 1.1464,
      "grad_norm": 0.06954289731233868,
      "kl": 0.1083984375,
      "learning_rate": 3.4047122747475227e-06,
      "loss": 0.0043,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1433
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 450.796875,
      "epoch": 1.1472,
      "grad_norm": 0.10797523245038422,
      "kl": 0.10888671875,
      "learning_rate": 3.402759514669694e-06,
      "loss": 0.0043,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1434
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 381.15625,
      "epoch": 1.148,
      "grad_norm": 0.4994777175919268,
      "kl": 0.10595703125,
      "learning_rate": 3.4008061210004872e-06,
      "loss": 0.0042,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1435
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 419.53125,
      "epoch": 1.1488,
      "grad_norm": 0.15145116729361308,
      "kl": 0.1162109375,
      "learning_rate": 3.3988520951108683e-06,
      "loss": 0.0047,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1436
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 450.1875,
      "epoch": 1.1496,
      "grad_norm": 0.453908518354673,
      "kl": 0.0966796875,
      "learning_rate": 3.3968974383722497e-06,
      "loss": 0.0039,
      "reward": 1.53125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1437
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 420.296875,
      "epoch": 1.1504,
      "grad_norm": 0.08309227989880606,
      "kl": 0.10791015625,
      "learning_rate": 3.3949421521564825e-06,
      "loss": 0.0043,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1438
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 465.265625,
      "epoch": 1.1512,
      "grad_norm": 1.0436547874273405,
      "kl": 0.109375,
      "learning_rate": 3.392986237835863e-06,
      "loss": 0.0044,
      "reward": 1.703125,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1439
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 461.875,
      "epoch": 1.152,
      "grad_norm": 0.6227114845892602,
      "kl": 0.0986328125,
      "learning_rate": 3.391029696783127e-06,
      "loss": 0.0039,
      "reward": 1.71875,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1440
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 417.78125,
      "epoch": 1.1528,
      "grad_norm": 0.06476405521823399,
      "kl": 0.09375,
      "learning_rate": 3.389072530371451e-06,
      "loss": 0.0037,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1441
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 441.453125,
      "epoch": 1.1536,
      "grad_norm": 0.9585479878506278,
      "kl": 0.1044921875,
      "learning_rate": 3.3871147399744482e-06,
      "loss": 0.0042,
      "reward": 1.703125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1442
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 424.28125,
      "epoch": 1.1544,
      "grad_norm": 0.06163943899833009,
      "kl": 0.1044921875,
      "learning_rate": 3.385156326966173e-06,
      "loss": 0.0042,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1443
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 433.78125,
      "epoch": 1.1552,
      "grad_norm": 1.1113670329603564,
      "kl": 0.11181640625,
      "learning_rate": 3.383197292721114e-06,
      "loss": 0.0045,
      "reward": 1.703125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1444
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 435.09375,
      "epoch": 1.156,
      "grad_norm": 0.10695649373521311,
      "kl": 0.10498046875,
      "learning_rate": 3.3812376386141966e-06,
      "loss": 0.0042,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1445
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 396.921875,
      "epoch": 1.1568,
      "grad_norm": 0.06801156277794704,
      "kl": 0.10009765625,
      "learning_rate": 3.379277366020782e-06,
      "loss": 0.004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1446
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 418.734375,
      "epoch": 1.1576,
      "grad_norm": 0.3580823813076655,
      "kl": 0.10498046875,
      "learning_rate": 3.3773164763166653e-06,
      "loss": 0.0042,
      "reward": 1.9375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 1447
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 464.828125,
      "epoch": 1.1584,
      "grad_norm": 0.663091035201307,
      "kl": 0.09716796875,
      "learning_rate": 3.3753549708780736e-06,
      "loss": 0.0039,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1448
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 373.609375,
      "epoch": 1.1592,
      "grad_norm": 0.6667749238816655,
      "kl": 0.1142578125,
      "learning_rate": 3.3733928510816677e-06,
      "loss": 0.0046,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1449
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 410.5,
      "epoch": 1.16,
      "grad_norm": 0.6968065727793826,
      "kl": 0.1015625,
      "learning_rate": 3.3714301183045382e-06,
      "loss": 0.0041,
      "reward": 1.578125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 1450
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 451.0625,
      "epoch": 1.1608,
      "grad_norm": 0.44453937218899353,
      "kl": 0.10107421875,
      "learning_rate": 3.369466773924207e-06,
      "loss": 0.004,
      "reward": 1.59375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1451
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 419.578125,
      "epoch": 1.1616,
      "grad_norm": 0.5322992184165442,
      "kl": 0.10546875,
      "learning_rate": 3.3675028193186243e-06,
      "loss": 0.0042,
      "reward": 1.796875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1452
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 413.265625,
      "epoch": 1.1623999999999999,
      "grad_norm": 0.4982598509555196,
      "kl": 0.1083984375,
      "learning_rate": 3.365538255866169e-06,
      "loss": 0.0043,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1453
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 454.90625,
      "epoch": 1.1632,
      "grad_norm": 0.66166595330765,
      "kl": 0.109375,
      "learning_rate": 3.3635730849456484e-06,
      "loss": 0.0044,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1454
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 440.59375,
      "epoch": 1.164,
      "grad_norm": 0.8889262086198133,
      "kl": 0.107421875,
      "learning_rate": 3.3616073079362925e-06,
      "loss": 0.0043,
      "reward": 1.765625,
      "reward_std": 0.17782479524612427,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1455
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 440.0,
      "epoch": 1.1648,
      "grad_norm": 2.5141442050434963,
      "kl": 0.099609375,
      "learning_rate": 3.3596409262177633e-06,
      "loss": 0.004,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1456
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 463.078125,
      "epoch": 1.1656,
      "grad_norm": 0.6616652595910705,
      "kl": 0.103515625,
      "learning_rate": 3.357673941170139e-06,
      "loss": 0.0041,
      "reward": 1.78125,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1457
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 406.59375,
      "epoch": 1.1663999999999999,
      "grad_norm": 0.6832318765926045,
      "kl": 0.11328125,
      "learning_rate": 3.3557063541739283e-06,
      "loss": 0.0045,
      "reward": 1.609375,
      "reward_std": 0.17782479524612427,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1458
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 476.078125,
      "epoch": 1.1672,
      "grad_norm": 0.6721377170179303,
      "kl": 0.10498046875,
      "learning_rate": 3.353738166610058e-06,
      "loss": 0.0042,
      "reward": 1.65625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1459
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 380.140625,
      "epoch": 1.168,
      "grad_norm": 0.6020334048192396,
      "kl": 0.1142578125,
      "learning_rate": 3.35176937985988e-06,
      "loss": 0.0046,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1460
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 384.109375,
      "epoch": 1.1688,
      "grad_norm": 0.48014412954771984,
      "kl": 0.109375,
      "learning_rate": 3.349799995305162e-06,
      "loss": 0.0044,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1461
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 414.734375,
      "epoch": 1.1696,
      "grad_norm": 0.341451409321828,
      "kl": 0.09130859375,
      "learning_rate": 3.3478300143280946e-06,
      "loss": 0.0036,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1462
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 453.0,
      "epoch": 1.1703999999999999,
      "grad_norm": 0.0639825437004834,
      "kl": 0.095703125,
      "learning_rate": 3.3458594383112868e-06,
      "loss": 0.0038,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1463
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 444.078125,
      "epoch": 1.1712,
      "grad_norm": 0.5032444666172071,
      "kl": 0.09326171875,
      "learning_rate": 3.343888268637765e-06,
      "loss": 0.0037,
      "reward": 1.65625,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1464
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 418.28125,
      "epoch": 1.172,
      "grad_norm": 0.44208712896677865,
      "kl": 0.10400390625,
      "learning_rate": 3.341916506690971e-06,
      "loss": 0.0042,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1465
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 441.703125,
      "epoch": 1.1728,
      "grad_norm": 0.36290573465269566,
      "kl": 0.0986328125,
      "learning_rate": 3.3399441538547638e-06,
      "loss": 0.0039,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1466
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 429.25,
      "epoch": 1.1736,
      "grad_norm": 1.100507848866221,
      "kl": 0.10302734375,
      "learning_rate": 3.337971211513417e-06,
      "loss": 0.0041,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1467
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 422.359375,
      "epoch": 1.1743999999999999,
      "grad_norm": 0.06864927055867691,
      "kl": 0.099609375,
      "learning_rate": 3.3359976810516164e-06,
      "loss": 0.004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1468
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 432.28125,
      "epoch": 1.1752,
      "grad_norm": 0.7475973430320392,
      "kl": 0.1181640625,
      "learning_rate": 3.3340235638544633e-06,
      "loss": 0.0047,
      "reward": 1.6875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1469
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 428.21875,
      "epoch": 1.176,
      "grad_norm": 1.3264637736722542,
      "kl": 0.09716796875,
      "learning_rate": 3.332048861307467e-06,
      "loss": 0.0039,
      "reward": 1.78125,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1470
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 383.96875,
      "epoch": 1.1768,
      "grad_norm": 0.6959989823906907,
      "kl": 0.1064453125,
      "learning_rate": 3.330073574796551e-06,
      "loss": 0.0043,
      "reward": 1.5,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1471
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 420.8125,
      "epoch": 1.1776,
      "grad_norm": 0.87072487103619,
      "kl": 0.107421875,
      "learning_rate": 3.328097705708047e-06,
      "loss": 0.0043,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1472
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 420.40625,
      "epoch": 1.1784,
      "grad_norm": 0.5768313122779916,
      "kl": 0.11181640625,
      "learning_rate": 3.3261212554286977e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1473
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 439.1875,
      "epoch": 1.1792,
      "grad_norm": 0.07694570407810193,
      "kl": 0.11328125,
      "learning_rate": 3.324144225345649e-06,
      "loss": 0.0045,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1474
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 440.515625,
      "epoch": 1.18,
      "grad_norm": 0.09631195615765224,
      "kl": 0.109375,
      "learning_rate": 3.3221666168464584e-06,
      "loss": 0.0044,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1475
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 384.5,
      "epoch": 1.1808,
      "grad_norm": 0.4622244083058771,
      "kl": 0.1220703125,
      "learning_rate": 3.320188431319088e-06,
      "loss": 0.0049,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1476
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 438.4375,
      "epoch": 1.1816,
      "grad_norm": 0.6939857811376071,
      "kl": 0.09912109375,
      "learning_rate": 3.318209670151904e-06,
      "loss": 0.004,
      "reward": 1.75,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1477
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 377.78125,
      "epoch": 1.1824,
      "grad_norm": 0.5453108696690133,
      "kl": 0.1025390625,
      "learning_rate": 3.3162303347336765e-06,
      "loss": 0.0041,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1478
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 405.703125,
      "epoch": 1.1832,
      "grad_norm": 0.7482532247766777,
      "kl": 0.10888671875,
      "learning_rate": 3.3142504264535808e-06,
      "loss": 0.0044,
      "reward": 1.609375,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1479
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 441.859375,
      "epoch": 1.184,
      "grad_norm": 0.7719143253082762,
      "kl": 0.1015625,
      "learning_rate": 3.3122699467011913e-06,
      "loss": 0.0041,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1480
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 425.578125,
      "epoch": 1.1848,
      "grad_norm": 0.07207211889396034,
      "kl": 0.1044921875,
      "learning_rate": 3.3102888968664857e-06,
      "loss": 0.0042,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1481
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 432.671875,
      "epoch": 1.1856,
      "grad_norm": 0.4440499987515191,
      "kl": 0.10009765625,
      "learning_rate": 3.308307278339842e-06,
      "loss": 0.004,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1482
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 458.9375,
      "epoch": 1.1864,
      "grad_norm": 0.44469879596118,
      "kl": 0.0966796875,
      "learning_rate": 3.306325092512034e-06,
      "loss": 0.0039,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1483
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 431.578125,
      "epoch": 1.1872,
      "grad_norm": 0.5874152839811555,
      "kl": 0.10205078125,
      "learning_rate": 3.3043423407742374e-06,
      "loss": 0.0041,
      "reward": 1.46875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 1484
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 435.671875,
      "epoch": 1.188,
      "grad_norm": 0.4305572452110268,
      "kl": 0.0966796875,
      "learning_rate": 3.3023590245180237e-06,
      "loss": 0.0039,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1485
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 413.5625,
      "epoch": 1.1888,
      "grad_norm": 0.6724993029168898,
      "kl": 0.1123046875,
      "learning_rate": 3.300375145135361e-06,
      "loss": 0.0045,
      "reward": 1.703125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 1486
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 455.5,
      "epoch": 1.1896,
      "grad_norm": 0.7374735676474072,
      "kl": 0.09375,
      "learning_rate": 3.2983907040186112e-06,
      "loss": 0.0037,
      "reward": 1.84375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1487
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 466.265625,
      "epoch": 1.1904,
      "grad_norm": 0.39588011924560756,
      "kl": 0.0947265625,
      "learning_rate": 3.296405702560532e-06,
      "loss": 0.0038,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1488
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 361.375,
      "epoch": 1.1912,
      "grad_norm": 0.4409189442063453,
      "kl": 0.11474609375,
      "learning_rate": 3.294420142154274e-06,
      "loss": 0.0046,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1489
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 436.71875,
      "epoch": 1.192,
      "grad_norm": 0.3752695753607167,
      "kl": 0.09326171875,
      "learning_rate": 3.29243402419338e-06,
      "loss": 0.0037,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1490
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 387.78125,
      "epoch": 1.1928,
      "grad_norm": 0.59116556013629,
      "kl": 0.0927734375,
      "learning_rate": 3.2904473500717826e-06,
      "loss": 0.0037,
      "reward": 1.78125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1491
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 378.640625,
      "epoch": 1.1936,
      "grad_norm": 0.8800352480705476,
      "kl": 0.11767578125,
      "learning_rate": 3.2884601211838087e-06,
      "loss": 0.0047,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1492
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 426.53125,
      "epoch": 1.1944,
      "grad_norm": 0.06276444903383693,
      "kl": 0.0986328125,
      "learning_rate": 3.2864723389241697e-06,
      "loss": 0.0039,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1493
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 405.96875,
      "epoch": 1.1952,
      "grad_norm": 0.637470041365548,
      "kl": 0.12109375,
      "learning_rate": 3.284484004687969e-06,
      "loss": 0.0049,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1494
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 372.65625,
      "epoch": 1.196,
      "grad_norm": 0.08157808922861641,
      "kl": 0.11279296875,
      "learning_rate": 3.2824951198706958e-06,
      "loss": 0.0045,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1495
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 399.1875,
      "epoch": 1.1968,
      "grad_norm": 0.5575219576538066,
      "kl": 0.095703125,
      "learning_rate": 3.280505685868226e-06,
      "loss": 0.0038,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1496
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 455.21875,
      "epoch": 1.1976,
      "grad_norm": 0.35317690682621256,
      "kl": 0.0947265625,
      "learning_rate": 3.278515704076821e-06,
      "loss": 0.0038,
      "reward": 1.9375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 1497
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 451.03125,
      "epoch": 1.1984,
      "grad_norm": 0.7645963540928052,
      "kl": 0.0947265625,
      "learning_rate": 3.276525175893126e-06,
      "loss": 0.0038,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1498
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.5,
      "completion_length": 389.84375,
      "epoch": 1.1992,
      "grad_norm": 0.1171126630216482,
      "kl": 0.1044921875,
      "learning_rate": 3.274534102714172e-06,
      "loss": 0.0042,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1499
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 459.328125,
      "epoch": 1.2,
      "grad_norm": 0.5239497582573815,
      "kl": 0.10302734375,
      "learning_rate": 3.272542485937369e-06,
      "loss": 0.0041,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1500
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 412.65625,
      "epoch": 1.2008,
      "grad_norm": 0.7470058520218157,
      "kl": 0.103515625,
      "learning_rate": 3.270550326960511e-06,
      "loss": 0.0041,
      "reward": 1.71875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1501
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 374.671875,
      "epoch": 1.2016,
      "grad_norm": 1.1478656805215206,
      "kl": 0.107421875,
      "learning_rate": 3.268557627181772e-06,
      "loss": 0.0043,
      "reward": 1.78125,
      "reward_std": 0.23827511072158813,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1502
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 479.53125,
      "epoch": 1.2024,
      "grad_norm": 1.2596014404408202,
      "kl": 0.111328125,
      "learning_rate": 3.2665643879997054e-06,
      "loss": 0.0044,
      "reward": 1.6875,
      "reward_std": 0.19506090879440308,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 1503
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 403.859375,
      "epoch": 1.2032,
      "grad_norm": 0.5155330501641545,
      "kl": 0.10498046875,
      "learning_rate": 3.2645706108132426e-06,
      "loss": 0.0042,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1504
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 420.8125,
      "epoch": 1.204,
      "grad_norm": 0.08151666120758788,
      "kl": 0.09326171875,
      "learning_rate": 3.2625762970216944e-06,
      "loss": 0.0037,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1505
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 395.65625,
      "epoch": 1.2048,
      "grad_norm": 0.07530023716969772,
      "kl": 0.111328125,
      "learning_rate": 3.2605814480247454e-06,
      "loss": 0.0044,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1506
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 431.234375,
      "epoch": 1.2056,
      "grad_norm": 0.5458788856505151,
      "kl": 0.09765625,
      "learning_rate": 3.258586065222459e-06,
      "loss": 0.0039,
      "reward": 1.5,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1507
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 409.5625,
      "epoch": 1.2064,
      "grad_norm": 0.6878005837117955,
      "kl": 0.10546875,
      "learning_rate": 3.2565901500152702e-06,
      "loss": 0.0042,
      "reward": 1.453125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 1508
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 449.34375,
      "epoch": 1.2072,
      "grad_norm": 1.9930446515014937,
      "kl": 0.09375,
      "learning_rate": 3.2545937038039904e-06,
      "loss": 0.0037,
      "reward": 1.796875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1509
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 415.359375,
      "epoch": 1.208,
      "grad_norm": 0.6289779042784038,
      "kl": 0.103515625,
      "learning_rate": 3.2525967279898017e-06,
      "loss": 0.0041,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1510
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 488.96875,
      "epoch": 1.2088,
      "grad_norm": 1.0644143284005663,
      "kl": 0.10400390625,
      "learning_rate": 3.2505992239742582e-06,
      "loss": 0.0042,
      "reward": 1.875,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1511
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 406.5625,
      "epoch": 1.2096,
      "grad_norm": 1.0202268224441047,
      "kl": 0.10498046875,
      "learning_rate": 3.2486011931592863e-06,
      "loss": 0.0042,
      "reward": 1.765625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1512
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 488.796875,
      "epoch": 1.2104,
      "grad_norm": 0.8755874946771128,
      "kl": 0.09765625,
      "learning_rate": 3.2466026369471804e-06,
      "loss": 0.0039,
      "reward": 1.796875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1513
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 408.140625,
      "epoch": 1.2112,
      "grad_norm": 0.49639688259820164,
      "kl": 0.1044921875,
      "learning_rate": 3.2446035567406033e-06,
      "loss": 0.0042,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1514
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 411.78125,
      "epoch": 1.212,
      "grad_norm": 0.491303707987397,
      "kl": 0.10009765625,
      "learning_rate": 3.2426039539425875e-06,
      "loss": 0.004,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1515
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 442.8125,
      "epoch": 1.2128,
      "grad_norm": 0.47695448585921424,
      "kl": 0.10009765625,
      "learning_rate": 3.240603829956531e-06,
      "loss": 0.004,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1516
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 404.375,
      "epoch": 1.2136,
      "grad_norm": 0.6604175255061485,
      "kl": 0.11083984375,
      "learning_rate": 3.238603186186198e-06,
      "loss": 0.0044,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1517
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 462.0,
      "epoch": 1.2144,
      "grad_norm": 0.7893523216783938,
      "kl": 0.099609375,
      "learning_rate": 3.2366020240357166e-06,
      "loss": 0.004,
      "reward": 1.75,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1518
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 384.5,
      "epoch": 1.2152,
      "grad_norm": 0.10841424005624302,
      "kl": 0.1005859375,
      "learning_rate": 3.2346003449095803e-06,
      "loss": 0.004,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1519
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 421.65625,
      "epoch": 1.216,
      "grad_norm": 0.09099275239237044,
      "kl": 0.087890625,
      "learning_rate": 3.2325981502126434e-06,
      "loss": 0.0035,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1520
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 422.765625,
      "epoch": 1.2168,
      "grad_norm": 0.6016652936245342,
      "kl": 0.09814453125,
      "learning_rate": 3.2305954413501252e-06,
      "loss": 0.0039,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1521
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 422.90625,
      "epoch": 1.2176,
      "grad_norm": 0.746887381046538,
      "kl": 0.08984375,
      "learning_rate": 3.228592219727602e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1522
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 467.671875,
      "epoch": 1.2184,
      "grad_norm": 0.4274887556891549,
      "kl": 0.0947265625,
      "learning_rate": 3.226588486751012e-06,
      "loss": 0.0038,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1523
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 375.046875,
      "epoch": 1.2192,
      "grad_norm": 0.4559184408879622,
      "kl": 0.0966796875,
      "learning_rate": 3.2245842438266526e-06,
      "loss": 0.0039,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1524
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 452.140625,
      "epoch": 1.22,
      "grad_norm": 0.060753100467566594,
      "kl": 0.10400390625,
      "learning_rate": 3.222579492361179e-06,
      "loss": 0.0042,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1525
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 409.40625,
      "epoch": 1.2208,
      "grad_norm": 1.7216627459251168,
      "kl": 0.10986328125,
      "learning_rate": 3.220574233761603e-06,
      "loss": 0.0044,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1526
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 454.296875,
      "epoch": 1.2216,
      "grad_norm": 0.4211796013305398,
      "kl": 0.10107421875,
      "learning_rate": 3.2185684694352913e-06,
      "loss": 0.004,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1527
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 413.875,
      "epoch": 1.2224,
      "grad_norm": 0.4899344234500141,
      "kl": 0.087890625,
      "learning_rate": 3.216562200789968e-06,
      "loss": 0.0035,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1528
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 452.265625,
      "epoch": 1.2232,
      "grad_norm": 0.6674345375776687,
      "kl": 0.09912109375,
      "learning_rate": 3.214555429233707e-06,
      "loss": 0.004,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1529
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 475.46875,
      "epoch": 1.224,
      "grad_norm": 0.43503399027021084,
      "kl": 0.0927734375,
      "learning_rate": 3.2125481561749406e-06,
      "loss": 0.0037,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1530
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 438.203125,
      "epoch": 1.2248,
      "grad_norm": 0.5721824849540433,
      "kl": 0.0966796875,
      "learning_rate": 3.210540383022449e-06,
      "loss": 0.0039,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1531
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 446.8125,
      "epoch": 1.2256,
      "grad_norm": 0.6566101912725456,
      "kl": 0.095703125,
      "learning_rate": 3.208532111185365e-06,
      "loss": 0.0038,
      "reward": 1.734375,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1532
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 422.125,
      "epoch": 1.2264,
      "grad_norm": 0.6024691885630784,
      "kl": 0.09423828125,
      "learning_rate": 3.2065233420731717e-06,
      "loss": 0.0038,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1533
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 404.328125,
      "epoch": 1.2272,
      "grad_norm": 0.37536801298633604,
      "kl": 0.09765625,
      "learning_rate": 3.2045140770956987e-06,
      "loss": 0.0039,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1534
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 423.96875,
      "epoch": 1.228,
      "grad_norm": 0.06282124859248339,
      "kl": 0.10107421875,
      "learning_rate": 3.2025043176631283e-06,
      "loss": 0.004,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1535
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 374.921875,
      "epoch": 1.2288000000000001,
      "grad_norm": 0.07562155529442534,
      "kl": 0.10400390625,
      "learning_rate": 3.2004940651859844e-06,
      "loss": 0.0042,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1536
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 427.5625,
      "epoch": 1.2296,
      "grad_norm": 0.9258511371248414,
      "kl": 0.0869140625,
      "learning_rate": 3.198483321075141e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1537
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 422.625,
      "epoch": 1.2304,
      "grad_norm": 0.07953989457996766,
      "kl": 0.09375,
      "learning_rate": 3.196472086741815e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1538
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 434.453125,
      "epoch": 1.2312,
      "grad_norm": 0.3698225889729225,
      "kl": 0.08935546875,
      "learning_rate": 3.194460363597569e-06,
      "loss": 0.0036,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1539
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 370.359375,
      "epoch": 1.232,
      "grad_norm": 0.7287457368623393,
      "kl": 0.11181640625,
      "learning_rate": 3.192448153054306e-06,
      "loss": 0.0045,
      "reward": 1.71875,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1540
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 460.703125,
      "epoch": 1.2328000000000001,
      "grad_norm": 1.4674304521597479,
      "kl": 0.08447265625,
      "learning_rate": 3.190435456524275e-06,
      "loss": 0.0034,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1541
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 447.65625,
      "epoch": 1.2336,
      "grad_norm": 0.6147338626869835,
      "kl": 0.09814453125,
      "learning_rate": 3.188422275420063e-06,
      "loss": 0.0039,
      "reward": 1.734375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1542
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 480.265625,
      "epoch": 1.2344,
      "grad_norm": 0.6045595821869633,
      "kl": 0.08837890625,
      "learning_rate": 3.186408611154597e-06,
      "loss": 0.0035,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1543
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 405.015625,
      "epoch": 1.2352,
      "grad_norm": 0.519945223557907,
      "kl": 0.095703125,
      "learning_rate": 3.184394465141146e-06,
      "loss": 0.0038,
      "reward": 1.9375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 1544
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 409.515625,
      "epoch": 1.236,
      "grad_norm": 0.9373246726421136,
      "kl": 0.09912109375,
      "learning_rate": 3.1823798387933134e-06,
      "loss": 0.004,
      "reward": 1.484375,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 1545
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 443.21875,
      "epoch": 1.2368000000000001,
      "grad_norm": 0.08531400250257483,
      "kl": 0.08349609375,
      "learning_rate": 3.180364733525043e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1546
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 374.984375,
      "epoch": 1.2376,
      "grad_norm": 0.06331314257077808,
      "kl": 0.099609375,
      "learning_rate": 3.178349150750612e-06,
      "loss": 0.004,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1547
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 385.9375,
      "epoch": 1.2384,
      "grad_norm": 0.6285410098420786,
      "kl": 0.1025390625,
      "learning_rate": 3.1763330918846347e-06,
      "loss": 0.0041,
      "reward": 1.78125,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1548
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 477.09375,
      "epoch": 1.2392,
      "grad_norm": 0.4345200262968506,
      "kl": 0.09521484375,
      "learning_rate": 3.1743165583420586e-06,
      "loss": 0.0038,
      "reward": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1549
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 439.703125,
      "epoch": 1.24,
      "grad_norm": 0.06940406756956047,
      "kl": 0.091796875,
      "learning_rate": 3.1722995515381644e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1550
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 373.421875,
      "epoch": 1.2408,
      "grad_norm": 0.7915566012453998,
      "kl": 0.11376953125,
      "learning_rate": 3.1702820728885657e-06,
      "loss": 0.0045,
      "reward": 1.5625,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1551
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 424.46875,
      "epoch": 1.2416,
      "grad_norm": 0.6619017754194848,
      "kl": 0.0927734375,
      "learning_rate": 3.1682641238092064e-06,
      "loss": 0.0037,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1552
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 383.375,
      "epoch": 1.2424,
      "grad_norm": 0.4864696985125599,
      "kl": 0.0859375,
      "learning_rate": 3.1662457057163603e-06,
      "loss": 0.0034,
      "reward": 1.71875,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1553
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 417.90625,
      "epoch": 1.2432,
      "grad_norm": 1.0619547205515607,
      "kl": 0.09716796875,
      "learning_rate": 3.164226820026632e-06,
      "loss": 0.0039,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1554
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 413.0,
      "epoch": 1.244,
      "grad_norm": 0.9950719878007777,
      "kl": 0.09765625,
      "learning_rate": 3.162207468156952e-06,
      "loss": 0.0039,
      "reward": 1.765625,
      "reward_std": 0.16887325048446655,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1555
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 398.859375,
      "epoch": 1.2448,
      "grad_norm": 1.025372167284398,
      "kl": 0.10546875,
      "learning_rate": 3.16018765152458e-06,
      "loss": 0.0042,
      "reward": 1.640625,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 1556
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 400.75,
      "epoch": 1.2456,
      "grad_norm": 0.48661050828794217,
      "kl": 0.10009765625,
      "learning_rate": 3.1581673715471007e-06,
      "loss": 0.004,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1557
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 433.546875,
      "epoch": 1.2464,
      "grad_norm": 0.5652765630430348,
      "kl": 0.08642578125,
      "learning_rate": 3.1561466296424247e-06,
      "loss": 0.0035,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1558
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 428.28125,
      "epoch": 1.2472,
      "grad_norm": 0.9513859294472352,
      "kl": 0.0947265625,
      "learning_rate": 3.154125427228786e-06,
      "loss": 0.0038,
      "reward": 1.75,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1559
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 360.296875,
      "epoch": 1.248,
      "grad_norm": 0.9009921272927971,
      "kl": 0.099609375,
      "learning_rate": 3.152103765724743e-06,
      "loss": 0.004,
      "reward": 1.671875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1560
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 378.09375,
      "epoch": 1.2488,
      "grad_norm": 0.8178067423041044,
      "kl": 0.10302734375,
      "learning_rate": 3.150081646549174e-06,
      "loss": 0.0041,
      "reward": 1.734375,
      "reward_std": 0.1893727034330368,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1561
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 427.34375,
      "epoch": 1.2496,
      "grad_norm": 0.38434557649415374,
      "kl": 0.09814453125,
      "learning_rate": 3.1480590711212823e-06,
      "loss": 0.0039,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1562
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 435.78125,
      "epoch": 1.2504,
      "grad_norm": 0.7944107285153403,
      "kl": 0.09130859375,
      "learning_rate": 3.1460360408605866e-06,
      "loss": 0.0037,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1563
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 408.609375,
      "epoch": 1.2511999999999999,
      "grad_norm": 0.06691646785805336,
      "kl": 0.10546875,
      "learning_rate": 3.144012557186931e-06,
      "loss": 0.0042,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1564
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 424.09375,
      "epoch": 1.252,
      "grad_norm": 0.7682463850209427,
      "kl": 0.1064453125,
      "learning_rate": 3.14198862152047e-06,
      "loss": 0.0043,
      "reward": 1.734375,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1565
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 426.046875,
      "epoch": 1.2528000000000001,
      "grad_norm": 0.06225457269389659,
      "kl": 0.08935546875,
      "learning_rate": 3.1399642352816825e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1566
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 390.9375,
      "epoch": 1.2536,
      "grad_norm": 0.9374596011892233,
      "kl": 0.103515625,
      "learning_rate": 3.1379393998913594e-06,
      "loss": 0.0041,
      "reward": 1.765625,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 1567
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 412.390625,
      "epoch": 1.2544,
      "grad_norm": 7.514759589636227,
      "kl": 0.09228515625,
      "learning_rate": 3.135914116770609e-06,
      "loss": 0.0037,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1568
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 392.34375,
      "epoch": 1.2551999999999999,
      "grad_norm": 0.6425695406518036,
      "kl": 0.10205078125,
      "learning_rate": 3.1338883873408517e-06,
      "loss": 0.0041,
      "reward": 1.890625,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1569
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 440.03125,
      "epoch": 1.256,
      "grad_norm": 0.4627676969714413,
      "kl": 0.087890625,
      "learning_rate": 3.1318622130238237e-06,
      "loss": 0.0035,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1570
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 433.25,
      "epoch": 1.2568,
      "grad_norm": 0.22107253422721868,
      "kl": 0.0966796875,
      "learning_rate": 3.1298355952415714e-06,
      "loss": 0.0039,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1571
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 428.0,
      "epoch": 1.2576,
      "grad_norm": 0.7907025613664066,
      "kl": 0.09716796875,
      "learning_rate": 3.127808535416454e-06,
      "loss": 0.0039,
      "reward": 1.890625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.984375,
      "step": 1572
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 449.359375,
      "epoch": 1.2584,
      "grad_norm": 0.06448354850068903,
      "kl": 0.0849609375,
      "learning_rate": 3.1257810349711388e-06,
      "loss": 0.0034,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1573
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 477.40625,
      "epoch": 1.2591999999999999,
      "grad_norm": 0.7031076936859635,
      "kl": 0.0927734375,
      "learning_rate": 3.1237530953286046e-06,
      "loss": 0.0037,
      "reward": 1.859375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1574
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 406.734375,
      "epoch": 1.26,
      "grad_norm": 0.5359139719280055,
      "kl": 0.0966796875,
      "learning_rate": 3.121724717912138e-06,
      "loss": 0.0039,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1575
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 478.421875,
      "epoch": 1.2608,
      "grad_norm": 1.1600258143376623,
      "kl": 0.09375,
      "learning_rate": 3.11969590414533e-06,
      "loss": 0.0037,
      "reward": 1.734375,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1576
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 442.21875,
      "epoch": 1.2616,
      "grad_norm": 0.5506945261688823,
      "kl": 0.09033203125,
      "learning_rate": 3.1176666554520827e-06,
      "loss": 0.0036,
      "reward": 1.734375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1577
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 435.484375,
      "epoch": 1.2624,
      "grad_norm": 1.0118338172531742,
      "kl": 0.10205078125,
      "learning_rate": 3.1156369732566006e-06,
      "loss": 0.0041,
      "reward": 1.75,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1578
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 444.390625,
      "epoch": 1.2631999999999999,
      "grad_norm": 0.73187066201207,
      "kl": 0.1025390625,
      "learning_rate": 3.113606858983391e-06,
      "loss": 0.0041,
      "reward": 1.765625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1579
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 465.265625,
      "epoch": 1.264,
      "grad_norm": 0.49609510574463794,
      "kl": 0.09765625,
      "learning_rate": 3.1115763140572686e-06,
      "loss": 0.0039,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1580
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 466.875,
      "epoch": 1.2648,
      "grad_norm": 0.4063841044076745,
      "kl": 0.08642578125,
      "learning_rate": 3.109545339903347e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1581
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 510.0,
      "epoch": 1.2656,
      "grad_norm": 0.06132187821039446,
      "kl": 0.09912109375,
      "learning_rate": 3.107513937947041e-06,
      "loss": 0.004,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1582
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 439.328125,
      "epoch": 1.2664,
      "grad_norm": 0.3537366580312689,
      "kl": 0.0927734375,
      "learning_rate": 3.1054821096140675e-06,
      "loss": 0.0037,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1583
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 401.171875,
      "epoch": 1.2671999999999999,
      "grad_norm": 0.4688301710358441,
      "kl": 0.10107421875,
      "learning_rate": 3.1034498563304435e-06,
      "loss": 0.004,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1584
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 407.703125,
      "epoch": 1.268,
      "grad_norm": 0.41789923092783626,
      "kl": 0.111328125,
      "learning_rate": 3.1014171795224794e-06,
      "loss": 0.0045,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1585
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 436.90625,
      "epoch": 1.2688,
      "grad_norm": 0.49923174987107194,
      "kl": 0.1025390625,
      "learning_rate": 3.0993840806167884e-06,
      "loss": 0.0041,
      "reward": 1.578125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 1586
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 453.90625,
      "epoch": 1.2696,
      "grad_norm": 0.5634745847349014,
      "kl": 0.08935546875,
      "learning_rate": 3.0973505610402767e-06,
      "loss": 0.0036,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1587
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 446.0625,
      "epoch": 1.2704,
      "grad_norm": 11.033931290745768,
      "kl": 0.1025390625,
      "learning_rate": 3.0953166222201474e-06,
      "loss": 0.0041,
      "reward": 1.921875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1588
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 417.484375,
      "epoch": 1.2711999999999999,
      "grad_norm": 0.6320445656018617,
      "kl": 0.1005859375,
      "learning_rate": 3.093282265583895e-06,
      "loss": 0.004,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1589
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 423.78125,
      "epoch": 1.272,
      "grad_norm": 0.5090770424043585,
      "kl": 0.0966796875,
      "learning_rate": 3.0912474925593124e-06,
      "loss": 0.0039,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1590
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 440.109375,
      "epoch": 1.2728,
      "grad_norm": 0.34673848255588985,
      "kl": 0.10107421875,
      "learning_rate": 3.0892123045744787e-06,
      "loss": 0.004,
      "reward": 1.8125,
      "reward_std": 0.09449111670255661,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 1591
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 515.8125,
      "epoch": 1.2736,
      "grad_norm": 0.5295230759151055,
      "kl": 0.09033203125,
      "learning_rate": 3.0871767030577686e-06,
      "loss": 0.0036,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1592
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 432.375,
      "epoch": 1.2744,
      "grad_norm": 0.6899275836361773,
      "kl": 0.0966796875,
      "learning_rate": 3.085140689437846e-06,
      "loss": 0.0039,
      "reward": 1.625,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1593
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 428.171875,
      "epoch": 1.2752,
      "grad_norm": 0.37769899158558207,
      "kl": 0.09716796875,
      "learning_rate": 3.0831042651436634e-06,
      "loss": 0.0039,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1594
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 381.75,
      "epoch": 1.276,
      "grad_norm": 1.3234644407004388,
      "kl": 0.10302734375,
      "learning_rate": 3.0810674316044602e-06,
      "loss": 0.0041,
      "reward": 1.5,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1595
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 380.609375,
      "epoch": 1.2768,
      "grad_norm": 1.2325874581687077,
      "kl": 0.11474609375,
      "learning_rate": 3.0790301902497664e-06,
      "loss": 0.0046,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1596
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 409.734375,
      "epoch": 1.2776,
      "grad_norm": 0.49425500489254154,
      "kl": 0.10400390625,
      "learning_rate": 3.076992542509396e-06,
      "loss": 0.0042,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1597
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 401.515625,
      "epoch": 1.2784,
      "grad_norm": 0.44881265921084124,
      "kl": 0.0927734375,
      "learning_rate": 3.0749544898134487e-06,
      "loss": 0.0037,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1598
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 427.03125,
      "epoch": 1.2792,
      "grad_norm": 0.611287629019588,
      "kl": 0.0966796875,
      "learning_rate": 3.072916033592307e-06,
      "loss": 0.0039,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1599
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 414.703125,
      "epoch": 1.28,
      "grad_norm": 0.5946965434991116,
      "kl": 0.09716796875,
      "learning_rate": 3.0708771752766397e-06,
      "loss": 0.0039,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1600
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 456.34375,
      "epoch": 1.2808,
      "grad_norm": 0.4426993076609644,
      "kl": 0.095703125,
      "learning_rate": 3.068837916297396e-06,
      "loss": 0.0038,
      "reward": 1.9375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 1601
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 422.859375,
      "epoch": 1.2816,
      "grad_norm": 1.1339435328216212,
      "kl": 0.099609375,
      "learning_rate": 3.0667982580858047e-06,
      "loss": 0.004,
      "reward": 1.53125,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 1602
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 416.890625,
      "epoch": 1.2824,
      "grad_norm": 0.9071102702262377,
      "kl": 0.095703125,
      "learning_rate": 3.0647582020733773e-06,
      "loss": 0.0038,
      "reward": 1.828125,
      "reward_std": 0.2109457403421402,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1603
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 419.9375,
      "epoch": 1.2832,
      "grad_norm": 0.0676820144422039,
      "kl": 0.09375,
      "learning_rate": 3.062717749691904e-06,
      "loss": 0.0038,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1604
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 459.234375,
      "epoch": 1.284,
      "grad_norm": 0.41953849970425167,
      "kl": 0.095703125,
      "learning_rate": 3.0606769023734535e-06,
      "loss": 0.0038,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1605
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 434.953125,
      "epoch": 1.2848,
      "grad_norm": 0.0648339639969181,
      "kl": 0.0849609375,
      "learning_rate": 3.0586356615503693e-06,
      "loss": 0.0034,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1606
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 399.515625,
      "epoch": 1.2856,
      "grad_norm": 0.48602338854803356,
      "kl": 0.10009765625,
      "learning_rate": 3.056594028655274e-06,
      "loss": 0.004,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1607
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 448.34375,
      "epoch": 1.2864,
      "grad_norm": 0.8451809274469703,
      "kl": 0.09033203125,
      "learning_rate": 3.0545520051210637e-06,
      "loss": 0.0036,
      "reward": 1.84375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1608
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 441.515625,
      "epoch": 1.2872,
      "grad_norm": 0.0589542855636287,
      "kl": 0.09326171875,
      "learning_rate": 3.052509592380909e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1609
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 443.734375,
      "epoch": 1.288,
      "grad_norm": 0.4313472556924139,
      "kl": 0.09619140625,
      "learning_rate": 3.050466791868254e-06,
      "loss": 0.0038,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1610
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 423.234375,
      "epoch": 1.2888,
      "grad_norm": 0.6668857371532481,
      "kl": 0.08642578125,
      "learning_rate": 3.048423605016815e-06,
      "loss": 0.0034,
      "reward": 1.84375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1611
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 451.671875,
      "epoch": 1.2896,
      "grad_norm": 1.3406958819436345,
      "kl": 0.08935546875,
      "learning_rate": 3.0463800332605787e-06,
      "loss": 0.0036,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1612
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 393.078125,
      "epoch": 1.2904,
      "grad_norm": 0.08148749383324354,
      "kl": 0.09814453125,
      "learning_rate": 3.0443360780338034e-06,
      "loss": 0.0039,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1613
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 436.015625,
      "epoch": 1.2912,
      "grad_norm": 0.4167582259726254,
      "kl": 0.0966796875,
      "learning_rate": 3.042291740771014e-06,
      "loss": 0.0039,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1614
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 440.234375,
      "epoch": 1.292,
      "grad_norm": 0.7375712223049959,
      "kl": 0.0927734375,
      "learning_rate": 3.0402470229070057e-06,
      "loss": 0.0037,
      "reward": 1.796875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1615
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 417.421875,
      "epoch": 1.2928,
      "grad_norm": 0.40930958366571135,
      "kl": 0.1015625,
      "learning_rate": 3.03820192587684e-06,
      "loss": 0.0041,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1616
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 451.46875,
      "epoch": 1.2936,
      "grad_norm": 0.06704584310068394,
      "kl": 0.09716796875,
      "learning_rate": 3.036156451115846e-06,
      "loss": 0.0039,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1617
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 461.078125,
      "epoch": 1.2944,
      "grad_norm": 0.8551137716897909,
      "kl": 0.09521484375,
      "learning_rate": 3.034110600059616e-06,
      "loss": 0.0038,
      "reward": 1.796875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1618
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 360.828125,
      "epoch": 1.2952,
      "grad_norm": 0.4107784594557192,
      "kl": 0.1044921875,
      "learning_rate": 3.0320643741440052e-06,
      "loss": 0.0042,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1619
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 424.703125,
      "epoch": 1.296,
      "grad_norm": 0.7297262530078372,
      "kl": 0.09765625,
      "learning_rate": 3.0300177748051375e-06,
      "loss": 0.0039,
      "reward": 1.453125,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 1620
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 448.6875,
      "epoch": 1.2968,
      "grad_norm": 0.5609991290061184,
      "kl": 0.09228515625,
      "learning_rate": 3.0279708034793907e-06,
      "loss": 0.0037,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1621
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 433.09375,
      "epoch": 1.2976,
      "grad_norm": 2.127767079084424,
      "kl": 0.099609375,
      "learning_rate": 3.025923461603412e-06,
      "loss": 0.004,
      "reward": 1.765625,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1622
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 415.671875,
      "epoch": 1.2984,
      "grad_norm": 0.421056995953176,
      "kl": 0.0849609375,
      "learning_rate": 3.0238757506141013e-06,
      "loss": 0.0034,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1623
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 405.9375,
      "epoch": 1.2992,
      "grad_norm": 0.4784112731263907,
      "kl": 0.0859375,
      "learning_rate": 3.0218276719486245e-06,
      "loss": 0.0034,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1624
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 397.9375,
      "epoch": 1.3,
      "grad_norm": 0.0846574884923956,
      "kl": 0.09912109375,
      "learning_rate": 3.019779227044398e-06,
      "loss": 0.004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1625
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 466.484375,
      "epoch": 1.3008,
      "grad_norm": 0.41397898699895536,
      "kl": 0.09423828125,
      "learning_rate": 3.0177304173391038e-06,
      "loss": 0.0038,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1626
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 456.078125,
      "epoch": 1.3016,
      "grad_norm": 0.091162832448217,
      "kl": 0.09228515625,
      "learning_rate": 3.015681244270672e-06,
      "loss": 0.0037,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1627
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 457.5625,
      "epoch": 1.3024,
      "grad_norm": 0.6367701683829089,
      "kl": 0.0908203125,
      "learning_rate": 3.0136317092772923e-06,
      "loss": 0.0036,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1628
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 511.109375,
      "epoch": 1.3032,
      "grad_norm": 0.3142596953350838,
      "kl": 0.0927734375,
      "learning_rate": 3.0115818137974066e-06,
      "loss": 0.0037,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1629
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 473.4375,
      "epoch": 1.304,
      "grad_norm": 0.2755177205130857,
      "kl": 0.08984375,
      "learning_rate": 3.0095315592697126e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1630
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 414.0,
      "epoch": 1.3048,
      "grad_norm": 0.621827152213513,
      "kl": 0.10302734375,
      "learning_rate": 3.007480947133155e-06,
      "loss": 0.0041,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1631
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 504.125,
      "epoch": 1.3056,
      "grad_norm": 4.9949937715132915,
      "kl": 0.0869140625,
      "learning_rate": 3.0054299788269343e-06,
      "loss": 0.0035,
      "reward": 1.65625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 1632
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 457.875,
      "epoch": 1.3064,
      "grad_norm": 0.4221012204468084,
      "kl": 0.0859375,
      "learning_rate": 3.0033786557904982e-06,
      "loss": 0.0034,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1633
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 459.609375,
      "epoch": 1.3072,
      "grad_norm": 0.3895949270319322,
      "kl": 0.09228515625,
      "learning_rate": 3.001326979463545e-06,
      "loss": 0.0037,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1634
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 455.984375,
      "epoch": 1.308,
      "grad_norm": 0.6182718078206811,
      "kl": 0.0908203125,
      "learning_rate": 2.9992749512860177e-06,
      "loss": 0.0036,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1635
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 459.984375,
      "epoch": 1.3088,
      "grad_norm": 0.7769829844057634,
      "kl": 0.0888671875,
      "learning_rate": 2.9972225726981114e-06,
      "loss": 0.0036,
      "reward": 1.640625,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1636
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 458.59375,
      "epoch": 1.3096,
      "grad_norm": 1.2178216736288152,
      "kl": 0.1044921875,
      "learning_rate": 2.995169845140264e-06,
      "loss": 0.0042,
      "reward": 1.5625,
      "reward_std": 0.213067427277565,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1637
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 435.921875,
      "epoch": 1.3104,
      "grad_norm": 0.7459185976110713,
      "kl": 0.099609375,
      "learning_rate": 2.9931167700531575e-06,
      "loss": 0.004,
      "reward": 1.75,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1638
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 431.890625,
      "epoch": 1.3112,
      "grad_norm": 0.5701855150813324,
      "kl": 0.1005859375,
      "learning_rate": 2.9910633488777198e-06,
      "loss": 0.004,
      "reward": 1.625,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1639
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 461.578125,
      "epoch": 1.312,
      "grad_norm": 0.5978223972000913,
      "kl": 0.0869140625,
      "learning_rate": 2.989009583055121e-06,
      "loss": 0.0035,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1640
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 424.921875,
      "epoch": 1.3128,
      "grad_norm": 0.6614721976106119,
      "kl": 0.103515625,
      "learning_rate": 2.9869554740267726e-06,
      "loss": 0.0041,
      "reward": 1.640625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1641
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 478.390625,
      "epoch": 1.3136,
      "grad_norm": 0.5065939763157713,
      "kl": 0.09375,
      "learning_rate": 2.9849010232343274e-06,
      "loss": 0.0037,
      "reward": 1.90625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1642
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 455.921875,
      "epoch": 1.3144,
      "grad_norm": 0.6517591893012402,
      "kl": 0.0927734375,
      "learning_rate": 2.982846232119679e-06,
      "loss": 0.0037,
      "reward": 1.90625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1643
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 477.328125,
      "epoch": 1.3152,
      "grad_norm": 0.7316421419175054,
      "kl": 0.08984375,
      "learning_rate": 2.9807911021249573e-06,
      "loss": 0.0036,
      "reward": 1.765625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1644
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 454.625,
      "epoch": 1.316,
      "grad_norm": 0.3839403671631433,
      "kl": 0.0947265625,
      "learning_rate": 2.9787356346925327e-06,
      "loss": 0.0038,
      "reward": 1.515625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 1645
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 418.859375,
      "epoch": 1.3168,
      "grad_norm": 0.7656226808781703,
      "kl": 0.0947265625,
      "learning_rate": 2.9766798312650112e-06,
      "loss": 0.0038,
      "reward": 1.515625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 1646
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 409.84375,
      "epoch": 1.3176,
      "grad_norm": 0.4979322214211116,
      "kl": 0.09912109375,
      "learning_rate": 2.9746236932852355e-06,
      "loss": 0.004,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1647
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 376.890625,
      "epoch": 1.3184,
      "grad_norm": 1.6603169356215501,
      "kl": 0.1015625,
      "learning_rate": 2.9725672221962804e-06,
      "loss": 0.0041,
      "reward": 1.71875,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1648
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 433.265625,
      "epoch": 1.3192,
      "grad_norm": 0.6779815964348609,
      "kl": 0.0947265625,
      "learning_rate": 2.9705104194414587e-06,
      "loss": 0.0038,
      "reward": 1.75,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1649
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 447.3125,
      "epoch": 1.32,
      "grad_norm": 0.09704947200588641,
      "kl": 0.099609375,
      "learning_rate": 2.9684532864643123e-06,
      "loss": 0.004,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1650
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 386.890625,
      "epoch": 1.3208,
      "grad_norm": 0.91615554598393,
      "kl": 0.09912109375,
      "learning_rate": 2.9663958247086165e-06,
      "loss": 0.004,
      "reward": 1.84375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1651
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 473.109375,
      "epoch": 1.3216,
      "grad_norm": 0.43300156125730144,
      "kl": 0.09375,
      "learning_rate": 2.964338035618378e-06,
      "loss": 0.0038,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1652
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 476.09375,
      "epoch": 1.3224,
      "grad_norm": 0.6125742503413522,
      "kl": 0.0888671875,
      "learning_rate": 2.9622799206378306e-06,
      "loss": 0.0036,
      "reward": 1.515625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 1653
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 422.140625,
      "epoch": 1.3232,
      "grad_norm": 0.6155846543805755,
      "kl": 0.0888671875,
      "learning_rate": 2.9602214812114414e-06,
      "loss": 0.0036,
      "reward": 1.890625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1654
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 436.09375,
      "epoch": 1.324,
      "grad_norm": 0.9448571772973201,
      "kl": 0.08984375,
      "learning_rate": 2.9581627187838997e-06,
      "loss": 0.0036,
      "reward": 1.6875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1655
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 441.0625,
      "epoch": 1.3248,
      "grad_norm": 0.8290792265291552,
      "kl": 0.09765625,
      "learning_rate": 2.956103634800126e-06,
      "loss": 0.0039,
      "reward": 1.8125,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1656
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 487.984375,
      "epoch": 1.3256000000000001,
      "grad_norm": 0.7710961065985185,
      "kl": 0.08935546875,
      "learning_rate": 2.9540442307052643e-06,
      "loss": 0.0036,
      "reward": 1.828125,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1657
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 468.515625,
      "epoch": 1.3264,
      "grad_norm": 0.4367040453358929,
      "kl": 0.09228515625,
      "learning_rate": 2.9519845079446824e-06,
      "loss": 0.0037,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1658
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 441.0,
      "epoch": 1.3272,
      "grad_norm": 1.7849653041160358,
      "kl": 0.09716796875,
      "learning_rate": 2.949924467963975e-06,
      "loss": 0.0039,
      "reward": 1.53125,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1659
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 456.859375,
      "epoch": 1.328,
      "grad_norm": 0.06616971197048722,
      "kl": 0.087890625,
      "learning_rate": 2.9478641122089563e-06,
      "loss": 0.0035,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1660
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 479.890625,
      "epoch": 1.3288,
      "grad_norm": 0.8634993072852989,
      "kl": 0.0888671875,
      "learning_rate": 2.945803442125663e-06,
      "loss": 0.0036,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1661
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 477.28125,
      "epoch": 1.3296000000000001,
      "grad_norm": 0.5858383428314752,
      "kl": 0.08740234375,
      "learning_rate": 2.943742459160354e-06,
      "loss": 0.0035,
      "reward": 1.59375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1662
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 440.59375,
      "epoch": 1.3304,
      "grad_norm": 0.6528977422099489,
      "kl": 0.0927734375,
      "learning_rate": 2.9416811647595052e-06,
      "loss": 0.0037,
      "reward": 1.828125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1663
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 435.96875,
      "epoch": 1.3312,
      "grad_norm": 0.07583206488855732,
      "kl": 0.10400390625,
      "learning_rate": 2.939619560369813e-06,
      "loss": 0.0042,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1664
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 453.296875,
      "epoch": 1.332,
      "grad_norm": 0.3460165537316144,
      "kl": 0.08447265625,
      "learning_rate": 2.9375576474381907e-06,
      "loss": 0.0034,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1665
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 422.140625,
      "epoch": 1.3328,
      "grad_norm": 0.38978421791431306,
      "kl": 0.10107421875,
      "learning_rate": 2.9354954274117683e-06,
      "loss": 0.004,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1666
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 466.96875,
      "epoch": 1.3336000000000001,
      "grad_norm": 1.4009124949937999,
      "kl": 0.1015625,
      "learning_rate": 2.9334329017378898e-06,
      "loss": 0.0041,
      "reward": 1.84375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1667
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 424.328125,
      "epoch": 1.3344,
      "grad_norm": 0.6815678642497196,
      "kl": 0.1005859375,
      "learning_rate": 2.9313700718641167e-06,
      "loss": 0.004,
      "reward": 1.78125,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1668
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 435.234375,
      "epoch": 1.3352,
      "grad_norm": 0.7279258818436678,
      "kl": 0.09716796875,
      "learning_rate": 2.9293069392382224e-06,
      "loss": 0.0039,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1669
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 436.203125,
      "epoch": 1.336,
      "grad_norm": 0.5515830996228361,
      "kl": 0.0947265625,
      "learning_rate": 2.927243505308192e-06,
      "loss": 0.0038,
      "reward": 1.53125,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1670
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 412.09375,
      "epoch": 1.3368,
      "grad_norm": 1.3628656070496277,
      "kl": 0.11474609375,
      "learning_rate": 2.925179771522223e-06,
      "loss": 0.0046,
      "reward": 1.75,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1671
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 431.4375,
      "epoch": 1.3376000000000001,
      "grad_norm": 0.09915645836765682,
      "kl": 0.0859375,
      "learning_rate": 2.9231157393287234e-06,
      "loss": 0.0034,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1672
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 427.8125,
      "epoch": 1.3384,
      "grad_norm": 0.06367935957263718,
      "kl": 0.10009765625,
      "learning_rate": 2.9210514101763116e-06,
      "loss": 0.004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1673
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 482.546875,
      "epoch": 1.3392,
      "grad_norm": 0.2979102609761941,
      "kl": 0.083984375,
      "learning_rate": 2.9189867855138103e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1674
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 377.171875,
      "epoch": 1.34,
      "grad_norm": 0.7054617593421745,
      "kl": 0.11474609375,
      "learning_rate": 2.9169218667902562e-06,
      "loss": 0.0046,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 1675
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 394.4375,
      "epoch": 1.3408,
      "grad_norm": 0.3841998484727046,
      "kl": 0.09375,
      "learning_rate": 2.9148566554548857e-06,
      "loss": 0.0037,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1676
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 389.53125,
      "epoch": 1.3416000000000001,
      "grad_norm": 0.4768179751975896,
      "kl": 0.08837890625,
      "learning_rate": 2.912791152957145e-06,
      "loss": 0.0035,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1677
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 396.78125,
      "epoch": 1.3424,
      "grad_norm": 0.6416350235556716,
      "kl": 0.1005859375,
      "learning_rate": 2.9107253607466833e-06,
      "loss": 0.004,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1678
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.5,
      "completion_length": 413.71875,
      "epoch": 1.3432,
      "grad_norm": 0.07621470210180957,
      "kl": 0.08740234375,
      "learning_rate": 2.908659280273354e-06,
      "loss": 0.0035,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1679
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 408.625,
      "epoch": 1.3439999999999999,
      "grad_norm": 0.07061972741807182,
      "kl": 0.0927734375,
      "learning_rate": 2.9065929129872097e-06,
      "loss": 0.0037,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1680
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 385.984375,
      "epoch": 1.3448,
      "grad_norm": 0.5972451450752345,
      "kl": 0.09912109375,
      "learning_rate": 2.9045262603385073e-06,
      "loss": 0.004,
      "reward": 1.78125,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1681
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 398.640625,
      "epoch": 1.3456000000000001,
      "grad_norm": 1.232740151493746,
      "kl": 0.09326171875,
      "learning_rate": 2.902459323777704e-06,
      "loss": 0.0037,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1682
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 406.03125,
      "epoch": 1.3464,
      "grad_norm": 0.6605666547913168,
      "kl": 0.091796875,
      "learning_rate": 2.900392104755455e-06,
      "loss": 0.0037,
      "reward": 1.78125,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1683
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 383.5625,
      "epoch": 1.3472,
      "grad_norm": 0.5299227472944182,
      "kl": 0.09716796875,
      "learning_rate": 2.8983246047226137e-06,
      "loss": 0.0039,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1684
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 406.203125,
      "epoch": 1.3479999999999999,
      "grad_norm": 0.06712588071041166,
      "kl": 0.087890625,
      "learning_rate": 2.8962568251302327e-06,
      "loss": 0.0035,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1685
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 440.078125,
      "epoch": 1.3488,
      "grad_norm": 0.6702177333942141,
      "kl": 0.09423828125,
      "learning_rate": 2.8941887674295573e-06,
      "loss": 0.0038,
      "reward": 1.921875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1686
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 459.375,
      "epoch": 1.3496000000000001,
      "grad_norm": 1.1857337277286952,
      "kl": 0.09375,
      "learning_rate": 2.892120433072031e-06,
      "loss": 0.0037,
      "reward": 1.484375,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 1687
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 350.953125,
      "epoch": 1.3504,
      "grad_norm": 1.139359898078901,
      "kl": 0.11279296875,
      "learning_rate": 2.8900518235092908e-06,
      "loss": 0.0045,
      "reward": 1.765625,
      "reward_std": 0.16887325048446655,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1688
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 443.28125,
      "epoch": 1.3512,
      "grad_norm": 0.6587354222238803,
      "kl": 0.08642578125,
      "learning_rate": 2.887982940193165e-06,
      "loss": 0.0035,
      "reward": 1.875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1689
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 376.0,
      "epoch": 1.3519999999999999,
      "grad_norm": 1.0438938529358286,
      "kl": 0.09814453125,
      "learning_rate": 2.8859137845756785e-06,
      "loss": 0.0039,
      "reward": 1.75,
      "reward_std": 0.30038219690322876,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1690
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 384.6875,
      "epoch": 1.3528,
      "grad_norm": 0.8948286229963006,
      "kl": 0.091796875,
      "learning_rate": 2.8838443581090415e-06,
      "loss": 0.0037,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1691
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 410.25,
      "epoch": 1.3536000000000001,
      "grad_norm": 0.567807663221168,
      "kl": 0.10009765625,
      "learning_rate": 2.8817746622456585e-06,
      "loss": 0.004,
      "reward": 1.453125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 1692
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 372.875,
      "epoch": 1.3544,
      "grad_norm": 0.519819832802404,
      "kl": 0.1044921875,
      "learning_rate": 2.879704698438121e-06,
      "loss": 0.0042,
      "reward": 1.484375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 1693
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 391.53125,
      "epoch": 1.3552,
      "grad_norm": 0.06979011480417248,
      "kl": 0.08349609375,
      "learning_rate": 2.8776344681392106e-06,
      "loss": 0.0033,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1694
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 417.828125,
      "epoch": 1.3559999999999999,
      "grad_norm": 0.8902488995754579,
      "kl": 0.09765625,
      "learning_rate": 2.875563972801893e-06,
      "loss": 0.0039,
      "reward": 1.578125,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 1695
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 517.671875,
      "epoch": 1.3568,
      "grad_norm": 0.43442008238301505,
      "kl": 0.08349609375,
      "learning_rate": 2.8734932138793226e-06,
      "loss": 0.0033,
      "reward": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1696
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 383.5625,
      "epoch": 1.3576,
      "grad_norm": 0.7640673678868046,
      "kl": 0.09814453125,
      "learning_rate": 2.871422192824837e-06,
      "loss": 0.0039,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1697
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 419.125,
      "epoch": 1.3584,
      "grad_norm": 0.08779466583102942,
      "kl": 0.09619140625,
      "learning_rate": 2.8693509110919597e-06,
      "loss": 0.0038,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1698
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 453.390625,
      "epoch": 1.3592,
      "grad_norm": 0.06334432150922421,
      "kl": 0.0830078125,
      "learning_rate": 2.867279370134395e-06,
      "loss": 0.0033,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1699
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 441.421875,
      "epoch": 1.3599999999999999,
      "grad_norm": 0.4094643398060491,
      "kl": 0.09765625,
      "learning_rate": 2.8652075714060296e-06,
      "loss": 0.0039,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1700
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 410.171875,
      "epoch": 1.3608,
      "grad_norm": 1.1037768105451344,
      "kl": 0.09716796875,
      "learning_rate": 2.863135516360932e-06,
      "loss": 0.0039,
      "reward": 1.53125,
      "reward_std": 0.2709311842918396,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1701
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 436.546875,
      "epoch": 1.3616,
      "grad_norm": 0.05567304665583726,
      "kl": 0.0869140625,
      "learning_rate": 2.8610632064533517e-06,
      "loss": 0.0035,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1702
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 505.3125,
      "epoch": 1.3624,
      "grad_norm": 0.3870815660672654,
      "kl": 0.08251953125,
      "learning_rate": 2.8589906431377133e-06,
      "loss": 0.0033,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1703
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 464.5,
      "epoch": 1.3632,
      "grad_norm": 0.05381483134898848,
      "kl": 0.0830078125,
      "learning_rate": 2.8569178278686222e-06,
      "loss": 0.0033,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1704
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 440.765625,
      "epoch": 1.3639999999999999,
      "grad_norm": 0.8768340466062517,
      "kl": 0.091796875,
      "learning_rate": 2.8548447621008614e-06,
      "loss": 0.0037,
      "reward": 1.90625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1705
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 440.109375,
      "epoch": 1.3648,
      "grad_norm": 0.616807366638633,
      "kl": 0.09716796875,
      "learning_rate": 2.8527714472893866e-06,
      "loss": 0.0039,
      "reward": 1.859375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1706
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 438.875,
      "epoch": 1.3656,
      "grad_norm": 0.6597623797833504,
      "kl": 0.0869140625,
      "learning_rate": 2.85069788488933e-06,
      "loss": 0.0035,
      "reward": 1.640625,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1707
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 444.953125,
      "epoch": 1.3664,
      "grad_norm": 0.07323190489046655,
      "kl": 0.08642578125,
      "learning_rate": 2.8486240763559984e-06,
      "loss": 0.0035,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1708
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 440.109375,
      "epoch": 1.3672,
      "grad_norm": 0.42004256302505294,
      "kl": 0.0830078125,
      "learning_rate": 2.8465500231448707e-06,
      "loss": 0.0033,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1709
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 433.03125,
      "epoch": 1.3679999999999999,
      "grad_norm": 1.6511001500878217,
      "kl": 0.0927734375,
      "learning_rate": 2.844475726711595e-06,
      "loss": 0.0037,
      "reward": 1.640625,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1710
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 381.53125,
      "epoch": 1.3688,
      "grad_norm": 0.6525848959359796,
      "kl": 0.09716796875,
      "learning_rate": 2.8424011885119956e-06,
      "loss": 0.0039,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1711
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 424.34375,
      "epoch": 1.3696,
      "grad_norm": 0.55412135648888,
      "kl": 0.08935546875,
      "learning_rate": 2.8403264100020613e-06,
      "loss": 0.0036,
      "reward": 1.5625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1712
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 428.84375,
      "epoch": 1.3704,
      "grad_norm": 0.6574047284766994,
      "kl": 0.09228515625,
      "learning_rate": 2.8382513926379508e-06,
      "loss": 0.0037,
      "reward": 1.546875,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 1713
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 387.578125,
      "epoch": 1.3712,
      "grad_norm": 0.4794407935749394,
      "kl": 0.09033203125,
      "learning_rate": 2.836176137875993e-06,
      "loss": 0.0036,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1714
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 424.703125,
      "epoch": 1.3719999999999999,
      "grad_norm": 0.6043851197564949,
      "kl": 0.08984375,
      "learning_rate": 2.8341006471726817e-06,
      "loss": 0.0036,
      "reward": 1.84375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1715
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 423.875,
      "epoch": 1.3728,
      "grad_norm": 0.05399729496075482,
      "kl": 0.087890625,
      "learning_rate": 2.832024921984674e-06,
      "loss": 0.0035,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1716
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 413.046875,
      "epoch": 1.3736,
      "grad_norm": 0.05820011528800716,
      "kl": 0.0830078125,
      "learning_rate": 2.8299489637687955e-06,
      "loss": 0.0033,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1717
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 395.140625,
      "epoch": 1.3744,
      "grad_norm": 1.930911031144314,
      "kl": 0.095703125,
      "learning_rate": 2.8278727739820334e-06,
      "loss": 0.0038,
      "reward": 1.5,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1718
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 436.328125,
      "epoch": 1.3752,
      "grad_norm": 0.17158954044269398,
      "kl": 0.09326171875,
      "learning_rate": 2.825796354081537e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1719
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 401.578125,
      "epoch": 1.376,
      "grad_norm": 0.6193317607808814,
      "kl": 0.09033203125,
      "learning_rate": 2.8237197055246175e-06,
      "loss": 0.0036,
      "reward": 1.765625,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1720
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 429.578125,
      "epoch": 1.3768,
      "grad_norm": 0.7098608229060337,
      "kl": 0.091796875,
      "learning_rate": 2.821642829768748e-06,
      "loss": 0.0037,
      "reward": 1.484375,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 1721
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 397.546875,
      "epoch": 1.3776,
      "grad_norm": 0.6271528558475724,
      "kl": 0.087890625,
      "learning_rate": 2.8195657282715595e-06,
      "loss": 0.0035,
      "reward": 1.875,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1722
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 419.96875,
      "epoch": 1.3784,
      "grad_norm": 0.07813253837379211,
      "kl": 0.09375,
      "learning_rate": 2.817488402490841e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1723
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 402.90625,
      "epoch": 1.3792,
      "grad_norm": 0.7704286028303243,
      "kl": 0.09521484375,
      "learning_rate": 2.8154108538845405e-06,
      "loss": 0.0038,
      "reward": 1.8125,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1724
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 473.140625,
      "epoch": 1.38,
      "grad_norm": 0.49829524199966546,
      "kl": 0.0888671875,
      "learning_rate": 2.813333083910761e-06,
      "loss": 0.0035,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1725
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 416.328125,
      "epoch": 1.3808,
      "grad_norm": 0.6052960995753209,
      "kl": 0.0966796875,
      "learning_rate": 2.8112550940277615e-06,
      "loss": 0.0039,
      "reward": 1.78125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1726
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 375.4375,
      "epoch": 1.3816,
      "grad_norm": 0.0658129561427548,
      "kl": 0.08984375,
      "learning_rate": 2.809176885693956e-06,
      "loss": 0.0036,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1727
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 393.65625,
      "epoch": 1.3824,
      "grad_norm": 1.0729215465529967,
      "kl": 0.10498046875,
      "learning_rate": 2.807098460367911e-06,
      "loss": 0.0042,
      "reward": 1.484375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 1728
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 378.78125,
      "epoch": 1.3832,
      "grad_norm": 0.961026547114402,
      "kl": 0.09228515625,
      "learning_rate": 2.8050198195083445e-06,
      "loss": 0.0037,
      "reward": 1.65625,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1729
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 446.390625,
      "epoch": 1.384,
      "grad_norm": 0.579067243181472,
      "kl": 0.08740234375,
      "learning_rate": 2.802940964574127e-06,
      "loss": 0.0035,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1730
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 441.609375,
      "epoch": 1.3848,
      "grad_norm": 0.35402072367858844,
      "kl": 0.0908203125,
      "learning_rate": 2.800861897024279e-06,
      "loss": 0.0036,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1731
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 455.03125,
      "epoch": 1.3856,
      "grad_norm": 0.8443935379882248,
      "kl": 0.0947265625,
      "learning_rate": 2.798782618317971e-06,
      "loss": 0.0038,
      "reward": 1.71875,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1732
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 429.390625,
      "epoch": 1.3864,
      "grad_norm": 0.1488616126210085,
      "kl": 0.08203125,
      "learning_rate": 2.796703129914519e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1733
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 475.734375,
      "epoch": 1.3872,
      "grad_norm": 0.5764467406833679,
      "kl": 0.087890625,
      "learning_rate": 2.79462343327339e-06,
      "loss": 0.0035,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1734
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 436.53125,
      "epoch": 1.388,
      "grad_norm": 0.9090572974719794,
      "kl": 0.091796875,
      "learning_rate": 2.7925435298541944e-06,
      "loss": 0.0037,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1735
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 408.34375,
      "epoch": 1.3888,
      "grad_norm": 0.7280829399817845,
      "kl": 0.09912109375,
      "learning_rate": 2.7904634211166877e-06,
      "loss": 0.004,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1736
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 434.859375,
      "epoch": 1.3896,
      "grad_norm": 0.43457076699696445,
      "kl": 0.0908203125,
      "learning_rate": 2.7883831085207707e-06,
      "loss": 0.0036,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1737
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 497.5625,
      "epoch": 1.3904,
      "grad_norm": 0.8210253431600771,
      "kl": 0.0859375,
      "learning_rate": 2.7863025935264876e-06,
      "loss": 0.0034,
      "reward": 1.796875,
      "reward_std": 0.2472364604473114,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1738
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 434.515625,
      "epoch": 1.3912,
      "grad_norm": 0.6231118757133141,
      "kl": 0.0849609375,
      "learning_rate": 2.784221877594024e-06,
      "loss": 0.0034,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1739
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 428.71875,
      "epoch": 1.392,
      "grad_norm": 0.3842278559460937,
      "kl": 0.08349609375,
      "learning_rate": 2.7821409621837042e-06,
      "loss": 0.0033,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1740
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 444.609375,
      "epoch": 1.3928,
      "grad_norm": 0.42824205799656606,
      "kl": 0.083984375,
      "learning_rate": 2.7800598487559976e-06,
      "loss": 0.0034,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1741
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 391.953125,
      "epoch": 1.3936,
      "grad_norm": 0.7379060895730872,
      "kl": 0.10302734375,
      "learning_rate": 2.777978538771508e-06,
      "loss": 0.0041,
      "reward": 1.40625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 1742
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 483.5,
      "epoch": 1.3944,
      "grad_norm": 0.2762651121228131,
      "kl": 0.0859375,
      "learning_rate": 2.7758970336909795e-06,
      "loss": 0.0034,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 1743
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 461.84375,
      "epoch": 1.3952,
      "grad_norm": 0.060911127843547894,
      "kl": 0.08251953125,
      "learning_rate": 2.7738153349752923e-06,
      "loss": 0.0033,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1744
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 453.09375,
      "epoch": 1.396,
      "grad_norm": 0.8246021984882184,
      "kl": 0.0986328125,
      "learning_rate": 2.7717334440854634e-06,
      "loss": 0.0039,
      "reward": 1.625,
      "reward_std": 0.1825428307056427,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1745
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 403.59375,
      "epoch": 1.3968,
      "grad_norm": 0.5415659251566713,
      "kl": 0.0927734375,
      "learning_rate": 2.7696513624826422e-06,
      "loss": 0.0037,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1746
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 457.546875,
      "epoch": 1.3976,
      "grad_norm": 1.0157142443046738,
      "kl": 0.09326171875,
      "learning_rate": 2.7675690916281158e-06,
      "loss": 0.0037,
      "reward": 1.421875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.984375,
      "step": 1747
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 420.890625,
      "epoch": 1.3984,
      "grad_norm": 0.7862791590866622,
      "kl": 0.0908203125,
      "learning_rate": 2.7654866329833e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1748
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 412.703125,
      "epoch": 1.3992,
      "grad_norm": 0.07807654781300286,
      "kl": 0.09765625,
      "learning_rate": 2.763403988009746e-06,
      "loss": 0.0039,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1749
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 456.25,
      "epoch": 1.4,
      "grad_norm": 0.054222595804571644,
      "kl": 0.08447265625,
      "learning_rate": 2.761321158169134e-06,
      "loss": 0.0034,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1750
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 497.96875,
      "epoch": 1.4008,
      "grad_norm": 1.0155958812389072,
      "kl": 0.08447265625,
      "learning_rate": 2.759238144923274e-06,
      "loss": 0.0034,
      "reward": 1.6875,
      "reward_std": 0.24608495831489563,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 1751
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 480.3125,
      "epoch": 1.4016,
      "grad_norm": 0.6716594212516473,
      "kl": 0.08544921875,
      "learning_rate": 2.7571549497341044e-06,
      "loss": 0.0034,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1752
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 395.953125,
      "epoch": 1.4024,
      "grad_norm": 0.7699765748252924,
      "kl": 0.09423828125,
      "learning_rate": 2.755071574063692e-06,
      "loss": 0.0038,
      "reward": 1.640625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1753
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 457.703125,
      "epoch": 1.4032,
      "grad_norm": 0.05632543107067612,
      "kl": 0.087890625,
      "learning_rate": 2.7529880193742297e-06,
      "loss": 0.0035,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1754
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 436.046875,
      "epoch": 1.404,
      "grad_norm": 0.09855268858275923,
      "kl": 0.0830078125,
      "learning_rate": 2.7509042871280373e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1755
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 439.703125,
      "epoch": 1.4048,
      "grad_norm": 0.6798272640898966,
      "kl": 0.0947265625,
      "learning_rate": 2.748820378787558e-06,
      "loss": 0.0038,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1756
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 453.609375,
      "epoch": 1.4056,
      "grad_norm": 0.48265125822759436,
      "kl": 0.07763671875,
      "learning_rate": 2.7467362958153585e-06,
      "loss": 0.0031,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1757
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 441.78125,
      "epoch": 1.4064,
      "grad_norm": 0.6355267246841728,
      "kl": 0.0986328125,
      "learning_rate": 2.7446520396741293e-06,
      "loss": 0.004,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1758
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 446.359375,
      "epoch": 1.4072,
      "grad_norm": 0.0589431499290098,
      "kl": 0.0927734375,
      "learning_rate": 2.742567611826681e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1759
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 455.921875,
      "epoch": 1.408,
      "grad_norm": 0.5294323876504748,
      "kl": 0.08935546875,
      "learning_rate": 2.7404830137359445e-06,
      "loss": 0.0036,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1760
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 474.171875,
      "epoch": 1.4088,
      "grad_norm": 0.5531081080315496,
      "kl": 0.095703125,
      "learning_rate": 2.7383982468649715e-06,
      "loss": 0.0038,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1761
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 444.5625,
      "epoch": 1.4096,
      "grad_norm": 0.43544886548557143,
      "kl": 0.08935546875,
      "learning_rate": 2.7363133126769326e-06,
      "loss": 0.0036,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1762
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 456.203125,
      "epoch": 1.4104,
      "grad_norm": 0.8314586552583844,
      "kl": 0.0927734375,
      "learning_rate": 2.7342282126351145e-06,
      "loss": 0.0037,
      "reward": 1.453125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 1763
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 419.53125,
      "epoch": 1.4112,
      "grad_norm": 0.39421918824872754,
      "kl": 0.1025390625,
      "learning_rate": 2.73214294820292e-06,
      "loss": 0.0041,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1764
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 461.359375,
      "epoch": 1.412,
      "grad_norm": 0.9457511440850183,
      "kl": 0.0810546875,
      "learning_rate": 2.7300575208438684e-06,
      "loss": 0.0032,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1765
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 447.125,
      "epoch": 1.4128,
      "grad_norm": 0.06588191069668192,
      "kl": 0.0888671875,
      "learning_rate": 2.7279719320215924e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1766
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 481.234375,
      "epoch": 1.4136,
      "grad_norm": 0.6801417772319411,
      "kl": 0.09228515625,
      "learning_rate": 2.725886183199839e-06,
      "loss": 0.0037,
      "reward": 1.734375,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1767
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 430.703125,
      "epoch": 1.4144,
      "grad_norm": 2.235008009368302,
      "kl": 0.09228515625,
      "learning_rate": 2.723800275842468e-06,
      "loss": 0.0037,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 1768
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 459.078125,
      "epoch": 1.4152,
      "grad_norm": 0.05632043931444275,
      "kl": 0.08447265625,
      "learning_rate": 2.7217142114134466e-06,
      "loss": 0.0034,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1769
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 380.6875,
      "epoch": 1.416,
      "grad_norm": 1.328281291993653,
      "kl": 0.10205078125,
      "learning_rate": 2.7196279913768587e-06,
      "loss": 0.0041,
      "reward": 1.765625,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1770
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 436.9375,
      "epoch": 1.4168,
      "grad_norm": 0.4187136145829767,
      "kl": 0.0869140625,
      "learning_rate": 2.717541617196891e-06,
      "loss": 0.0035,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1771
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 421.125,
      "epoch": 1.4176,
      "grad_norm": 0.3430143246769699,
      "kl": 0.08544921875,
      "learning_rate": 2.7154550903378425e-06,
      "loss": 0.0034,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1772
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 442.640625,
      "epoch": 1.4184,
      "grad_norm": 0.75589745660042,
      "kl": 0.0947265625,
      "learning_rate": 2.713368412264118e-06,
      "loss": 0.0038,
      "reward": 1.65625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1773
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 481.84375,
      "epoch": 1.4192,
      "grad_norm": 0.5936135246264803,
      "kl": 0.07666015625,
      "learning_rate": 2.711281584440228e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1774
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 401.3125,
      "epoch": 1.42,
      "grad_norm": 0.7427802776635967,
      "kl": 0.09423828125,
      "learning_rate": 2.70919460833079e-06,
      "loss": 0.0038,
      "reward": 1.65625,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1775
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 447.25,
      "epoch": 1.4208,
      "grad_norm": 1.0457149215633064,
      "kl": 0.08642578125,
      "learning_rate": 2.7071074854005206e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1776
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 429.53125,
      "epoch": 1.4216,
      "grad_norm": 0.11534240201715483,
      "kl": 0.08544921875,
      "learning_rate": 2.705020217114248e-06,
      "loss": 0.0034,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1777
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 404.046875,
      "epoch": 1.4224,
      "grad_norm": 2.0687053977515215,
      "kl": 0.087890625,
      "learning_rate": 2.7029328049368942e-06,
      "loss": 0.0035,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1778
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 404.859375,
      "epoch": 1.4232,
      "grad_norm": 0.6773605944996723,
      "kl": 0.0908203125,
      "learning_rate": 2.700845250333486e-06,
      "loss": 0.0036,
      "reward": 1.796875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1779
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 454.21875,
      "epoch": 1.424,
      "grad_norm": 0.37142243554998144,
      "kl": 0.0859375,
      "learning_rate": 2.69875755476915e-06,
      "loss": 0.0034,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1780
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 421.9375,
      "epoch": 1.4248,
      "grad_norm": 0.842197329868124,
      "kl": 0.0849609375,
      "learning_rate": 2.696669719709111e-06,
      "loss": 0.0034,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1781
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 428.28125,
      "epoch": 1.4256,
      "grad_norm": 0.37781768013698297,
      "kl": 0.08056640625,
      "learning_rate": 2.694581746618691e-06,
      "loss": 0.0032,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1782
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 412.703125,
      "epoch": 1.4264000000000001,
      "grad_norm": 0.9742505049336619,
      "kl": 0.08740234375,
      "learning_rate": 2.6924936369633126e-06,
      "loss": 0.0035,
      "reward": 1.796875,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1783
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 409.765625,
      "epoch": 1.4272,
      "grad_norm": 0.36508109585285203,
      "kl": 0.0908203125,
      "learning_rate": 2.6904053922084893e-06,
      "loss": 0.0036,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1784
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 399.53125,
      "epoch": 1.428,
      "grad_norm": 0.4482960129779903,
      "kl": 0.0908203125,
      "learning_rate": 2.688317013819832e-06,
      "loss": 0.0036,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1785
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 436.46875,
      "epoch": 1.4288,
      "grad_norm": 0.8436666351865185,
      "kl": 0.08251953125,
      "learning_rate": 2.686228503263045e-06,
      "loss": 0.0033,
      "reward": 1.78125,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1786
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 412.828125,
      "epoch": 1.4296,
      "grad_norm": 0.6532852990214628,
      "kl": 0.0888671875,
      "learning_rate": 2.684139862003927e-06,
      "loss": 0.0035,
      "reward": 1.578125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 1787
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 412.21875,
      "epoch": 1.4304000000000001,
      "grad_norm": 0.7029649307478942,
      "kl": 0.09765625,
      "learning_rate": 2.682051091508365e-06,
      "loss": 0.0039,
      "reward": 1.734375,
      "reward_std": 0.1893727034330368,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1788
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 477.46875,
      "epoch": 1.4312,
      "grad_norm": 0.5554355074355926,
      "kl": 0.08837890625,
      "learning_rate": 2.679962193242338e-06,
      "loss": 0.0035,
      "reward": 1.734375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1789
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 476.203125,
      "epoch": 1.432,
      "grad_norm": 9.292448459279282,
      "kl": 0.08251953125,
      "learning_rate": 2.6778731686719177e-06,
      "loss": 0.0033,
      "reward": 1.796875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1790
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 407.59375,
      "epoch": 1.4328,
      "grad_norm": 1.1170165060861008,
      "kl": 0.08837890625,
      "learning_rate": 2.67578401926326e-06,
      "loss": 0.0035,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 1791
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 387.5625,
      "epoch": 1.4336,
      "grad_norm": 0.6628566904467228,
      "kl": 0.08740234375,
      "learning_rate": 2.6736947464826107e-06,
      "loss": 0.0035,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1792
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 416.09375,
      "epoch": 1.4344000000000001,
      "grad_norm": 0.5003754997616967,
      "kl": 0.08447265625,
      "learning_rate": 2.671605351796302e-06,
      "loss": 0.0034,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1793
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 397.109375,
      "epoch": 1.4352,
      "grad_norm": 0.6028123645605987,
      "kl": 0.0966796875,
      "learning_rate": 2.6695158366707526e-06,
      "loss": 0.0039,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1794
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 445.53125,
      "epoch": 1.436,
      "grad_norm": 3.057199036053953,
      "kl": 0.1005859375,
      "learning_rate": 2.667426202572463e-06,
      "loss": 0.004,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1795
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 408.40625,
      "epoch": 1.4368,
      "grad_norm": 0.5067201512233437,
      "kl": 0.07763671875,
      "learning_rate": 2.665336450968019e-06,
      "loss": 0.0031,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1796
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 429.234375,
      "epoch": 1.4376,
      "grad_norm": 0.06801069394458993,
      "kl": 0.08544921875,
      "learning_rate": 2.6632465833240895e-06,
      "loss": 0.0034,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1797
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 460.75,
      "epoch": 1.4384000000000001,
      "grad_norm": 0.054330115276332414,
      "kl": 0.08447265625,
      "learning_rate": 2.661156601107424e-06,
      "loss": 0.0034,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1798
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 442.328125,
      "epoch": 1.4392,
      "grad_norm": 0.7862043988917632,
      "kl": 0.08203125,
      "learning_rate": 2.659066505784852e-06,
      "loss": 0.0033,
      "reward": 1.859375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1799
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 388.171875,
      "epoch": 1.44,
      "grad_norm": 0.5702632707864189,
      "kl": 0.09619140625,
      "learning_rate": 2.6569762988232838e-06,
      "loss": 0.0039,
      "reward": 1.640625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1800
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 465.203125,
      "epoch": 1.4408,
      "grad_norm": 0.8445473876693124,
      "kl": 0.0810546875,
      "learning_rate": 2.654885981689706e-06,
      "loss": 0.0032,
      "reward": 1.9375,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 1801
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 423.828125,
      "epoch": 1.4416,
      "grad_norm": 0.7352932222659931,
      "kl": 0.08935546875,
      "learning_rate": 2.652795555851184e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1802
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 418.375,
      "epoch": 1.4424000000000001,
      "grad_norm": 0.6185246891708823,
      "kl": 0.095703125,
      "learning_rate": 2.6507050227748595e-06,
      "loss": 0.0038,
      "reward": 1.90625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.984375,
      "step": 1803
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 415.0625,
      "epoch": 1.4432,
      "grad_norm": 0.5966741557649632,
      "kl": 0.083984375,
      "learning_rate": 2.648614383927949e-06,
      "loss": 0.0034,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1804
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 435.609375,
      "epoch": 1.444,
      "grad_norm": 0.6485992869672412,
      "kl": 0.09912109375,
      "learning_rate": 2.646523640777741e-06,
      "loss": 0.004,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 1805
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 456.40625,
      "epoch": 1.4447999999999999,
      "grad_norm": 0.055847950381667615,
      "kl": 0.08251953125,
      "learning_rate": 2.6444327947916037e-06,
      "loss": 0.0033,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1806
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 413.15625,
      "epoch": 1.4456,
      "grad_norm": 1.909258839064335,
      "kl": 0.08740234375,
      "learning_rate": 2.6423418474369707e-06,
      "loss": 0.0035,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1807
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 477.390625,
      "epoch": 1.4464000000000001,
      "grad_norm": 0.06290973328158704,
      "kl": 0.07421875,
      "learning_rate": 2.64025080018135e-06,
      "loss": 0.003,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1808
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 390.625,
      "epoch": 1.4472,
      "grad_norm": 0.4668977788776293,
      "kl": 0.10107421875,
      "learning_rate": 2.6381596544923184e-06,
      "loss": 0.004,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 1809
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 513.4375,
      "epoch": 1.448,
      "grad_norm": 0.4406609866412533,
      "kl": 0.080078125,
      "learning_rate": 2.636068411837523e-06,
      "loss": 0.0032,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1810
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 494.234375,
      "epoch": 1.4487999999999999,
      "grad_norm": 0.5725715470521852,
      "kl": 0.0859375,
      "learning_rate": 2.6339770736846794e-06,
      "loss": 0.0034,
      "reward": 1.671875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1811
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 448.875,
      "epoch": 1.4496,
      "grad_norm": 0.8695663469311959,
      "kl": 0.08447265625,
      "learning_rate": 2.6318856415015664e-06,
      "loss": 0.0034,
      "reward": 1.671875,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1812
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 403.90625,
      "epoch": 1.4504000000000001,
      "grad_norm": 0.0594596234641751,
      "kl": 0.0986328125,
      "learning_rate": 2.629794116756035e-06,
      "loss": 0.004,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1813
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 493.359375,
      "epoch": 1.4512,
      "grad_norm": 0.8626395875074092,
      "kl": 0.07861328125,
      "learning_rate": 2.627702500915995e-06,
      "loss": 0.0031,
      "reward": 1.71875,
      "reward_std": 0.2709311544895172,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 1814
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 430.0,
      "epoch": 1.452,
      "grad_norm": 1.2437546406975093,
      "kl": 0.091796875,
      "learning_rate": 2.625610795449424e-06,
      "loss": 0.0037,
      "reward": 1.796875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1815
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 442.984375,
      "epoch": 1.4527999999999999,
      "grad_norm": 0.4148455903939452,
      "kl": 0.08447265625,
      "learning_rate": 2.6235190018243623e-06,
      "loss": 0.0034,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1816
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 489.46875,
      "epoch": 1.4536,
      "grad_norm": 0.4913454489457642,
      "kl": 0.09033203125,
      "learning_rate": 2.6214271215089106e-06,
      "loss": 0.0036,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1817
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 451.984375,
      "epoch": 1.4544000000000001,
      "grad_norm": 0.3579503990202543,
      "kl": 0.0908203125,
      "learning_rate": 2.6193351559712294e-06,
      "loss": 0.0036,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1818
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 424.015625,
      "epoch": 1.4552,
      "grad_norm": 0.9040421571837298,
      "kl": 0.080078125,
      "learning_rate": 2.6172431066795428e-06,
      "loss": 0.0032,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1819
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 467.390625,
      "epoch": 1.456,
      "grad_norm": 0.5527848660887122,
      "kl": 0.08935546875,
      "learning_rate": 2.6151509751021307e-06,
      "loss": 0.0036,
      "reward": 1.921875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 1820
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 452.609375,
      "epoch": 1.4567999999999999,
      "grad_norm": 0.7159570949496135,
      "kl": 0.08447265625,
      "learning_rate": 2.6130587627073315e-06,
      "loss": 0.0034,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1821
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 503.59375,
      "epoch": 1.4576,
      "grad_norm": 0.7045527173346348,
      "kl": 0.0869140625,
      "learning_rate": 2.6109664709635413e-06,
      "loss": 0.0035,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1822
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 458.953125,
      "epoch": 1.4584,
      "grad_norm": 1.3746716929196108,
      "kl": 0.07373046875,
      "learning_rate": 2.60887410133921e-06,
      "loss": 0.003,
      "reward": 1.84375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1823
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 401.984375,
      "epoch": 1.4592,
      "grad_norm": 0.7099269175226316,
      "kl": 0.0966796875,
      "learning_rate": 2.606781655302843e-06,
      "loss": 0.0039,
      "reward": 1.734375,
      "reward_std": 0.16887325048446655,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1824
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 446.734375,
      "epoch": 1.46,
      "grad_norm": 0.29515016967963137,
      "kl": 0.09228515625,
      "learning_rate": 2.604689134322999e-06,
      "loss": 0.0037,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1825
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 439.578125,
      "epoch": 1.4607999999999999,
      "grad_norm": 0.4957503499839096,
      "kl": 0.0849609375,
      "learning_rate": 2.602596539868292e-06,
      "loss": 0.0034,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1826
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 471.640625,
      "epoch": 1.4616,
      "grad_norm": 0.05056978228204779,
      "kl": 0.08642578125,
      "learning_rate": 2.6005038734073833e-06,
      "loss": 0.0035,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1827
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 469.625,
      "epoch": 1.4624,
      "grad_norm": 0.05118351605817079,
      "kl": 0.080078125,
      "learning_rate": 2.5984111364089875e-06,
      "loss": 0.0032,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1828
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 439.546875,
      "epoch": 1.4632,
      "grad_norm": 0.5244917655001043,
      "kl": 0.08740234375,
      "learning_rate": 2.5963183303418682e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1829
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.5,
      "completion_length": 378.703125,
      "epoch": 1.464,
      "grad_norm": 0.06660306641892735,
      "kl": 0.09619140625,
      "learning_rate": 2.594225456674837e-06,
      "loss": 0.0038,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1830
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 377.71875,
      "epoch": 1.4647999999999999,
      "grad_norm": 0.4164352473607868,
      "kl": 0.10595703125,
      "learning_rate": 2.592132516876753e-06,
      "loss": 0.0042,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1831
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 447.671875,
      "epoch": 1.4656,
      "grad_norm": 0.5268936714361685,
      "kl": 0.0859375,
      "learning_rate": 2.5900395124165216e-06,
      "loss": 0.0034,
      "reward": 1.796875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1832
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 436.890625,
      "epoch": 1.4664,
      "grad_norm": 0.39118333712594194,
      "kl": 0.0869140625,
      "learning_rate": 2.5879464447630947e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1833
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 460.828125,
      "epoch": 1.4672,
      "grad_norm": 0.055892104785312396,
      "kl": 0.087890625,
      "learning_rate": 2.5858533153854676e-06,
      "loss": 0.0035,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1834
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 468.828125,
      "epoch": 1.468,
      "grad_norm": 0.30366899064146025,
      "kl": 0.0849609375,
      "learning_rate": 2.583760125752679e-06,
      "loss": 0.0034,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1835
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 490.140625,
      "epoch": 1.4687999999999999,
      "grad_norm": 0.3984764927244527,
      "kl": 0.0869140625,
      "learning_rate": 2.58166687733381e-06,
      "loss": 0.0035,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1836
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 434.34375,
      "epoch": 1.4696,
      "grad_norm": 0.09308225263921162,
      "kl": 0.080078125,
      "learning_rate": 2.5795735715979826e-06,
      "loss": 0.0032,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1837
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 398.3125,
      "epoch": 1.4704,
      "grad_norm": 0.5889483500307229,
      "kl": 0.0927734375,
      "learning_rate": 2.577480210014359e-06,
      "loss": 0.0037,
      "reward": 1.796875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1838
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 397.015625,
      "epoch": 1.4712,
      "grad_norm": 0.9711711579355499,
      "kl": 0.103515625,
      "learning_rate": 2.575386794052142e-06,
      "loss": 0.0041,
      "reward": 1.609375,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 1839
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 438.125,
      "epoch": 1.472,
      "grad_norm": 0.0572974281718638,
      "kl": 0.080078125,
      "learning_rate": 2.5732933251805716e-06,
      "loss": 0.0032,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1840
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 424.859375,
      "epoch": 1.4727999999999999,
      "grad_norm": 0.48171769956286503,
      "kl": 0.083984375,
      "learning_rate": 2.571199804868923e-06,
      "loss": 0.0034,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1841
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 408.515625,
      "epoch": 1.4736,
      "grad_norm": 0.39939450673132837,
      "kl": 0.08349609375,
      "learning_rate": 2.569106234586511e-06,
      "loss": 0.0033,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.984375,
      "step": 1842
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 423.640625,
      "epoch": 1.4744,
      "grad_norm": 0.3497836011832466,
      "kl": 0.0947265625,
      "learning_rate": 2.5670126158026843e-06,
      "loss": 0.0038,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1843
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 471.9375,
      "epoch": 1.4752,
      "grad_norm": 0.05418370592255706,
      "kl": 0.08837890625,
      "learning_rate": 2.5649189499868233e-06,
      "loss": 0.0035,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1844
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 436.859375,
      "epoch": 1.476,
      "grad_norm": 0.053063064671305116,
      "kl": 0.0888671875,
      "learning_rate": 2.5628252386083443e-06,
      "loss": 0.0036,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1845
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 462.796875,
      "epoch": 1.4768,
      "grad_norm": 0.30970356384425807,
      "kl": 0.08447265625,
      "learning_rate": 2.560731483136694e-06,
      "loss": 0.0034,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1846
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 413.28125,
      "epoch": 1.4776,
      "grad_norm": 0.9125615897854776,
      "kl": 0.0830078125,
      "learning_rate": 2.558637685041352e-06,
      "loss": 0.0033,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1847
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 454.984375,
      "epoch": 1.4784,
      "grad_norm": 0.059595331973601735,
      "kl": 0.076171875,
      "learning_rate": 2.5565438457918247e-06,
      "loss": 0.003,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1848
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 391.359375,
      "epoch": 1.4792,
      "grad_norm": 0.06795832349931433,
      "kl": 0.0888671875,
      "learning_rate": 2.5544499668576508e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1849
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 420.890625,
      "epoch": 1.48,
      "grad_norm": 0.5189598870665199,
      "kl": 0.0810546875,
      "learning_rate": 2.5523560497083927e-06,
      "loss": 0.0032,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1850
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 430.5,
      "epoch": 1.4808,
      "grad_norm": 0.4665591820452326,
      "kl": 0.06689453125,
      "learning_rate": 2.5502620958136444e-06,
      "loss": 0.0027,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1851
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 465.4375,
      "epoch": 1.4816,
      "grad_norm": 0.6207213717246918,
      "kl": 0.0810546875,
      "learning_rate": 2.548168106643022e-06,
      "loss": 0.0032,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1852
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 452.5625,
      "epoch": 1.4824,
      "grad_norm": 0.5084056440420548,
      "kl": 0.0859375,
      "learning_rate": 2.546074083666169e-06,
      "loss": 0.0034,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1853
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 401.078125,
      "epoch": 1.4832,
      "grad_norm": 0.060662205606532446,
      "kl": 0.08837890625,
      "learning_rate": 2.5439800283527495e-06,
      "loss": 0.0035,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1854
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 463.203125,
      "epoch": 1.484,
      "grad_norm": 0.07666590310346533,
      "kl": 0.07373046875,
      "learning_rate": 2.541885942172454e-06,
      "loss": 0.0029,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1855
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 400.890625,
      "epoch": 1.4848,
      "grad_norm": 0.39402392081461957,
      "kl": 0.099609375,
      "learning_rate": 2.539791826594991e-06,
      "loss": 0.004,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1856
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 455.21875,
      "epoch": 1.4856,
      "grad_norm": 0.3416794358100851,
      "kl": 0.09375,
      "learning_rate": 2.537697683090093e-06,
      "loss": 0.0037,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1857
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 392.34375,
      "epoch": 1.4864,
      "grad_norm": 2.1323883472952603,
      "kl": 0.0927734375,
      "learning_rate": 2.5356035131275096e-06,
      "loss": 0.0037,
      "reward": 1.8125,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1858
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 447.953125,
      "epoch": 1.4872,
      "grad_norm": 0.505078824930963,
      "kl": 0.0869140625,
      "learning_rate": 2.5335093181770105e-06,
      "loss": 0.0035,
      "reward": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1859
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 473.828125,
      "epoch": 1.488,
      "grad_norm": 0.44477228397087665,
      "kl": 0.08203125,
      "learning_rate": 2.531415099708382e-06,
      "loss": 0.0033,
      "reward": 1.84375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1860
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 471.59375,
      "epoch": 1.4888,
      "grad_norm": 0.31768271281471744,
      "kl": 0.0869140625,
      "learning_rate": 2.5293208591914265e-06,
      "loss": 0.0035,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1861
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 396.0625,
      "epoch": 1.4896,
      "grad_norm": 0.37940950625472025,
      "kl": 0.08154296875,
      "learning_rate": 2.5272265980959644e-06,
      "loss": 0.0033,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1862
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 396.71875,
      "epoch": 1.4904,
      "grad_norm": 0.6872315143129755,
      "kl": 0.10009765625,
      "learning_rate": 2.525132317891827e-06,
      "loss": 0.004,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1863
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 451.375,
      "epoch": 1.4912,
      "grad_norm": 0.05494493440457726,
      "kl": 0.08154296875,
      "learning_rate": 2.523038020048861e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1864
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 405.609375,
      "epoch": 1.492,
      "grad_norm": 2.3759659888477676,
      "kl": 0.08740234375,
      "learning_rate": 2.5209437060369266e-06,
      "loss": 0.0035,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1865
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 407.1875,
      "epoch": 1.4928,
      "grad_norm": 0.5496704832725965,
      "kl": 0.09130859375,
      "learning_rate": 2.518849377325893e-06,
      "loss": 0.0037,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1866
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 460.796875,
      "epoch": 1.4936,
      "grad_norm": 0.04423375224248584,
      "kl": 0.07373046875,
      "learning_rate": 2.51675503538564e-06,
      "loss": 0.0029,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1867
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 422.875,
      "epoch": 1.4944,
      "grad_norm": 0.4092693934966897,
      "kl": 0.09326171875,
      "learning_rate": 2.5146606816860597e-06,
      "loss": 0.0037,
      "reward": 1.53125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1868
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 454.484375,
      "epoch": 1.4952,
      "grad_norm": 0.6088729253934295,
      "kl": 0.08251953125,
      "learning_rate": 2.5125663176970475e-06,
      "loss": 0.0033,
      "reward": 1.6875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1869
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 413.09375,
      "epoch": 1.496,
      "grad_norm": 0.05835888209531134,
      "kl": 0.0869140625,
      "learning_rate": 2.5104719448885103e-06,
      "loss": 0.0035,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1870
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 410.03125,
      "epoch": 1.4968,
      "grad_norm": 0.6104164350909421,
      "kl": 0.09033203125,
      "learning_rate": 2.5083775647303583e-06,
      "loss": 0.0036,
      "reward": 1.640625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1871
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 430.265625,
      "epoch": 1.4976,
      "grad_norm": 0.7829505465205233,
      "kl": 0.08740234375,
      "learning_rate": 2.5062831786925102e-06,
      "loss": 0.0035,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 1872
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 484.140625,
      "epoch": 1.4984,
      "grad_norm": 0.7670852263386284,
      "kl": 0.0771484375,
      "learning_rate": 2.5041887882448845e-06,
      "loss": 0.0031,
      "reward": 1.71875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 1873
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 407.1875,
      "epoch": 1.4992,
      "grad_norm": 0.7791601483538768,
      "kl": 0.1005859375,
      "learning_rate": 2.5020943948574056e-06,
      "loss": 0.004,
      "reward": 1.90625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1874
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 414.4375,
      "epoch": 1.5,
      "grad_norm": 0.07225995331618189,
      "kl": 0.08544921875,
      "learning_rate": 2.5e-06,
      "loss": 0.0034,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1875
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 461.9375,
      "epoch": 1.5008,
      "grad_norm": 1.03905440902492,
      "kl": 0.08935546875,
      "learning_rate": 2.497905605142595e-06,
      "loss": 0.0036,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1876
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 432.75,
      "epoch": 1.5016,
      "grad_norm": 0.7955818423033687,
      "kl": 0.09228515625,
      "learning_rate": 2.4958112117551163e-06,
      "loss": 0.0037,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1877
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 403.46875,
      "epoch": 1.5024,
      "grad_norm": 0.4000701195771805,
      "kl": 0.08154296875,
      "learning_rate": 2.4937168213074906e-06,
      "loss": 0.0033,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.984375,
      "step": 1878
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 434.890625,
      "epoch": 1.5032,
      "grad_norm": 0.05769996956502378,
      "kl": 0.07861328125,
      "learning_rate": 2.491622435269642e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1879
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 419.9375,
      "epoch": 1.504,
      "grad_norm": 0.658628747605266,
      "kl": 0.0859375,
      "learning_rate": 2.489528055111491e-06,
      "loss": 0.0034,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1880
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 427.671875,
      "epoch": 1.5048,
      "grad_norm": 0.6500906281135059,
      "kl": 0.0888671875,
      "learning_rate": 2.487433682302953e-06,
      "loss": 0.0036,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1881
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 458.0625,
      "epoch": 1.5056,
      "grad_norm": 0.047598480794597976,
      "kl": 0.068359375,
      "learning_rate": 2.485339318313941e-06,
      "loss": 0.0027,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1882
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 460.5625,
      "epoch": 1.5064,
      "grad_norm": 0.0754762844906546,
      "kl": 0.08349609375,
      "learning_rate": 2.4832449646143605e-06,
      "loss": 0.0033,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1883
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 469.203125,
      "epoch": 1.5072,
      "grad_norm": 0.4099922193713641,
      "kl": 0.08203125,
      "learning_rate": 2.4811506226741077e-06,
      "loss": 0.0033,
      "reward": 1.9375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 1884
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 394.65625,
      "epoch": 1.508,
      "grad_norm": 0.8625269812715508,
      "kl": 0.0927734375,
      "learning_rate": 2.4790562939630738e-06,
      "loss": 0.0037,
      "reward": 1.90625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1885
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 472.6875,
      "epoch": 1.5088,
      "grad_norm": 0.05813491582761802,
      "kl": 0.0732421875,
      "learning_rate": 2.4769619799511392e-06,
      "loss": 0.0029,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1886
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 474.53125,
      "epoch": 1.5096,
      "grad_norm": 0.6129709930819037,
      "kl": 0.09521484375,
      "learning_rate": 2.474867682108174e-06,
      "loss": 0.0038,
      "reward": 1.796875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1887
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 426.90625,
      "epoch": 1.5104,
      "grad_norm": 0.7629134340766548,
      "kl": 0.087890625,
      "learning_rate": 2.472773401904037e-06,
      "loss": 0.0035,
      "reward": 1.65625,
      "reward_std": 0.1735912710428238,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1888
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 463.953125,
      "epoch": 1.5112,
      "grad_norm": 0.4520750873435136,
      "kl": 0.07861328125,
      "learning_rate": 2.470679140808574e-06,
      "loss": 0.0031,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1889
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 458.625,
      "epoch": 1.512,
      "grad_norm": 0.05248061071666733,
      "kl": 0.08056640625,
      "learning_rate": 2.4685849002916184e-06,
      "loss": 0.0032,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1890
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 439.375,
      "epoch": 1.5128,
      "grad_norm": 0.058220772450739776,
      "kl": 0.07763671875,
      "learning_rate": 2.4664906818229903e-06,
      "loss": 0.0031,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1891
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 466.09375,
      "epoch": 1.5135999999999998,
      "grad_norm": 0.05223533553545021,
      "kl": 0.08642578125,
      "learning_rate": 2.4643964868724916e-06,
      "loss": 0.0035,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1892
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 452.53125,
      "epoch": 1.5144,
      "grad_norm": 1.512917485006429,
      "kl": 0.08203125,
      "learning_rate": 2.4623023169099074e-06,
      "loss": 0.0033,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 1893
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 465.609375,
      "epoch": 1.5152,
      "grad_norm": 0.5777589521384713,
      "kl": 0.07275390625,
      "learning_rate": 2.4602081734050093e-06,
      "loss": 0.0029,
      "reward": 1.859375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1894
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 454.9375,
      "epoch": 1.516,
      "grad_norm": 0.04983621844004456,
      "kl": 0.0869140625,
      "learning_rate": 2.4581140578275473e-06,
      "loss": 0.0035,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1895
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 418.609375,
      "epoch": 1.5168,
      "grad_norm": 0.05418391240913127,
      "kl": 0.0859375,
      "learning_rate": 2.456019971647251e-06,
      "loss": 0.0034,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1896
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 419.578125,
      "epoch": 1.5175999999999998,
      "grad_norm": 0.4708469803329581,
      "kl": 0.09912109375,
      "learning_rate": 2.4539259163338317e-06,
      "loss": 0.004,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1897
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 404.96875,
      "epoch": 1.5184,
      "grad_norm": 0.20812983894555992,
      "kl": 0.0908203125,
      "learning_rate": 2.4518318933569786e-06,
      "loss": 0.0036,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1898
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 448.40625,
      "epoch": 1.5192,
      "grad_norm": 0.4885372425816075,
      "kl": 0.08544921875,
      "learning_rate": 2.449737904186357e-06,
      "loss": 0.0034,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1899
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 487.625,
      "epoch": 1.52,
      "grad_norm": 0.6225599802503589,
      "kl": 0.0751953125,
      "learning_rate": 2.447643950291608e-06,
      "loss": 0.003,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1900
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 450.5625,
      "epoch": 1.5208,
      "grad_norm": 0.44505940091203605,
      "kl": 0.08056640625,
      "learning_rate": 2.4455500331423505e-06,
      "loss": 0.0032,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1901
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 426.828125,
      "epoch": 1.5215999999999998,
      "grad_norm": 1.5724566153791564,
      "kl": 0.08642578125,
      "learning_rate": 2.4434561542081765e-06,
      "loss": 0.0035,
      "reward": 1.53125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1902
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 412.15625,
      "epoch": 1.5224,
      "grad_norm": 0.0688764837836802,
      "kl": 0.083984375,
      "learning_rate": 2.441362314958649e-06,
      "loss": 0.0034,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1903
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 406.8125,
      "epoch": 1.5232,
      "grad_norm": 0.4062787098036763,
      "kl": 0.0849609375,
      "learning_rate": 2.439268516863306e-06,
      "loss": 0.0034,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1904
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 405.1875,
      "epoch": 1.524,
      "grad_norm": 0.056109183998569095,
      "kl": 0.07958984375,
      "learning_rate": 2.4371747613916566e-06,
      "loss": 0.0032,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1905
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 398.421875,
      "epoch": 1.5248,
      "grad_norm": 0.5402923435551574,
      "kl": 0.09814453125,
      "learning_rate": 2.4350810500131776e-06,
      "loss": 0.0039,
      "reward": 1.515625,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 1906
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 451.921875,
      "epoch": 1.5255999999999998,
      "grad_norm": 0.06078793551415715,
      "kl": 0.0830078125,
      "learning_rate": 2.4329873841973174e-06,
      "loss": 0.0033,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1907
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 414.578125,
      "epoch": 1.5264,
      "grad_norm": 0.43927563665387126,
      "kl": 0.08349609375,
      "learning_rate": 2.4308937654134893e-06,
      "loss": 0.0033,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1908
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 373.65625,
      "epoch": 1.5272000000000001,
      "grad_norm": 0.977754562420843,
      "kl": 0.099609375,
      "learning_rate": 2.428800195131078e-06,
      "loss": 0.004,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1909
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 446.90625,
      "epoch": 1.528,
      "grad_norm": 0.43253855303777056,
      "kl": 0.08642578125,
      "learning_rate": 2.4267066748194297e-06,
      "loss": 0.0035,
      "reward": 1.78125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1910
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 419.453125,
      "epoch": 1.5288,
      "grad_norm": 0.05995847869070541,
      "kl": 0.08203125,
      "learning_rate": 2.4246132059478582e-06,
      "loss": 0.0033,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1911
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 481.9375,
      "epoch": 1.5295999999999998,
      "grad_norm": 0.05057924473336973,
      "kl": 0.087890625,
      "learning_rate": 2.4225197899856416e-06,
      "loss": 0.0035,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1912
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 431.359375,
      "epoch": 1.5304,
      "grad_norm": 0.7218737826161329,
      "kl": 0.08349609375,
      "learning_rate": 2.4204264284020182e-06,
      "loss": 0.0033,
      "reward": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 1913
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 424.796875,
      "epoch": 1.5312000000000001,
      "grad_norm": 1.083573297307257,
      "kl": 0.10546875,
      "learning_rate": 2.4183331226661913e-06,
      "loss": 0.0042,
      "reward": 1.578125,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 1914
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 461.015625,
      "epoch": 1.532,
      "grad_norm": 0.903437762615361,
      "kl": 0.08251953125,
      "learning_rate": 2.4162398742473216e-06,
      "loss": 0.0033,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1915
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 447.6875,
      "epoch": 1.5328,
      "grad_norm": 0.05691963307194323,
      "kl": 0.09326171875,
      "learning_rate": 2.4141466846145332e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1916
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 437.796875,
      "epoch": 1.5335999999999999,
      "grad_norm": 0.6243776782610874,
      "kl": 0.08935546875,
      "learning_rate": 2.4120535552369057e-06,
      "loss": 0.0036,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 1917
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 455.8125,
      "epoch": 1.5344,
      "grad_norm": 0.05135058165921652,
      "kl": 0.0751953125,
      "learning_rate": 2.4099604875834796e-06,
      "loss": 0.003,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1918
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 408.390625,
      "epoch": 1.5352000000000001,
      "grad_norm": 0.5251489637543353,
      "kl": 0.09033203125,
      "learning_rate": 2.407867483123248e-06,
      "loss": 0.0036,
      "reward": 1.90625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1919
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 470.125,
      "epoch": 1.536,
      "grad_norm": 0.4654445293962623,
      "kl": 0.08056640625,
      "learning_rate": 2.4057745433251637e-06,
      "loss": 0.0032,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1920
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 451.96875,
      "epoch": 1.5368,
      "grad_norm": 0.3780116267775583,
      "kl": 0.07666015625,
      "learning_rate": 2.4036816696581326e-06,
      "loss": 0.0031,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1921
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 452.015625,
      "epoch": 1.5375999999999999,
      "grad_norm": 0.04968332132414957,
      "kl": 0.078125,
      "learning_rate": 2.401588863591013e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1922
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 422.390625,
      "epoch": 1.5384,
      "grad_norm": 0.5597047714476183,
      "kl": 0.0751953125,
      "learning_rate": 2.3994961265926166e-06,
      "loss": 0.003,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 1923
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 401.703125,
      "epoch": 1.5392000000000001,
      "grad_norm": 0.42614631105928746,
      "kl": 0.080078125,
      "learning_rate": 2.3974034601317085e-06,
      "loss": 0.0032,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1924
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 473.9375,
      "epoch": 1.54,
      "grad_norm": 0.7449681026166073,
      "kl": 0.08447265625,
      "learning_rate": 2.3953108656770018e-06,
      "loss": 0.0034,
      "reward": 1.75,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1925
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 505.890625,
      "epoch": 1.5408,
      "grad_norm": 0.3399618763451221,
      "kl": 0.0771484375,
      "learning_rate": 2.3932183446971584e-06,
      "loss": 0.0031,
      "reward": 1.421875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 1926
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 405.40625,
      "epoch": 1.5415999999999999,
      "grad_norm": 0.048034442827705685,
      "kl": 0.076171875,
      "learning_rate": 2.3911258986607907e-06,
      "loss": 0.0031,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1927
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 412.78125,
      "epoch": 1.5424,
      "grad_norm": 0.5970261805180694,
      "kl": 0.091796875,
      "learning_rate": 2.3890335290364596e-06,
      "loss": 0.0037,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1928
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 396.421875,
      "epoch": 1.5432000000000001,
      "grad_norm": 1.2450282207798113,
      "kl": 0.08740234375,
      "learning_rate": 2.386941237292669e-06,
      "loss": 0.0035,
      "reward": 1.546875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 1929
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 377.140625,
      "epoch": 1.544,
      "grad_norm": 0.4069990345139908,
      "kl": 0.09228515625,
      "learning_rate": 2.3848490248978693e-06,
      "loss": 0.0037,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1930
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 376.75,
      "epoch": 1.5448,
      "grad_norm": 0.41353856720159765,
      "kl": 0.08642578125,
      "learning_rate": 2.3827568933204576e-06,
      "loss": 0.0035,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1931
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 440.6875,
      "epoch": 1.5455999999999999,
      "grad_norm": 0.3722024727369536,
      "kl": 0.09423828125,
      "learning_rate": 2.3806648440287715e-06,
      "loss": 0.0038,
      "reward": 1.59375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 1932
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 444.390625,
      "epoch": 1.5464,
      "grad_norm": 0.29316398916960057,
      "kl": 0.07568359375,
      "learning_rate": 2.378572878491091e-06,
      "loss": 0.003,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1933
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.5,
      "completion_length": 411.875,
      "epoch": 1.5472000000000001,
      "grad_norm": 0.05588848785297944,
      "kl": 0.0947265625,
      "learning_rate": 2.376480998175638e-06,
      "loss": 0.0038,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 1934
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 408.796875,
      "epoch": 1.548,
      "grad_norm": 0.3869799472767756,
      "kl": 0.0732421875,
      "learning_rate": 2.3743892045505764e-06,
      "loss": 0.0029,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 1935
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 411.203125,
      "epoch": 1.5488,
      "grad_norm": 0.05040098743195126,
      "kl": 0.08154296875,
      "learning_rate": 2.372297499084006e-06,
      "loss": 0.0033,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1936
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 414.5,
      "epoch": 1.5495999999999999,
      "grad_norm": 0.09941455691374751,
      "kl": 0.080078125,
      "learning_rate": 2.3702058832439667e-06,
      "loss": 0.0032,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1937
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 410.390625,
      "epoch": 1.5504,
      "grad_norm": 0.049595646976077816,
      "kl": 0.0810546875,
      "learning_rate": 2.368114358498434e-06,
      "loss": 0.0032,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1938
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 489.515625,
      "epoch": 1.5512000000000001,
      "grad_norm": 0.058668874610572117,
      "kl": 0.07080078125,
      "learning_rate": 2.366022926315322e-06,
      "loss": 0.0028,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1939
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 423.4375,
      "epoch": 1.552,
      "grad_norm": 0.4576779353248496,
      "kl": 0.09521484375,
      "learning_rate": 2.3639315881624776e-06,
      "loss": 0.0038,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.984375,
      "step": 1940
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 385.453125,
      "epoch": 1.5528,
      "grad_norm": 0.41280531013291405,
      "kl": 0.1005859375,
      "learning_rate": 2.361840345507683e-06,
      "loss": 0.004,
      "reward": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 1941
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 456.75,
      "epoch": 1.5535999999999999,
      "grad_norm": 0.051528662637933556,
      "kl": 0.083984375,
      "learning_rate": 2.359749199818651e-06,
      "loss": 0.0034,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1942
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 402.359375,
      "epoch": 1.5544,
      "grad_norm": 0.44677184269484693,
      "kl": 0.08544921875,
      "learning_rate": 2.3576581525630297e-06,
      "loss": 0.0034,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1943
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 428.171875,
      "epoch": 1.5552000000000001,
      "grad_norm": 0.4934843447792334,
      "kl": 0.08154296875,
      "learning_rate": 2.355567205208397e-06,
      "loss": 0.0033,
      "reward": 1.6875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1944
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 445.40625,
      "epoch": 1.556,
      "grad_norm": 1.2941746037157797,
      "kl": 0.0849609375,
      "learning_rate": 2.353476359222259e-06,
      "loss": 0.0034,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1945
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 411.671875,
      "epoch": 1.5568,
      "grad_norm": 0.7087891414321927,
      "kl": 0.09521484375,
      "learning_rate": 2.351385616072052e-06,
      "loss": 0.0038,
      "reward": 1.796875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 1946
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 441.796875,
      "epoch": 1.5575999999999999,
      "grad_norm": 0.7033943265399606,
      "kl": 0.08935546875,
      "learning_rate": 2.3492949772251418e-06,
      "loss": 0.0036,
      "reward": 1.671875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 1947
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 435.203125,
      "epoch": 1.5584,
      "grad_norm": 0.517283215443088,
      "kl": 0.08349609375,
      "learning_rate": 2.3472044441488175e-06,
      "loss": 0.0033,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1948
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 422.203125,
      "epoch": 1.5592000000000001,
      "grad_norm": 0.32980988793749694,
      "kl": 0.0947265625,
      "learning_rate": 2.345114018310295e-06,
      "loss": 0.0038,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1949
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 520.0625,
      "epoch": 1.56,
      "grad_norm": 0.3253568610855632,
      "kl": 0.0771484375,
      "learning_rate": 2.3430237011767166e-06,
      "loss": 0.0031,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 1950
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 427.390625,
      "epoch": 1.5608,
      "grad_norm": 0.9188973196173563,
      "kl": 0.09716796875,
      "learning_rate": 2.3409334942151485e-06,
      "loss": 0.0039,
      "reward": 1.734375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 1951
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 398.109375,
      "epoch": 1.5615999999999999,
      "grad_norm": 0.05237912104856825,
      "kl": 0.08251953125,
      "learning_rate": 2.3388433988925767e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1952
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 434.0625,
      "epoch": 1.5624,
      "grad_norm": 0.8733536196484487,
      "kl": 0.09765625,
      "learning_rate": 2.3367534166759105e-06,
      "loss": 0.0039,
      "reward": 1.9375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 1953
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 481.765625,
      "epoch": 1.5632000000000001,
      "grad_norm": 0.3414651117889524,
      "kl": 0.087890625,
      "learning_rate": 2.3346635490319815e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 1954
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 429.015625,
      "epoch": 1.564,
      "grad_norm": 0.8019378872289062,
      "kl": 0.09033203125,
      "learning_rate": 2.3325737974275382e-06,
      "loss": 0.0036,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 1955
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 450.71875,
      "epoch": 1.5648,
      "grad_norm": 0.37614842106850166,
      "kl": 0.09130859375,
      "learning_rate": 2.3304841633292487e-06,
      "loss": 0.0037,
      "reward": 1.53125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1956
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 413.90625,
      "epoch": 1.5655999999999999,
      "grad_norm": 0.920505096249008,
      "kl": 0.08740234375,
      "learning_rate": 2.328394648203698e-06,
      "loss": 0.0035,
      "reward": 1.828125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1957
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 469.453125,
      "epoch": 1.5664,
      "grad_norm": 0.06176893859386287,
      "kl": 0.09375,
      "learning_rate": 2.32630525351739e-06,
      "loss": 0.0038,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1958
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 461.671875,
      "epoch": 1.5672000000000001,
      "grad_norm": 0.6721447259674301,
      "kl": 0.09765625,
      "learning_rate": 2.324215980736741e-06,
      "loss": 0.0039,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1959
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 438.40625,
      "epoch": 1.568,
      "grad_norm": 0.6466685139029245,
      "kl": 0.0869140625,
      "learning_rate": 2.3221268313280836e-06,
      "loss": 0.0035,
      "reward": 1.84375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1960
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 444.53125,
      "epoch": 1.5688,
      "grad_norm": 0.6677164989481668,
      "kl": 0.0830078125,
      "learning_rate": 2.320037806757662e-06,
      "loss": 0.0033,
      "reward": 1.84375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1961
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 410.640625,
      "epoch": 1.5695999999999999,
      "grad_norm": 0.05529813702949351,
      "kl": 0.08251953125,
      "learning_rate": 2.317948908491636e-06,
      "loss": 0.0033,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1962
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 415.234375,
      "epoch": 1.5704,
      "grad_norm": 0.3663790734074994,
      "kl": 0.09375,
      "learning_rate": 2.315860137996074e-06,
      "loss": 0.0038,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 1963
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 518.40625,
      "epoch": 1.5712000000000002,
      "grad_norm": 0.3820730654813393,
      "kl": 0.08349609375,
      "learning_rate": 2.3137714967369544e-06,
      "loss": 0.0033,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1964
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 400.359375,
      "epoch": 1.572,
      "grad_norm": 0.5072571106126055,
      "kl": 0.0888671875,
      "learning_rate": 2.3116829861801687e-06,
      "loss": 0.0036,
      "reward": 1.578125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 1965
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 453.640625,
      "epoch": 1.5728,
      "grad_norm": 0.7357031421006998,
      "kl": 0.099609375,
      "learning_rate": 2.3095946077915115e-06,
      "loss": 0.004,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1966
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 392.828125,
      "epoch": 1.5735999999999999,
      "grad_norm": 0.5710762790424448,
      "kl": 0.10009765625,
      "learning_rate": 2.307506363036688e-06,
      "loss": 0.004,
      "reward": 1.65625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1967
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 406.640625,
      "epoch": 1.5744,
      "grad_norm": 0.6056496600123321,
      "kl": 0.10546875,
      "learning_rate": 2.305418253381309e-06,
      "loss": 0.0042,
      "reward": 1.65625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 1968
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 423.578125,
      "epoch": 1.5752000000000002,
      "grad_norm": 0.7841380166434793,
      "kl": 0.1171875,
      "learning_rate": 2.3033302802908895e-06,
      "loss": 0.0047,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1969
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 427.953125,
      "epoch": 1.576,
      "grad_norm": 0.7099695625206764,
      "kl": 0.0908203125,
      "learning_rate": 2.301242445230851e-06,
      "loss": 0.0036,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1970
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 439.375,
      "epoch": 1.5768,
      "grad_norm": 0.48466629449866044,
      "kl": 0.0869140625,
      "learning_rate": 2.299154749666515e-06,
      "loss": 0.0035,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1971
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 442.78125,
      "epoch": 1.5776,
      "grad_norm": 0.8487171031141673,
      "kl": 0.09619140625,
      "learning_rate": 2.2970671950631066e-06,
      "loss": 0.0038,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1972
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 399.15625,
      "epoch": 1.5784,
      "grad_norm": 0.6884572205620748,
      "kl": 0.09423828125,
      "learning_rate": 2.2949797828857527e-06,
      "loss": 0.0038,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 1973
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 403.109375,
      "epoch": 1.5792000000000002,
      "grad_norm": 0.4376278554728628,
      "kl": 0.08935546875,
      "learning_rate": 2.2928925145994798e-06,
      "loss": 0.0036,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 1974
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 473.484375,
      "epoch": 1.58,
      "grad_norm": 0.612301840169861,
      "kl": 0.08740234375,
      "learning_rate": 2.290805391669212e-06,
      "loss": 0.0035,
      "reward": 1.6875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 1975
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 461.765625,
      "epoch": 1.5808,
      "grad_norm": 1.1403517593852999,
      "kl": 0.1015625,
      "learning_rate": 2.2887184155597725e-06,
      "loss": 0.0041,
      "reward": 1.875,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1976
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 432.234375,
      "epoch": 1.5816,
      "grad_norm": 0.6715587617536881,
      "kl": 0.0908203125,
      "learning_rate": 2.286631587735883e-06,
      "loss": 0.0036,
      "reward": 1.859375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 1977
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 394.484375,
      "epoch": 1.5824,
      "grad_norm": 0.5746937057612103,
      "kl": 0.095703125,
      "learning_rate": 2.2845449096621583e-06,
      "loss": 0.0038,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 1978
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 374.25,
      "epoch": 1.5832000000000002,
      "grad_norm": 0.6515198071616948,
      "kl": 0.10009765625,
      "learning_rate": 2.282458382803109e-06,
      "loss": 0.004,
      "reward": 1.84375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 1979
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 418.390625,
      "epoch": 1.584,
      "grad_norm": 0.06183753410933011,
      "kl": 0.08837890625,
      "learning_rate": 2.280372008623142e-06,
      "loss": 0.0035,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1980
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 419.859375,
      "epoch": 1.5848,
      "grad_norm": 0.7247649129352906,
      "kl": 0.10009765625,
      "learning_rate": 2.2782857885865538e-06,
      "loss": 0.004,
      "reward": 1.765625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1981
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 425.546875,
      "epoch": 1.5856,
      "grad_norm": 0.058504336371987986,
      "kl": 0.07861328125,
      "learning_rate": 2.2761997241575335e-06,
      "loss": 0.0031,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1982
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 464.09375,
      "epoch": 1.5864,
      "grad_norm": 0.059325817514382304,
      "kl": 0.0927734375,
      "learning_rate": 2.274113816800161e-06,
      "loss": 0.0037,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1983
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 427.15625,
      "epoch": 1.5872000000000002,
      "grad_norm": 0.44834264209595615,
      "kl": 0.08154296875,
      "learning_rate": 2.272028067978408e-06,
      "loss": 0.0033,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1984
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 380.671875,
      "epoch": 1.588,
      "grad_norm": 0.07685436523479029,
      "kl": 0.09326171875,
      "learning_rate": 2.2699424791561324e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 1985
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 447.171875,
      "epoch": 1.5888,
      "grad_norm": 0.060732157570829066,
      "kl": 0.0966796875,
      "learning_rate": 2.267857051797081e-06,
      "loss": 0.0039,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1986
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 419.40625,
      "epoch": 1.5896,
      "grad_norm": 0.680513554192407,
      "kl": 0.0947265625,
      "learning_rate": 2.265771787364886e-06,
      "loss": 0.0038,
      "reward": 1.671875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 1987
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 419.5625,
      "epoch": 1.5904,
      "grad_norm": 0.6776229162099154,
      "kl": 0.10595703125,
      "learning_rate": 2.263686687323068e-06,
      "loss": 0.0042,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1988
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 428.96875,
      "epoch": 1.5912,
      "grad_norm": 0.05980076735866545,
      "kl": 0.0830078125,
      "learning_rate": 2.261601753135029e-06,
      "loss": 0.0033,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1989
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 427.640625,
      "epoch": 1.592,
      "grad_norm": 0.562923329652774,
      "kl": 0.10400390625,
      "learning_rate": 2.259516986264057e-06,
      "loss": 0.0042,
      "reward": 1.703125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 1990
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 402.796875,
      "epoch": 1.5928,
      "grad_norm": 0.6307637783596173,
      "kl": 0.09033203125,
      "learning_rate": 2.2574323881733202e-06,
      "loss": 0.0036,
      "reward": 1.78125,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 1991
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 423.421875,
      "epoch": 1.5936,
      "grad_norm": 0.6263483609701712,
      "kl": 0.07861328125,
      "learning_rate": 2.255347960325871e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 1992
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 438.640625,
      "epoch": 1.5944,
      "grad_norm": 0.5561684101608247,
      "kl": 0.0703125,
      "learning_rate": 2.2532637041846423e-06,
      "loss": 0.0028,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 1993
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 434.765625,
      "epoch": 1.5952,
      "grad_norm": 2.007077470371618,
      "kl": 0.1044921875,
      "learning_rate": 2.2511796212124424e-06,
      "loss": 0.0042,
      "reward": 1.53125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 1994
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 403.265625,
      "epoch": 1.596,
      "grad_norm": 0.052762762834036386,
      "kl": 0.07763671875,
      "learning_rate": 2.2490957128719627e-06,
      "loss": 0.0031,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 1995
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 431.09375,
      "epoch": 1.5968,
      "grad_norm": 0.540183431697575,
      "kl": 0.0927734375,
      "learning_rate": 2.247011980625771e-06,
      "loss": 0.0037,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 1996
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 399.546875,
      "epoch": 1.5976,
      "grad_norm": 0.709139819373853,
      "kl": 0.1083984375,
      "learning_rate": 2.2449284259363093e-06,
      "loss": 0.0043,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 1997
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 398.703125,
      "epoch": 1.5984,
      "grad_norm": 0.6191389046837881,
      "kl": 0.0908203125,
      "learning_rate": 2.2428450502658964e-06,
      "loss": 0.0036,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 1998
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 423.515625,
      "epoch": 1.5992,
      "grad_norm": 0.6661522188284288,
      "kl": 0.09375,
      "learning_rate": 2.240761855076727e-06,
      "loss": 0.0037,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 1999
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 472.109375,
      "epoch": 1.6,
      "grad_norm": 0.6794216918715973,
      "kl": 0.09228515625,
      "learning_rate": 2.238678841830867e-06,
      "loss": 0.0037,
      "reward": 1.390625,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 2000
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 502.25,
      "epoch": 1.6008,
      "grad_norm": 0.4924265749469055,
      "kl": 0.0859375,
      "learning_rate": 2.2365960119902543e-06,
      "loss": 0.0034,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 2001
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 441.125,
      "epoch": 1.6016,
      "grad_norm": 0.6632221972914333,
      "kl": 0.08984375,
      "learning_rate": 2.2345133670167e-06,
      "loss": 0.0036,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2002
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 403.984375,
      "epoch": 1.6024,
      "grad_norm": 0.5731803399766031,
      "kl": 0.09521484375,
      "learning_rate": 2.232430908371885e-06,
      "loss": 0.0038,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 2003
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 429.59375,
      "epoch": 1.6032,
      "grad_norm": 0.05592152651951858,
      "kl": 0.08984375,
      "learning_rate": 2.2303486375173586e-06,
      "loss": 0.0036,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2004
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 393.484375,
      "epoch": 1.604,
      "grad_norm": 0.5149817346204347,
      "kl": 0.0888671875,
      "learning_rate": 2.228266555914538e-06,
      "loss": 0.0035,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 2005
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 420.921875,
      "epoch": 1.6048,
      "grad_norm": 0.5618040729575181,
      "kl": 0.1083984375,
      "learning_rate": 2.2261846650247077e-06,
      "loss": 0.0043,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 2006
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 431.25,
      "epoch": 1.6056,
      "grad_norm": 0.5471707557096966,
      "kl": 0.08154296875,
      "learning_rate": 2.224102966309021e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2007
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 364.625,
      "epoch": 1.6064,
      "grad_norm": 0.5750125022504474,
      "kl": 0.083984375,
      "learning_rate": 2.2220214612284925e-06,
      "loss": 0.0034,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2008
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 408.53125,
      "epoch": 1.6072,
      "grad_norm": 0.5028381419417359,
      "kl": 0.0966796875,
      "learning_rate": 2.2199401512440037e-06,
      "loss": 0.0039,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2009
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 417.796875,
      "epoch": 1.608,
      "grad_norm": 1.0662632177333176,
      "kl": 0.08984375,
      "learning_rate": 2.2178590378162957e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2010
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 442.4375,
      "epoch": 1.6088,
      "grad_norm": 0.809973613432914,
      "kl": 0.109375,
      "learning_rate": 2.215778122405977e-06,
      "loss": 0.0044,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 2011
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 402.453125,
      "epoch": 1.6096,
      "grad_norm": 0.40635075586272323,
      "kl": 0.10107421875,
      "learning_rate": 2.2136974064735132e-06,
      "loss": 0.0041,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2012
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 454.96875,
      "epoch": 1.6104,
      "grad_norm": 0.3458725202502868,
      "kl": 0.08544921875,
      "learning_rate": 2.2116168914792293e-06,
      "loss": 0.0034,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 2013
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 412.15625,
      "epoch": 1.6112,
      "grad_norm": 0.7825353499057853,
      "kl": 0.09716796875,
      "learning_rate": 2.209536578883313e-06,
      "loss": 0.0039,
      "reward": 1.90625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 2014
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 450.75,
      "epoch": 1.612,
      "grad_norm": 0.40853927313163246,
      "kl": 0.099609375,
      "learning_rate": 2.207456470145807e-06,
      "loss": 0.004,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2015
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 398.46875,
      "epoch": 1.6128,
      "grad_norm": 0.39259991374650216,
      "kl": 0.09375,
      "learning_rate": 2.205376566726611e-06,
      "loss": 0.0037,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2016
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 403.765625,
      "epoch": 1.6136,
      "grad_norm": 0.38809053469550847,
      "kl": 0.0703125,
      "learning_rate": 2.2032968700854813e-06,
      "loss": 0.0028,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2017
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 430.359375,
      "epoch": 1.6143999999999998,
      "grad_norm": 1.6390571938325302,
      "kl": 0.11865234375,
      "learning_rate": 2.2012173816820297e-06,
      "loss": 0.0047,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2018
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 437.4375,
      "epoch": 1.6152,
      "grad_norm": 0.76433564696172,
      "kl": 0.11376953125,
      "learning_rate": 2.1991381029757216e-06,
      "loss": 0.0046,
      "reward": 1.515625,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 2019
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 466.34375,
      "epoch": 1.616,
      "grad_norm": 0.05320376064530623,
      "kl": 0.07763671875,
      "learning_rate": 2.1970590354258745e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2020
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 389.640625,
      "epoch": 1.6168,
      "grad_norm": 0.39452858608194213,
      "kl": 0.099609375,
      "learning_rate": 2.1949801804916563e-06,
      "loss": 0.004,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2021
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 409.953125,
      "epoch": 1.6176,
      "grad_norm": 0.6108606028152092,
      "kl": 0.099609375,
      "learning_rate": 2.19290153963209e-06,
      "loss": 0.004,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2022
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 387.34375,
      "epoch": 1.6183999999999998,
      "grad_norm": 0.07378419376928622,
      "kl": 0.08349609375,
      "learning_rate": 2.190823114306045e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2023
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 434.09375,
      "epoch": 1.6192,
      "grad_norm": 0.06119824242919308,
      "kl": 0.0810546875,
      "learning_rate": 2.188744905972239e-06,
      "loss": 0.0032,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2024
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 479.453125,
      "epoch": 1.62,
      "grad_norm": 0.07068022480178383,
      "kl": 0.09375,
      "learning_rate": 2.186666916089239e-06,
      "loss": 0.0037,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2025
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 375.671875,
      "epoch": 1.6208,
      "grad_norm": 0.6221384797782726,
      "kl": 0.1123046875,
      "learning_rate": 2.1845891461154604e-06,
      "loss": 0.0045,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2026
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 473.515625,
      "epoch": 1.6216,
      "grad_norm": 0.3793672283807978,
      "kl": 0.11962890625,
      "learning_rate": 2.1825115975091594e-06,
      "loss": 0.0048,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 2027
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 439.125,
      "epoch": 1.6223999999999998,
      "grad_norm": 0.6990150534191653,
      "kl": 0.09765625,
      "learning_rate": 2.1804342717284414e-06,
      "loss": 0.0039,
      "reward": 1.75,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2028
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 425.828125,
      "epoch": 1.6232,
      "grad_norm": 0.5126155154548316,
      "kl": 0.12255859375,
      "learning_rate": 2.1783571702312523e-06,
      "loss": 0.0049,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 2029
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 430.96875,
      "epoch": 1.624,
      "grad_norm": 1.1625177114219285,
      "kl": 0.11181640625,
      "learning_rate": 2.176280294475383e-06,
      "loss": 0.0045,
      "reward": 1.703125,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 2030
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 451.265625,
      "epoch": 1.6248,
      "grad_norm": 0.0790836077361706,
      "kl": 0.09228515625,
      "learning_rate": 2.174203645918464e-06,
      "loss": 0.0037,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 2031
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 456.578125,
      "epoch": 1.6256,
      "grad_norm": 0.5180100414711042,
      "kl": 0.1376953125,
      "learning_rate": 2.172127226017967e-06,
      "loss": 0.0055,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.984375,
      "step": 2032
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 491.734375,
      "epoch": 1.6263999999999998,
      "grad_norm": 0.5959103272952211,
      "kl": 0.10302734375,
      "learning_rate": 2.1700510362312053e-06,
      "loss": 0.0041,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 2033
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 409.890625,
      "epoch": 1.6272,
      "grad_norm": 1.0119783425654472,
      "kl": 0.1015625,
      "learning_rate": 2.1679750780153265e-06,
      "loss": 0.0041,
      "reward": 1.828125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2034
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 459.078125,
      "epoch": 1.6280000000000001,
      "grad_norm": 0.6102145825112454,
      "kl": 0.1337890625,
      "learning_rate": 2.1658993528273196e-06,
      "loss": 0.0053,
      "reward": 1.875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2035
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 458.703125,
      "epoch": 1.6288,
      "grad_norm": 0.670395486538238,
      "kl": 0.11376953125,
      "learning_rate": 2.163823862124007e-06,
      "loss": 0.0045,
      "reward": 1.671875,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2036
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 426.609375,
      "epoch": 1.6296,
      "grad_norm": 0.09793812265645296,
      "kl": 0.1123046875,
      "learning_rate": 2.1617486073620496e-06,
      "loss": 0.0045,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2037
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 485.6875,
      "epoch": 1.6303999999999998,
      "grad_norm": 0.846967509312758,
      "kl": 0.10302734375,
      "learning_rate": 2.15967358999794e-06,
      "loss": 0.0041,
      "reward": 1.890625,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 2038
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 415.6875,
      "epoch": 1.6312,
      "grad_norm": 1.185056589277107,
      "kl": 0.1630859375,
      "learning_rate": 2.1575988114880057e-06,
      "loss": 0.0065,
      "reward": 1.5625,
      "reward_std": 0.2709311842918396,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 2039
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 405.671875,
      "epoch": 1.6320000000000001,
      "grad_norm": 0.7997307398404204,
      "kl": 0.1572265625,
      "learning_rate": 2.155524273288405e-06,
      "loss": 0.0063,
      "reward": 1.515625,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 2040
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 490.234375,
      "epoch": 1.6328,
      "grad_norm": 0.758577387618897,
      "kl": 0.09375,
      "learning_rate": 2.15344997685513e-06,
      "loss": 0.0037,
      "reward": 1.625,
      "reward_std": 0.20337893068790436,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 2041
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 408.890625,
      "epoch": 1.6336,
      "grad_norm": 0.7824015696554165,
      "kl": 0.1591796875,
      "learning_rate": 2.1513759236440024e-06,
      "loss": 0.0064,
      "reward": 1.421875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 2042
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 438.03125,
      "epoch": 1.6343999999999999,
      "grad_norm": 1.1683127251260226,
      "kl": 0.1474609375,
      "learning_rate": 2.1493021151106704e-06,
      "loss": 0.0059,
      "reward": 1.65625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 2043
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 412.96875,
      "epoch": 1.6352,
      "grad_norm": 0.14790733519340812,
      "kl": 0.1484375,
      "learning_rate": 2.147228552710614e-06,
      "loss": 0.0059,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2044
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 414.109375,
      "epoch": 1.6360000000000001,
      "grad_norm": 0.8931486283271186,
      "kl": 0.1474609375,
      "learning_rate": 2.145155237899139e-06,
      "loss": 0.0059,
      "reward": 1.640625,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 2045
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 424.15625,
      "epoch": 1.6368,
      "grad_norm": 1.3417958553699005,
      "kl": 0.1201171875,
      "learning_rate": 2.143082172131378e-06,
      "loss": 0.0048,
      "reward": 1.859375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2046
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 435.046875,
      "epoch": 1.6376,
      "grad_norm": 1.0701853002218584,
      "kl": 0.1748046875,
      "learning_rate": 2.141009356862288e-06,
      "loss": 0.007,
      "reward": 1.9375,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 2047
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 446.375,
      "epoch": 1.6383999999999999,
      "grad_norm": 0.8067529629088794,
      "kl": 0.2001953125,
      "learning_rate": 2.138936793546649e-06,
      "loss": 0.008,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 2048
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 423.09375,
      "epoch": 1.6392,
      "grad_norm": 0.9583270906668814,
      "kl": 0.1552734375,
      "learning_rate": 2.1368644836390684e-06,
      "loss": 0.0062,
      "reward": 1.859375,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 2049
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 448.421875,
      "epoch": 1.6400000000000001,
      "grad_norm": 0.6525017486732866,
      "kl": 0.1923828125,
      "learning_rate": 2.134792428593971e-06,
      "loss": 0.0077,
      "reward": 1.5625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 2050
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 406.828125,
      "epoch": 1.6408,
      "grad_norm": 1.138673694517526,
      "kl": 0.2177734375,
      "learning_rate": 2.1327206298656055e-06,
      "loss": 0.0087,
      "reward": 1.6875,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 2051
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 395.40625,
      "epoch": 1.6416,
      "grad_norm": 0.8979409996731252,
      "kl": 0.2421875,
      "learning_rate": 2.130649088908041e-06,
      "loss": 0.0097,
      "reward": 1.71875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 2052
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 386.75,
      "epoch": 1.6423999999999999,
      "grad_norm": 1.351684688785436,
      "kl": 0.357421875,
      "learning_rate": 2.1285778071751638e-06,
      "loss": 0.0143,
      "reward": 1.734375,
      "reward_std": 0.2993341088294983,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.96875,
      "step": 2053
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 382.53125,
      "epoch": 1.6432,
      "grad_norm": 0.9290746829045567,
      "kl": 0.353515625,
      "learning_rate": 2.126506786120678e-06,
      "loss": 0.0141,
      "reward": 1.703125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 2054
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 377.34375,
      "epoch": 1.6440000000000001,
      "grad_norm": 0.8947963003590713,
      "kl": 0.275390625,
      "learning_rate": 2.1244360271981073e-06,
      "loss": 0.0111,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2055
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 387.140625,
      "epoch": 1.6448,
      "grad_norm": 0.8756873414122475,
      "kl": 0.400390625,
      "learning_rate": 2.1223655318607907e-06,
      "loss": 0.016,
      "reward": 1.5,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.984375,
      "step": 2056
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 446.96875,
      "epoch": 1.6456,
      "grad_norm": 1.9104153923874023,
      "kl": 0.40625,
      "learning_rate": 2.1202953015618794e-06,
      "loss": 0.0162,
      "reward": 1.71875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 2057
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 406.5625,
      "epoch": 1.6463999999999999,
      "grad_norm": 2.7428791774243075,
      "kl": 1.28125,
      "learning_rate": 2.1182253377543428e-06,
      "loss": 0.0514,
      "reward": 1.734375,
      "reward_std": 0.4639715254306793,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.875,
      "step": 2058
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 423.796875,
      "epoch": 1.6472,
      "grad_norm": 2.8132777340641075,
      "kl": 1.6640625,
      "learning_rate": 2.116155641890959e-06,
      "loss": 0.0665,
      "reward": 1.453125,
      "reward_std": 0.5810762643814087,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.84375,
      "step": 2059
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 440.75,
      "epoch": 1.6480000000000001,
      "grad_norm": 2.618284937298759,
      "kl": 0.609375,
      "learning_rate": 2.1140862154243223e-06,
      "loss": 0.0244,
      "reward": 1.734375,
      "reward_std": 0.2877861559391022,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.953125,
      "step": 2060
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 491.140625,
      "epoch": 1.6488,
      "grad_norm": 2.929192733231867,
      "kl": 2.5,
      "learning_rate": 2.1120170598068353e-06,
      "loss": 0.0998,
      "reward": 1.640625,
      "reward_std": 0.4592459797859192,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.859375,
      "step": 2061
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 453.375,
      "epoch": 1.6496,
      "grad_norm": 0.8376506704489841,
      "kl": 0.83984375,
      "learning_rate": 2.109948176490711e-06,
      "loss": 0.0335,
      "reward": 1.859375,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.96875,
      "step": 2062
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 471.546875,
      "epoch": 1.6503999999999999,
      "grad_norm": 3.354611778255606,
      "kl": 46.5,
      "learning_rate": 2.10787956692797e-06,
      "loss": 1.8653,
      "reward": 1.546875,
      "reward_std": 0.5070846080780029,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.828125,
      "step": 2063
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 433.234375,
      "epoch": 1.6512,
      "grad_norm": 2.632392548689691,
      "kl": 24.125,
      "learning_rate": 2.1058112325704436e-06,
      "loss": 0.9703,
      "reward": 1.578125,
      "reward_std": 0.5543705821037292,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.875,
      "step": 2064
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 495.40625,
      "epoch": 1.6520000000000001,
      "grad_norm": 6.459755037363299,
      "kl": 8.375,
      "learning_rate": 2.103743174869769e-06,
      "loss": 0.3359,
      "reward": 1.421875,
      "reward_std": 0.5879159569740295,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.78125,
      "step": 2065
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 389.734375,
      "epoch": 1.6528,
      "grad_norm": 0.37028672863703244,
      "kl": 0.0888671875,
      "learning_rate": 2.1016753952773867e-06,
      "loss": 0.0036,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2066
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 367.15625,
      "epoch": 1.6536,
      "grad_norm": 1.4978545902795564,
      "kl": 1.0078125,
      "learning_rate": 2.0996078952445453e-06,
      "loss": 0.0403,
      "reward": 1.609375,
      "reward_std": 0.22340349853038788,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.953125,
      "step": 2067
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 351.84375,
      "epoch": 1.6543999999999999,
      "grad_norm": 2.3919512724033165,
      "kl": 1.2421875,
      "learning_rate": 2.0975406762222966e-06,
      "loss": 0.0496,
      "reward": 1.671875,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.953125,
      "step": 2068
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 377.90625,
      "epoch": 1.6552,
      "grad_norm": 0.13528200674066437,
      "kl": 0.0908203125,
      "learning_rate": 2.095473739661494e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2069
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 322.953125,
      "epoch": 1.6560000000000001,
      "grad_norm": 1.3661852926913154,
      "kl": 0.423828125,
      "learning_rate": 2.093407087012791e-06,
      "loss": 0.017,
      "reward": 1.921875,
      "reward_std": 0.09300297498703003,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 2070
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 355.8125,
      "epoch": 1.6568,
      "grad_norm": 1.2447334924326254,
      "kl": 0.2001953125,
      "learning_rate": 2.091340719726647e-06,
      "loss": 0.008,
      "reward": 1.671875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2071
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 367.859375,
      "epoch": 1.6576,
      "grad_norm": 2.8788088395701164,
      "kl": 0.431640625,
      "learning_rate": 2.089274639253317e-06,
      "loss": 0.0172,
      "reward": 1.75,
      "reward_std": 0.3230288028717041,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.90625,
      "step": 2072
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 344.859375,
      "epoch": 1.6583999999999999,
      "grad_norm": 0.4312088131038686,
      "kl": 0.103515625,
      "learning_rate": 2.0872088470428553e-06,
      "loss": 0.0041,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2073
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 333.4375,
      "epoch": 1.6592,
      "grad_norm": 0.712412782267239,
      "kl": 0.09423828125,
      "learning_rate": 2.0851433445451142e-06,
      "loss": 0.0038,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2074
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 342.3125,
      "epoch": 1.6600000000000001,
      "grad_norm": 0.47028550784080814,
      "kl": 0.103515625,
      "learning_rate": 2.0830781332097446e-06,
      "loss": 0.0042,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 2075
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 364.90625,
      "epoch": 1.6608,
      "grad_norm": 0.11942239325951794,
      "kl": 0.0927734375,
      "learning_rate": 2.08101321448619e-06,
      "loss": 0.0037,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 2076
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 380.5,
      "epoch": 1.6616,
      "grad_norm": 0.08161399529058654,
      "kl": 0.07373046875,
      "learning_rate": 2.0789485898236897e-06,
      "loss": 0.0029,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2077
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 343.03125,
      "epoch": 1.6623999999999999,
      "grad_norm": 0.49450583451663443,
      "kl": 0.1669921875,
      "learning_rate": 2.076884260671276e-06,
      "loss": 0.0067,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 2078
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 322.0,
      "epoch": 1.6632,
      "grad_norm": 1.4657197200618237,
      "kl": 0.111328125,
      "learning_rate": 2.0748202284777775e-06,
      "loss": 0.0044,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 2079
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 361.5,
      "epoch": 1.6640000000000001,
      "grad_norm": 1.0973359588279956,
      "kl": 0.10009765625,
      "learning_rate": 2.072756494691809e-06,
      "loss": 0.004,
      "reward": 1.71875,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 2080
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 380.71875,
      "epoch": 1.6648,
      "grad_norm": 1.3076777684445047,
      "kl": 0.1689453125,
      "learning_rate": 2.070693060761779e-06,
      "loss": 0.0068,
      "reward": 1.640625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 2081
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 383.640625,
      "epoch": 1.6656,
      "grad_norm": 2.6447020055058146,
      "kl": 0.53515625,
      "learning_rate": 2.0686299281358837e-06,
      "loss": 0.0214,
      "reward": 1.71875,
      "reward_std": 0.3503679633140564,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.953125,
      "step": 2082
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 407.1875,
      "epoch": 1.6663999999999999,
      "grad_norm": 0.8757899604388453,
      "kl": 0.390625,
      "learning_rate": 2.0665670982621107e-06,
      "loss": 0.0156,
      "reward": 1.625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 2083
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 403.03125,
      "epoch": 1.6672,
      "grad_norm": 1.1131451419047147,
      "kl": 0.453125,
      "learning_rate": 2.0645045725882334e-06,
      "loss": 0.0181,
      "reward": 1.90625,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.984375,
      "step": 2084
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 423.828125,
      "epoch": 1.6680000000000001,
      "grad_norm": 3.3255435641845077,
      "kl": 1.9921875,
      "learning_rate": 2.0624423525618097e-06,
      "loss": 0.0799,
      "reward": 1.65625,
      "reward_std": 0.4106212854385376,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.890625,
      "step": 2085
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 418.234375,
      "epoch": 1.6688,
      "grad_norm": 2.875306385663048,
      "kl": 1.8828125,
      "learning_rate": 2.0603804396301875e-06,
      "loss": 0.0752,
      "reward": 1.84375,
      "reward_std": 0.3198433816432953,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.90625,
      "step": 2086
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 488.234375,
      "epoch": 1.6696,
      "grad_norm": 5.811269178043532,
      "kl": 7.5,
      "learning_rate": 2.058318835240495e-06,
      "loss": 0.2997,
      "reward": 1.40625,
      "reward_std": 0.5636438131332397,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.71875,
      "step": 2087
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 578.171875,
      "epoch": 1.6703999999999999,
      "grad_norm": 3.063940567925349,
      "kl": 6.625,
      "learning_rate": 2.0562575408396475e-06,
      "loss": 0.266,
      "reward": 1.359375,
      "reward_std": 0.5273522138595581,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.765625,
      "step": 2088
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 488.8125,
      "epoch": 1.6712,
      "grad_norm": 2.1562511039521093,
      "kl": 4.6875,
      "learning_rate": 2.0541965578743373e-06,
      "loss": 0.1873,
      "reward": 1.421875,
      "reward_std": 0.5159524083137512,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.84375,
      "step": 2089
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 506.21875,
      "epoch": 1.6720000000000002,
      "grad_norm": 5.100514446358345,
      "kl": 4.28125,
      "learning_rate": 2.0521358877910446e-06,
      "loss": 0.1713,
      "reward": 1.359375,
      "reward_std": 0.7028868794441223,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.796875,
      "step": 2090
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 466.578125,
      "epoch": 1.6728,
      "grad_norm": 2.844359638606817,
      "kl": 3.6875,
      "learning_rate": 2.0500755320360263e-06,
      "loss": 0.1472,
      "reward": 1.59375,
      "reward_std": 0.506583571434021,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.859375,
      "step": 2091
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 430.75,
      "epoch": 1.6736,
      "grad_norm": 4.278366066617067,
      "kl": 3.265625,
      "learning_rate": 2.048015492055319e-06,
      "loss": 0.1303,
      "reward": 1.546875,
      "reward_std": 0.5261501669883728,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.859375,
      "step": 2092
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 402.59375,
      "epoch": 1.6743999999999999,
      "grad_norm": 3.062156897960549,
      "kl": 3.53125,
      "learning_rate": 2.045955769294737e-06,
      "loss": 0.1412,
      "reward": 1.765625,
      "reward_std": 0.3825550079345703,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.84375,
      "step": 2093
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 477.671875,
      "epoch": 1.6752,
      "grad_norm": 519.8739486273251,
      "kl": 2.421875,
      "learning_rate": 2.0438963651998747e-06,
      "loss": 0.0967,
      "reward": 1.578125,
      "reward_std": 0.4517311453819275,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.84375,
      "step": 2094
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 459.109375,
      "epoch": 1.6760000000000002,
      "grad_norm": 34.63765130965278,
      "kl": 0.69140625,
      "learning_rate": 2.0418372812161015e-06,
      "loss": 0.0277,
      "reward": 1.875,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.984375,
      "step": 2095
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 491.09375,
      "epoch": 1.6768,
      "grad_norm": 1.6566859147044086,
      "kl": 1.4140625,
      "learning_rate": 2.03977851878856e-06,
      "loss": 0.0569,
      "reward": 1.90625,
      "reward_std": 0.22558549046516418,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.953125,
      "step": 2096
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.5,
      "completion_length": 460.578125,
      "epoch": 1.6776,
      "grad_norm": 1.091428748065709,
      "kl": 0.28515625,
      "learning_rate": 2.0377200793621694e-06,
      "loss": 0.0114,
      "reward": 1.4375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.984375,
      "step": 2097
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 394.78125,
      "epoch": 1.6784,
      "grad_norm": 0.7501097936511554,
      "kl": 0.60546875,
      "learning_rate": 2.0356619643816234e-06,
      "loss": 0.0242,
      "reward": 1.59375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 2098
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 432.40625,
      "epoch": 1.6792,
      "grad_norm": 0.9020856731720874,
      "kl": 0.337890625,
      "learning_rate": 2.0336041752913843e-06,
      "loss": 0.0135,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 2099
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 483.65625,
      "epoch": 1.6800000000000002,
      "grad_norm": 1.533215888822757,
      "kl": 0.37109375,
      "learning_rate": 2.031546713535688e-06,
      "loss": 0.0149,
      "reward": 1.6875,
      "reward_std": 0.23319074511528015,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 2100
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 397.328125,
      "epoch": 1.6808,
      "grad_norm": 0.7005043360325897,
      "kl": 0.130859375,
      "learning_rate": 2.029489580558542e-06,
      "loss": 0.0052,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2101
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 458.390625,
      "epoch": 1.6816,
      "grad_norm": 1.4970864399606052,
      "kl": 0.2890625,
      "learning_rate": 2.0274327778037204e-06,
      "loss": 0.0115,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 2102
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 443.75,
      "epoch": 1.6824,
      "grad_norm": 0.3186074496487898,
      "kl": 0.111328125,
      "learning_rate": 2.0253763067147657e-06,
      "loss": 0.0045,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 2103
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 415.15625,
      "epoch": 1.6832,
      "grad_norm": 1.6010840415017298,
      "kl": 1.140625,
      "learning_rate": 2.0233201687349888e-06,
      "loss": 0.0458,
      "reward": 1.84375,
      "reward_std": 0.21556037664413452,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.9375,
      "step": 2104
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 420.390625,
      "epoch": 1.6840000000000002,
      "grad_norm": 0.35779095114683007,
      "kl": 0.0830078125,
      "learning_rate": 2.0212643653074677e-06,
      "loss": 0.0033,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2105
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 429.953125,
      "epoch": 1.6848,
      "grad_norm": 2.534925458175817,
      "kl": 0.8125,
      "learning_rate": 2.019208897875043e-06,
      "loss": 0.0325,
      "reward": 1.9375,
      "reward_std": 0.13719715178012848,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.953125,
      "step": 2106
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 391.9375,
      "epoch": 1.6856,
      "grad_norm": 3.2418327402054525,
      "kl": 1.5,
      "learning_rate": 2.0171537678803222e-06,
      "loss": 0.0597,
      "reward": 1.609375,
      "reward_std": 0.2993341088294983,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.9375,
      "step": 2107
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.25,
      "completion_length": 341.703125,
      "epoch": 1.6864,
      "grad_norm": 2.2100911582973635,
      "kl": 0.51171875,
      "learning_rate": 2.015098976765673e-06,
      "loss": 0.0205,
      "reward": 1.390625,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.96875,
      "step": 2108
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 384.046875,
      "epoch": 1.6872,
      "grad_norm": 1.8927276557942287,
      "kl": 1.0625,
      "learning_rate": 2.0130445259732282e-06,
      "loss": 0.0425,
      "reward": 1.65625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.953125,
      "step": 2109
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 433.734375,
      "epoch": 1.688,
      "grad_norm": 1.5667912067089171,
      "kl": 0.9140625,
      "learning_rate": 2.01099041694488e-06,
      "loss": 0.0365,
      "reward": 1.8125,
      "reward_std": 0.13719715178012848,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.953125,
      "step": 2110
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 414.578125,
      "epoch": 1.6888,
      "grad_norm": 1.7986987795251153,
      "kl": 2.359375,
      "learning_rate": 2.0089366511222815e-06,
      "loss": 0.0943,
      "reward": 1.78125,
      "reward_std": 0.3652152717113495,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.921875,
      "step": 2111
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 419.828125,
      "epoch": 1.6896,
      "grad_norm": 1.217112278948557,
      "kl": 0.58203125,
      "learning_rate": 2.006883229946843e-06,
      "loss": 0.0233,
      "reward": 1.703125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.984375,
      "step": 2112
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 344.703125,
      "epoch": 1.6904,
      "grad_norm": 1.0434906078163486,
      "kl": 0.462890625,
      "learning_rate": 2.0048301548597365e-06,
      "loss": 0.0185,
      "reward": 1.515625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 2113
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 427.375,
      "epoch": 1.6912,
      "grad_norm": 1.7048318956597763,
      "kl": 2.5625,
      "learning_rate": 2.0027774273018894e-06,
      "loss": 0.1026,
      "reward": 1.53125,
      "reward_std": 0.24969476461410522,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.921875,
      "step": 2114
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 382.5,
      "epoch": 1.692,
      "grad_norm": 1.120614382118344,
      "kl": 1.421875,
      "learning_rate": 2.0007250487139827e-06,
      "loss": 0.0566,
      "reward": 1.890625,
      "reward_std": 0.27883461117744446,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.96875,
      "step": 2115
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 459.375,
      "epoch": 1.6928,
      "grad_norm": 1.2304593184727952,
      "kl": 3.140625,
      "learning_rate": 1.998673020536456e-06,
      "loss": 0.1261,
      "reward": 1.71875,
      "reward_std": 0.4387563467025757,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.921875,
      "step": 2116
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 364.484375,
      "epoch": 1.6936,
      "grad_norm": 1.0479482288555169,
      "kl": 2.203125,
      "learning_rate": 1.996621344209503e-06,
      "loss": 0.088,
      "reward": 1.734375,
      "reward_std": 0.2880638837814331,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.9375,
      "step": 2117
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 425.171875,
      "epoch": 1.6944,
      "grad_norm": 0.7930626251555342,
      "kl": 1.3984375,
      "learning_rate": 1.994570021173067e-06,
      "loss": 0.056,
      "reward": 1.640625,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.96875,
      "step": 2118
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 395.546875,
      "epoch": 1.6952,
      "grad_norm": 0.41743046947985185,
      "kl": 0.56640625,
      "learning_rate": 1.9925190528668455e-06,
      "loss": 0.0227,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 2119
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 383.140625,
      "epoch": 1.696,
      "grad_norm": 0.3176723997061967,
      "kl": 0.77734375,
      "learning_rate": 1.990468440730288e-06,
      "loss": 0.0311,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 2120
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 339.5,
      "epoch": 1.6968,
      "grad_norm": 0.9748433744860779,
      "kl": 0.4296875,
      "learning_rate": 1.9884181862025938e-06,
      "loss": 0.0172,
      "reward": 1.578125,
      "reward_std": 0.23531240224838257,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 2121
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 377.84375,
      "epoch": 1.6976,
      "grad_norm": 1.5175450035007318,
      "kl": 2.15625,
      "learning_rate": 1.986368290722709e-06,
      "loss": 0.0863,
      "reward": 1.890625,
      "reward_std": 0.23925507068634033,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.96875,
      "step": 2122
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 380.90625,
      "epoch": 1.6984,
      "grad_norm": 0.391819515717612,
      "kl": 0.609375,
      "learning_rate": 1.9843187557293286e-06,
      "loss": 0.0244,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 2123
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 397.234375,
      "epoch": 1.6992,
      "grad_norm": 1.6924632622452174,
      "kl": 2.609375,
      "learning_rate": 1.9822695826608975e-06,
      "loss": 0.1046,
      "reward": 1.75,
      "reward_std": 0.20751865208148956,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.9375,
      "step": 2124
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 374.078125,
      "epoch": 1.7,
      "grad_norm": 1.052184524765829,
      "kl": 1.109375,
      "learning_rate": 1.9802207729556023e-06,
      "loss": 0.0444,
      "reward": 1.78125,
      "reward_std": 0.13719715178012848,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 2125
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 391.65625,
      "epoch": 1.7008,
      "grad_norm": 0.34578416675042317,
      "kl": 0.55859375,
      "learning_rate": 1.978172328051377e-06,
      "loss": 0.0222,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 2126
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 396.484375,
      "epoch": 1.7016,
      "grad_norm": 0.49065077345498653,
      "kl": 0.7109375,
      "learning_rate": 1.9761242493858987e-06,
      "loss": 0.0284,
      "reward": 1.921875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 2127
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 382.90625,
      "epoch": 1.7024,
      "grad_norm": 0.6389077473900239,
      "kl": 0.8125,
      "learning_rate": 1.9740765383965894e-06,
      "loss": 0.0326,
      "reward": 1.6875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 2128
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 355.1875,
      "epoch": 1.7032,
      "grad_norm": 0.5299376628103857,
      "kl": 0.1005859375,
      "learning_rate": 1.9720291965206097e-06,
      "loss": 0.004,
      "reward": 1.65625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 2129
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 377.875,
      "epoch": 1.704,
      "grad_norm": 0.832740783607803,
      "kl": 0.53125,
      "learning_rate": 1.969982225194864e-06,
      "loss": 0.0213,
      "reward": 1.546875,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 2130
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 394.90625,
      "epoch": 1.7048,
      "grad_norm": 0.4899814699346314,
      "kl": 0.0849609375,
      "learning_rate": 1.9679356258559943e-06,
      "loss": 0.0034,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2131
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 419.4375,
      "epoch": 1.7056,
      "grad_norm": 0.7187222946588002,
      "kl": 0.0849609375,
      "learning_rate": 1.9658893999403847e-06,
      "loss": 0.0034,
      "reward": 1.65625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 2132
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 372.109375,
      "epoch": 1.7064,
      "grad_norm": 0.7439981408427864,
      "kl": 0.61328125,
      "learning_rate": 1.9638435488841543e-06,
      "loss": 0.0246,
      "reward": 1.734375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 2133
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 402.40625,
      "epoch": 1.7072,
      "grad_norm": 0.09514080244625805,
      "kl": 0.091796875,
      "learning_rate": 1.96179807412316e-06,
      "loss": 0.0037,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2134
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 382.40625,
      "epoch": 1.708,
      "grad_norm": 1.3045182798093014,
      "kl": 0.09375,
      "learning_rate": 1.959752977092995e-06,
      "loss": 0.0037,
      "reward": 1.828125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2135
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 364.5,
      "epoch": 1.7088,
      "grad_norm": 5.048103835913532,
      "kl": 0.111328125,
      "learning_rate": 1.957708259228987e-06,
      "loss": 0.0045,
      "reward": 1.75,
      "reward_std": 0.19506090879440308,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 2136
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 386.265625,
      "epoch": 1.7096,
      "grad_norm": 0.8767383880884204,
      "kl": 0.08935546875,
      "learning_rate": 1.9556639219661983e-06,
      "loss": 0.0036,
      "reward": 1.671875,
      "reward_std": 0.189372718334198,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2137
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 350.6875,
      "epoch": 1.7104,
      "grad_norm": 0.6028009139225727,
      "kl": 0.10888671875,
      "learning_rate": 1.9536199667394217e-06,
      "loss": 0.0044,
      "reward": 1.859375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2138
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 398.984375,
      "epoch": 1.7112,
      "grad_norm": 0.41365095907955113,
      "kl": 0.0869140625,
      "learning_rate": 1.9515763949831852e-06,
      "loss": 0.0035,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2139
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 379.25,
      "epoch": 1.712,
      "grad_norm": 1.0507864788730774,
      "kl": 0.08837890625,
      "learning_rate": 1.9495332081317466e-06,
      "loss": 0.0035,
      "reward": 1.84375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2140
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 396.28125,
      "epoch": 1.7128,
      "grad_norm": 0.5168401126078777,
      "kl": 0.08056640625,
      "learning_rate": 1.947490407619092e-06,
      "loss": 0.0032,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2141
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 426.796875,
      "epoch": 1.7136,
      "grad_norm": 1.1847996150566735,
      "kl": 0.09033203125,
      "learning_rate": 1.945447994878937e-06,
      "loss": 0.0036,
      "reward": 1.890625,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 2142
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 391.34375,
      "epoch": 1.7144,
      "grad_norm": 0.5054062802849875,
      "kl": 0.314453125,
      "learning_rate": 1.9434059713447264e-06,
      "loss": 0.0126,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 2143
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 345.25,
      "epoch": 1.7151999999999998,
      "grad_norm": 0.42991081054431823,
      "kl": 0.0947265625,
      "learning_rate": 1.9413643384496315e-06,
      "loss": 0.0038,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2144
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 389.453125,
      "epoch": 1.716,
      "grad_norm": 2.6317180049845845,
      "kl": 0.83984375,
      "learning_rate": 1.9393230976265478e-06,
      "loss": 0.0336,
      "reward": 1.75,
      "reward_std": 0.342454731464386,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.96875,
      "step": 2145
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 451.390625,
      "epoch": 1.7168,
      "grad_norm": 0.6238490237819564,
      "kl": 0.09130859375,
      "learning_rate": 1.937282250308096e-06,
      "loss": 0.0036,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2146
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 395.265625,
      "epoch": 1.7176,
      "grad_norm": 0.6754300967166547,
      "kl": 0.0810546875,
      "learning_rate": 1.935241797926623e-06,
      "loss": 0.0032,
      "reward": 1.765625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 2147
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 407.265625,
      "epoch": 1.7184,
      "grad_norm": 0.37420052763490447,
      "kl": 0.07763671875,
      "learning_rate": 1.933201741914196e-06,
      "loss": 0.0031,
      "reward": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 2148
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 410.3125,
      "epoch": 1.7191999999999998,
      "grad_norm": 0.5465000812012062,
      "kl": 0.08154296875,
      "learning_rate": 1.931162083702606e-06,
      "loss": 0.0033,
      "reward": 1.59375,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 2149
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 420.71875,
      "epoch": 1.72,
      "grad_norm": 2.069261642593193,
      "kl": 0.875,
      "learning_rate": 1.9291228247233607e-06,
      "loss": 0.035,
      "reward": 1.75,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 2150
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 373.15625,
      "epoch": 1.7208,
      "grad_norm": 0.7448041846892124,
      "kl": 0.462890625,
      "learning_rate": 1.9270839664076937e-06,
      "loss": 0.0186,
      "reward": 1.6875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 2151
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 367.046875,
      "epoch": 1.7216,
      "grad_norm": 0.6583434051201349,
      "kl": 0.09765625,
      "learning_rate": 1.9250455101865526e-06,
      "loss": 0.0039,
      "reward": 1.765625,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 2152
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 376.9375,
      "epoch": 1.7224,
      "grad_norm": 0.4778357308685756,
      "kl": 0.62890625,
      "learning_rate": 1.9230074574906043e-06,
      "loss": 0.025,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 2153
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 388.734375,
      "epoch": 1.7231999999999998,
      "grad_norm": 0.6613729323890468,
      "kl": 0.1640625,
      "learning_rate": 1.920969809750234e-06,
      "loss": 0.0066,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 2154
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 359.734375,
      "epoch": 1.724,
      "grad_norm": 0.8442695124394836,
      "kl": 0.8203125,
      "learning_rate": 1.91893256839554e-06,
      "loss": 0.033,
      "reward": 1.609375,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.984375,
      "step": 2155
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 361.890625,
      "epoch": 1.7248,
      "grad_norm": 0.5860484477873993,
      "kl": 1.5234375,
      "learning_rate": 1.916895734856338e-06,
      "loss": 0.0606,
      "reward": 1.6875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 2156
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 426.65625,
      "epoch": 1.7256,
      "grad_norm": 0.527963775207647,
      "kl": 0.08056640625,
      "learning_rate": 1.9148593105621542e-06,
      "loss": 0.0032,
      "reward": 1.6875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 2157
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 375.28125,
      "epoch": 1.7264,
      "grad_norm": 0.6531439609473019,
      "kl": 1.0078125,
      "learning_rate": 1.9128232969422318e-06,
      "loss": 0.0402,
      "reward": 1.890625,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.984375,
      "step": 2158
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.625,
      "completion_length": 338.625,
      "epoch": 1.7271999999999998,
      "grad_norm": 0.07875094697090004,
      "kl": 0.0947265625,
      "learning_rate": 1.9107876954255217e-06,
      "loss": 0.0038,
      "reward": 1.375,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 2159
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 387.421875,
      "epoch": 1.728,
      "grad_norm": 0.9866165387525001,
      "kl": 0.68359375,
      "learning_rate": 1.908752507440689e-06,
      "loss": 0.0273,
      "reward": 1.890625,
      "reward_std": 0.15981829166412354,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.984375,
      "step": 2160
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 338.765625,
      "epoch": 1.7288000000000001,
      "grad_norm": 0.5917012212023924,
      "kl": 0.08984375,
      "learning_rate": 1.906717734416105e-06,
      "loss": 0.0036,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 2161
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 423.0625,
      "epoch": 1.7296,
      "grad_norm": 0.37594943811425857,
      "kl": 0.08154296875,
      "learning_rate": 1.9046833777798534e-06,
      "loss": 0.0033,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 2162
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 346.453125,
      "epoch": 1.7304,
      "grad_norm": 0.3544332730968081,
      "kl": 0.0849609375,
      "learning_rate": 1.9026494389597239e-06,
      "loss": 0.0034,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2163
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 360.234375,
      "epoch": 1.7311999999999999,
      "grad_norm": 1.0241005608809086,
      "kl": 1.3515625,
      "learning_rate": 1.9006159193832124e-06,
      "loss": 0.0538,
      "reward": 1.8125,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 2164
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 390.109375,
      "epoch": 1.732,
      "grad_norm": 0.10141318590330797,
      "kl": 0.0869140625,
      "learning_rate": 1.8985828204775206e-06,
      "loss": 0.0035,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 2165
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 400.109375,
      "epoch": 1.7328000000000001,
      "grad_norm": 0.5447422323142942,
      "kl": 0.07958984375,
      "learning_rate": 1.8965501436695578e-06,
      "loss": 0.0032,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 2166
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 379.25,
      "epoch": 1.7336,
      "grad_norm": 1.3390343253268027,
      "kl": 1.6640625,
      "learning_rate": 1.894517890385933e-06,
      "loss": 0.0668,
      "reward": 1.71875,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 2167
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 385.140625,
      "epoch": 1.7344,
      "grad_norm": 0.6592344074028343,
      "kl": 0.07763671875,
      "learning_rate": 1.8924860620529594e-06,
      "loss": 0.0031,
      "reward": 1.71875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 2168
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 367.09375,
      "epoch": 1.7351999999999999,
      "grad_norm": 0.6831807214742347,
      "kl": 0.349609375,
      "learning_rate": 1.8904546600966539e-06,
      "loss": 0.0139,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2169
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 363.9375,
      "epoch": 1.736,
      "grad_norm": 0.8033956249930624,
      "kl": 0.076171875,
      "learning_rate": 1.888423685942732e-06,
      "loss": 0.003,
      "reward": 1.828125,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2170
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 378.5,
      "epoch": 1.7368000000000001,
      "grad_norm": 7.109417305470336,
      "kl": 66.5,
      "learning_rate": 1.886393141016609e-06,
      "loss": 2.6712,
      "reward": 1.53125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.984375,
      "step": 2171
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 391.953125,
      "epoch": 1.7376,
      "grad_norm": 0.6615096403758822,
      "kl": 0.0849609375,
      "learning_rate": 1.8843630267434e-06,
      "loss": 0.0034,
      "reward": 1.609375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 2172
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 395.203125,
      "epoch": 1.7384,
      "grad_norm": 0.6867397743025585,
      "kl": 0.08984375,
      "learning_rate": 1.8823333445479175e-06,
      "loss": 0.0036,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 2173
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 378.640625,
      "epoch": 1.7391999999999999,
      "grad_norm": 0.3295383515321191,
      "kl": 0.703125,
      "learning_rate": 1.8803040958546708e-06,
      "loss": 0.0282,
      "reward": 1.828125,
      "reward_std": 0.09300297498703003,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 2174
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 412.25,
      "epoch": 1.74,
      "grad_norm": 0.5054531410755875,
      "kl": 0.0810546875,
      "learning_rate": 1.8782752820878636e-06,
      "loss": 0.0032,
      "reward": 1.921875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2175
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 370.4375,
      "epoch": 1.7408000000000001,
      "grad_norm": 0.07730153858880742,
      "kl": 0.080078125,
      "learning_rate": 1.8762469046713954e-06,
      "loss": 0.0032,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2176
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 358.6875,
      "epoch": 1.7416,
      "grad_norm": 0.06700469393968977,
      "kl": 0.0830078125,
      "learning_rate": 1.8742189650288617e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2177
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 364.015625,
      "epoch": 1.7424,
      "grad_norm": 0.7741486266428659,
      "kl": 0.61328125,
      "learning_rate": 1.872191464583547e-06,
      "loss": 0.0248,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 2178
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 425.65625,
      "epoch": 1.7431999999999999,
      "grad_norm": 0.5533172483910037,
      "kl": 0.91796875,
      "learning_rate": 1.8701644047584294e-06,
      "loss": 0.0368,
      "reward": 1.640625,
      "reward_std": 0.2109457403421402,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 2179
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 384.484375,
      "epoch": 1.744,
      "grad_norm": 0.7222343989628767,
      "kl": 0.11376953125,
      "learning_rate": 1.868137786976177e-06,
      "loss": 0.0046,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2180
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 379.484375,
      "epoch": 1.7448000000000001,
      "grad_norm": 0.7613424119483693,
      "kl": 0.087890625,
      "learning_rate": 1.8661116126591492e-06,
      "loss": 0.0035,
      "reward": 1.828125,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2181
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 360.078125,
      "epoch": 1.7456,
      "grad_norm": 0.364184606976644,
      "kl": 0.0771484375,
      "learning_rate": 1.8640858832293924e-06,
      "loss": 0.0031,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2182
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 423.640625,
      "epoch": 1.7464,
      "grad_norm": 0.06684451176827239,
      "kl": 0.07666015625,
      "learning_rate": 1.8620606001086423e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2183
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 415.921875,
      "epoch": 1.7471999999999999,
      "grad_norm": 1.9474665806526612,
      "kl": 0.078125,
      "learning_rate": 1.8600357647183188e-06,
      "loss": 0.0031,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2184
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 363.21875,
      "epoch": 1.748,
      "grad_norm": 0.8366932360932623,
      "kl": 0.078125,
      "learning_rate": 1.8580113784795306e-06,
      "loss": 0.0031,
      "reward": 1.671875,
      "reward_std": 0.24039676785469055,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2185
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 353.453125,
      "epoch": 1.7488000000000001,
      "grad_norm": 0.07387785973137999,
      "kl": 0.0810546875,
      "learning_rate": 1.8559874428130708e-06,
      "loss": 0.0032,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2186
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 389.6875,
      "epoch": 1.7496,
      "grad_norm": 0.3920833617687332,
      "kl": 0.0771484375,
      "learning_rate": 1.8539639591394131e-06,
      "loss": 0.0031,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 2187
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 369.765625,
      "epoch": 1.7504,
      "grad_norm": 0.6329375099735762,
      "kl": 0.08740234375,
      "learning_rate": 1.8519409288787182e-06,
      "loss": 0.0035,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2188
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 319.765625,
      "epoch": 1.7511999999999999,
      "grad_norm": 0.7345984322249883,
      "kl": 0.08203125,
      "learning_rate": 1.8499183534508263e-06,
      "loss": 0.0033,
      "reward": 1.796875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2189
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 357.28125,
      "epoch": 1.752,
      "grad_norm": 0.33114526841709796,
      "kl": 0.09423828125,
      "learning_rate": 1.8478962342752584e-06,
      "loss": 0.0038,
      "reward": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 2190
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 442.625,
      "epoch": 1.7528000000000001,
      "grad_norm": 0.35950129781701856,
      "kl": 0.197265625,
      "learning_rate": 1.8458745727712142e-06,
      "loss": 0.0079,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 2191
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 385.09375,
      "epoch": 1.7536,
      "grad_norm": 0.638379994349957,
      "kl": 0.1337890625,
      "learning_rate": 1.8438533703575757e-06,
      "loss": 0.0053,
      "reward": 1.5,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 2192
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 381.984375,
      "epoch": 1.7544,
      "grad_norm": 0.656625662268667,
      "kl": 0.1787109375,
      "learning_rate": 1.8418326284528997e-06,
      "loss": 0.0071,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 2193
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 398.640625,
      "epoch": 1.7551999999999999,
      "grad_norm": 0.77059664015117,
      "kl": 0.0859375,
      "learning_rate": 1.8398123484754204e-06,
      "loss": 0.0034,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2194
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 356.890625,
      "epoch": 1.756,
      "grad_norm": 0.6530003369178063,
      "kl": 0.087890625,
      "learning_rate": 1.8377925318430478e-06,
      "loss": 0.0035,
      "reward": 1.71875,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 2195
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 397.796875,
      "epoch": 1.7568000000000001,
      "grad_norm": 0.7639542174461229,
      "kl": 0.08349609375,
      "learning_rate": 1.8357731799733686e-06,
      "loss": 0.0033,
      "reward": 1.84375,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2196
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 346.46875,
      "epoch": 1.7576,
      "grad_norm": 0.8352966618780917,
      "kl": 0.341796875,
      "learning_rate": 1.8337542942836406e-06,
      "loss": 0.0137,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 2197
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 379.734375,
      "epoch": 1.7584,
      "grad_norm": 0.4202004345708123,
      "kl": 0.08837890625,
      "learning_rate": 1.8317358761907945e-06,
      "loss": 0.0035,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2198
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 352.59375,
      "epoch": 1.7591999999999999,
      "grad_norm": 1.5169317862327436,
      "kl": 0.11474609375,
      "learning_rate": 1.8297179271114345e-06,
      "loss": 0.0046,
      "reward": 1.515625,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 2199
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 372.078125,
      "epoch": 1.76,
      "grad_norm": 0.5583614560311482,
      "kl": 0.0810546875,
      "learning_rate": 1.827700448461836e-06,
      "loss": 0.0032,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2200
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 338.21875,
      "epoch": 1.7608000000000001,
      "grad_norm": 0.09188466516776025,
      "kl": 0.0888671875,
      "learning_rate": 1.8256834416579423e-06,
      "loss": 0.0036,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2201
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 415.96875,
      "epoch": 1.7616,
      "grad_norm": 1.3590539546953042,
      "kl": 0.0830078125,
      "learning_rate": 1.8236669081153657e-06,
      "loss": 0.0033,
      "reward": 1.640625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 2202
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 450.640625,
      "epoch": 1.7624,
      "grad_norm": 0.25075375774385944,
      "kl": 0.07666015625,
      "learning_rate": 1.8216508492493887e-06,
      "loss": 0.0031,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 2203
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 417.84375,
      "epoch": 1.7631999999999999,
      "grad_norm": 0.5709628925941539,
      "kl": 0.07177734375,
      "learning_rate": 1.8196352664749578e-06,
      "loss": 0.0029,
      "reward": 1.65625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 2204
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 412.390625,
      "epoch": 1.764,
      "grad_norm": 0.6348322494182044,
      "kl": 0.0849609375,
      "learning_rate": 1.8176201612066874e-06,
      "loss": 0.0034,
      "reward": 1.59375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 2205
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 376.890625,
      "epoch": 1.7648000000000001,
      "grad_norm": 0.4552307632423137,
      "kl": 0.1455078125,
      "learning_rate": 1.8156055348588548e-06,
      "loss": 0.0058,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2206
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 424.484375,
      "epoch": 1.7656,
      "grad_norm": 0.49988439873976975,
      "kl": 0.08447265625,
      "learning_rate": 1.8135913888454034e-06,
      "loss": 0.0034,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2207
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 425.640625,
      "epoch": 1.7664,
      "grad_norm": 0.726632658712714,
      "kl": 0.0771484375,
      "learning_rate": 1.8115777245799383e-06,
      "loss": 0.0031,
      "reward": 1.796875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2208
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 382.53125,
      "epoch": 1.7671999999999999,
      "grad_norm": 0.8967043761757607,
      "kl": 0.2353515625,
      "learning_rate": 1.8095645434757261e-06,
      "loss": 0.0094,
      "reward": 1.78125,
      "reward_std": 0.15769661962985992,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 2209
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 431.984375,
      "epoch": 1.768,
      "grad_norm": 0.3909682790220696,
      "kl": 0.078125,
      "learning_rate": 1.8075518469456944e-06,
      "loss": 0.0031,
      "reward": 1.9375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 2210
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.625,
      "completion_length": 427.5,
      "epoch": 1.7688000000000001,
      "grad_norm": 0.3092307066350754,
      "kl": 0.08837890625,
      "learning_rate": 1.8055396364024318e-06,
      "loss": 0.0035,
      "reward": 1.265625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 1.0,
      "step": 2211
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 403.359375,
      "epoch": 1.7696,
      "grad_norm": 0.06132860723139154,
      "kl": 0.0810546875,
      "learning_rate": 1.803527913258186e-06,
      "loss": 0.0033,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2212
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 413.421875,
      "epoch": 1.7704,
      "grad_norm": 1.3164007009289929,
      "kl": 0.09375,
      "learning_rate": 1.8015166789248606e-06,
      "loss": 0.0037,
      "reward": 1.6875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 2213
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 391.28125,
      "epoch": 1.7711999999999999,
      "grad_norm": 0.6314313796956886,
      "kl": 0.083984375,
      "learning_rate": 1.7995059348140165e-06,
      "loss": 0.0034,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 2214
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 427.28125,
      "epoch": 1.772,
      "grad_norm": 0.7757636034561768,
      "kl": 0.2314453125,
      "learning_rate": 1.7974956823368728e-06,
      "loss": 0.0093,
      "reward": 1.609375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 2215
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 425.75,
      "epoch": 1.7728000000000002,
      "grad_norm": 0.07203304999848316,
      "kl": 0.07421875,
      "learning_rate": 1.7954859229043017e-06,
      "loss": 0.003,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 2216
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 424.40625,
      "epoch": 1.7736,
      "grad_norm": 0.6206520422132313,
      "kl": 0.0908203125,
      "learning_rate": 1.7934766579268292e-06,
      "loss": 0.0036,
      "reward": 1.609375,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 2217
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 386.59375,
      "epoch": 1.7744,
      "grad_norm": 0.5964038193468935,
      "kl": 0.09375,
      "learning_rate": 1.7914678888146347e-06,
      "loss": 0.0038,
      "reward": 1.65625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 2218
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 348.359375,
      "epoch": 1.7752,
      "grad_norm": 0.8000847788397882,
      "kl": 0.08837890625,
      "learning_rate": 1.7894596169775514e-06,
      "loss": 0.0035,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2219
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 442.9375,
      "epoch": 1.776,
      "grad_norm": 0.05735166309966034,
      "kl": 0.0830078125,
      "learning_rate": 1.7874518438250598e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2220
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 396.3125,
      "epoch": 1.7768000000000002,
      "grad_norm": 0.377393691190953,
      "kl": 0.09619140625,
      "learning_rate": 1.785444570766293e-06,
      "loss": 0.0038,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 2221
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.5,
      "completion_length": 472.71875,
      "epoch": 1.7776,
      "grad_norm": 0.5801925247004116,
      "kl": 0.08251953125,
      "learning_rate": 1.7834377992100332e-06,
      "loss": 0.0033,
      "reward": 1.328125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 2222
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 399.140625,
      "epoch": 1.7784,
      "grad_norm": 0.9329010475938972,
      "kl": 0.33203125,
      "learning_rate": 1.7814315305647095e-06,
      "loss": 0.0133,
      "reward": 1.71875,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 2223
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 385.734375,
      "epoch": 1.7792,
      "grad_norm": 1.8708027844615398,
      "kl": 0.185546875,
      "learning_rate": 1.779425766238398e-06,
      "loss": 0.0074,
      "reward": 1.609375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 2224
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 394.125,
      "epoch": 1.78,
      "grad_norm": 0.498139251920171,
      "kl": 0.28515625,
      "learning_rate": 1.7774205076388207e-06,
      "loss": 0.0114,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2225
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 401.25,
      "epoch": 1.7808000000000002,
      "grad_norm": 0.5619444912439748,
      "kl": 0.08349609375,
      "learning_rate": 1.7754157561733476e-06,
      "loss": 0.0033,
      "reward": 1.84375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2226
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 404.3125,
      "epoch": 1.7816,
      "grad_norm": 0.4210320302566394,
      "kl": 0.083984375,
      "learning_rate": 1.7734115132489887e-06,
      "loss": 0.0034,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2227
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 422.375,
      "epoch": 1.7824,
      "grad_norm": 0.7367942735127898,
      "kl": 0.07421875,
      "learning_rate": 1.7714077802723994e-06,
      "loss": 0.003,
      "reward": 1.671875,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2228
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 452.296875,
      "epoch": 1.7832,
      "grad_norm": 0.9890232405149658,
      "kl": 0.0888671875,
      "learning_rate": 1.7694045586498754e-06,
      "loss": 0.0035,
      "reward": 1.65625,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 2229
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 377.46875,
      "epoch": 1.784,
      "grad_norm": 0.6324103271657705,
      "kl": 0.1689453125,
      "learning_rate": 1.7674018497873568e-06,
      "loss": 0.0067,
      "reward": 1.921875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2230
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 405.5625,
      "epoch": 1.7848000000000002,
      "grad_norm": 3.941341814515546,
      "kl": 0.08984375,
      "learning_rate": 1.7653996550904208e-06,
      "loss": 0.0036,
      "reward": 1.59375,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 2231
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 378.21875,
      "epoch": 1.7856,
      "grad_norm": 1.0721577397180355,
      "kl": 0.08544921875,
      "learning_rate": 1.7633979759642844e-06,
      "loss": 0.0034,
      "reward": 1.875,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2232
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 450.75,
      "epoch": 1.7864,
      "grad_norm": 0.4838969957345881,
      "kl": 0.0888671875,
      "learning_rate": 1.7613968138138027e-06,
      "loss": 0.0036,
      "reward": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 2233
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 414.46875,
      "epoch": 1.7872,
      "grad_norm": 0.35336711014719263,
      "kl": 0.0751953125,
      "learning_rate": 1.7593961700434692e-06,
      "loss": 0.003,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2234
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 461.8125,
      "epoch": 1.788,
      "grad_norm": 0.48168024944088667,
      "kl": 0.08349609375,
      "learning_rate": 1.7573960460574133e-06,
      "loss": 0.0033,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2235
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 408.453125,
      "epoch": 1.7888,
      "grad_norm": 0.611225545331788,
      "kl": 0.56640625,
      "learning_rate": 1.7553964432593976e-06,
      "loss": 0.0227,
      "reward": 1.640625,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 2236
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 414.359375,
      "epoch": 1.7896,
      "grad_norm": 0.9386424953791263,
      "kl": 0.302734375,
      "learning_rate": 1.75339736305282e-06,
      "loss": 0.0121,
      "reward": 1.625,
      "reward_std": 0.1243029236793518,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.96875,
      "step": 2237
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 460.40625,
      "epoch": 1.7904,
      "grad_norm": 1.1941673294724382,
      "kl": 0.349609375,
      "learning_rate": 1.7513988068407145e-06,
      "loss": 0.0141,
      "reward": 1.6875,
      "reward_std": 0.23827511072158813,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 2238
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 406.109375,
      "epoch": 1.7912,
      "grad_norm": 0.8289088427251986,
      "kl": 0.458984375,
      "learning_rate": 1.7494007760257428e-06,
      "loss": 0.0183,
      "reward": 1.625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 2239
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 411.828125,
      "epoch": 1.792,
      "grad_norm": 0.18174947692607274,
      "kl": 0.08251953125,
      "learning_rate": 1.7474032720101991e-06,
      "loss": 0.0033,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 2240
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 410.03125,
      "epoch": 1.7928,
      "grad_norm": 0.057623689715676836,
      "kl": 0.0830078125,
      "learning_rate": 1.7454062961960102e-06,
      "loss": 0.0033,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2241
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 415.84375,
      "epoch": 1.7936,
      "grad_norm": 0.47957047822833476,
      "kl": 0.08837890625,
      "learning_rate": 1.7434098499847308e-06,
      "loss": 0.0035,
      "reward": 1.703125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 2242
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 411.78125,
      "epoch": 1.7944,
      "grad_norm": 1.4510650561256406,
      "kl": 1.3359375,
      "learning_rate": 1.7414139347775423e-06,
      "loss": 0.0534,
      "reward": 1.71875,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.953125,
      "step": 2243
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 411.53125,
      "epoch": 1.7952,
      "grad_norm": 0.7926264380225096,
      "kl": 0.1015625,
      "learning_rate": 1.7394185519752546e-06,
      "loss": 0.0041,
      "reward": 1.78125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 2244
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 432.640625,
      "epoch": 1.796,
      "grad_norm": 0.45536677576866846,
      "kl": 0.19921875,
      "learning_rate": 1.7374237029783064e-06,
      "loss": 0.008,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 2245
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 441.421875,
      "epoch": 1.7968,
      "grad_norm": 0.6059867585897916,
      "kl": 0.5859375,
      "learning_rate": 1.7354293891867582e-06,
      "loss": 0.0235,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 2246
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 389.59375,
      "epoch": 1.7976,
      "grad_norm": 0.41467690318077355,
      "kl": 0.0869140625,
      "learning_rate": 1.7334356120002956e-06,
      "loss": 0.0035,
      "reward": 1.65625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 2247
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 454.015625,
      "epoch": 1.7984,
      "grad_norm": 0.767967960568488,
      "kl": 0.07275390625,
      "learning_rate": 1.7314423728182283e-06,
      "loss": 0.0029,
      "reward": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2248
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 452.609375,
      "epoch": 1.7992,
      "grad_norm": 0.37091586756919365,
      "kl": 0.68359375,
      "learning_rate": 1.7294496730394897e-06,
      "loss": 0.0274,
      "reward": 1.75,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 2249
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 411.921875,
      "epoch": 1.8,
      "grad_norm": 1.0544059277895323,
      "kl": 0.08251953125,
      "learning_rate": 1.7274575140626318e-06,
      "loss": 0.0033,
      "reward": 1.46875,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 2250
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 415.109375,
      "epoch": 1.8008,
      "grad_norm": 0.05339237676383853,
      "kl": 0.06982421875,
      "learning_rate": 1.7254658972858293e-06,
      "loss": 0.0028,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 2251
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 466.4375,
      "epoch": 1.8016,
      "grad_norm": 0.4810915562817627,
      "kl": 0.71875,
      "learning_rate": 1.7234748241068742e-06,
      "loss": 0.0289,
      "reward": 1.921875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 2252
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 398.375,
      "epoch": 1.8024,
      "grad_norm": 0.08754555373606211,
      "kl": 0.0771484375,
      "learning_rate": 1.7214842959231796e-06,
      "loss": 0.0031,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2253
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 433.921875,
      "epoch": 1.8032,
      "grad_norm": 1.353657433311371,
      "kl": 1.0390625,
      "learning_rate": 1.719494314131775e-06,
      "loss": 0.0416,
      "reward": 1.796875,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 2254
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 386.1875,
      "epoch": 1.804,
      "grad_norm": 1.7745382663360545,
      "kl": 0.09228515625,
      "learning_rate": 1.7175048801293042e-06,
      "loss": 0.0037,
      "reward": 1.84375,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2255
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 400.8125,
      "epoch": 1.8048,
      "grad_norm": 0.8812652450707908,
      "kl": 0.09814453125,
      "learning_rate": 1.7155159953120315e-06,
      "loss": 0.0039,
      "reward": 1.65625,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 2256
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 380.875,
      "epoch": 1.8056,
      "grad_norm": 0.8448951799668749,
      "kl": 0.0869140625,
      "learning_rate": 1.7135276610758309e-06,
      "loss": 0.0035,
      "reward": 1.546875,
      "reward_std": 0.2472364753484726,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 2257
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 382.640625,
      "epoch": 1.8064,
      "grad_norm": 0.7179314122332302,
      "kl": 0.078125,
      "learning_rate": 1.7115398788161923e-06,
      "loss": 0.0031,
      "reward": 1.640625,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 2258
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 364.296875,
      "epoch": 1.8072,
      "grad_norm": 0.7423929414923691,
      "kl": 0.2001953125,
      "learning_rate": 1.7095526499282172e-06,
      "loss": 0.008,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2259
    },
    {
      "all_correct": 1.0,
      "all_wrong": 0.0,
      "completion_length": 409.28125,
      "epoch": 1.808,
      "grad_norm": 0.07914383247635483,
      "kl": 0.07958984375,
      "learning_rate": 1.7075659758066207e-06,
      "loss": 0.0032,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 2260
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 381.15625,
      "epoch": 1.8088,
      "grad_norm": 2.0865411261662095,
      "kl": 0.08740234375,
      "learning_rate": 1.7055798578457267e-06,
      "loss": 0.0035,
      "reward": 1.53125,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 2261
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 430.578125,
      "epoch": 1.8096,
      "grad_norm": 0.5517740052570347,
      "kl": 0.06982421875,
      "learning_rate": 1.703594297439469e-06,
      "loss": 0.0028,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 2262
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 423.578125,
      "epoch": 1.8104,
      "grad_norm": 1.258041362856456,
      "kl": 0.6796875,
      "learning_rate": 1.7016092959813892e-06,
      "loss": 0.0273,
      "reward": 1.875,
      "reward_std": 0.2177756279706955,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.984375,
      "step": 2263
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 396.53125,
      "epoch": 1.8112,
      "grad_norm": 0.5529226969659027,
      "kl": 0.09521484375,
      "learning_rate": 1.6996248548646393e-06,
      "loss": 0.0038,
      "reward": 1.796875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2264
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 397.453125,
      "epoch": 1.812,
      "grad_norm": 0.5097453340811813,
      "kl": 0.0830078125,
      "learning_rate": 1.6976409754819767e-06,
      "loss": 0.0033,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2265
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 416.9375,
      "epoch": 1.8128,
      "grad_norm": 0.06305387379928601,
      "kl": 0.0771484375,
      "learning_rate": 1.6956576592257635e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2266
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 397.3125,
      "epoch": 1.8136,
      "grad_norm": 0.06838940090814126,
      "kl": 0.08251953125,
      "learning_rate": 1.6936749074879663e-06,
      "loss": 0.0033,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2267
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 419.75,
      "epoch": 1.8144,
      "grad_norm": 0.37466414719961527,
      "kl": 0.076171875,
      "learning_rate": 1.6916927216601593e-06,
      "loss": 0.0031,
      "reward": 1.578125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 2268
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 366.796875,
      "epoch": 1.8152,
      "grad_norm": 0.6063340028821798,
      "kl": 0.07861328125,
      "learning_rate": 1.6897111031335145e-06,
      "loss": 0.0031,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2269
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 428.421875,
      "epoch": 1.8159999999999998,
      "grad_norm": 0.6073598853765154,
      "kl": 0.515625,
      "learning_rate": 1.6877300532988095e-06,
      "loss": 0.0206,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 2270
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 388.390625,
      "epoch": 1.8168,
      "grad_norm": 2.323754122284559,
      "kl": 0.0888671875,
      "learning_rate": 1.6857495735464196e-06,
      "loss": 0.0035,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2271
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 404.109375,
      "epoch": 1.8176,
      "grad_norm": 0.6804114586495742,
      "kl": 0.1220703125,
      "learning_rate": 1.6837696652663244e-06,
      "loss": 0.0049,
      "reward": 1.578125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 2272
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 426.6875,
      "epoch": 1.8184,
      "grad_norm": 0.7756025874676311,
      "kl": 0.08984375,
      "learning_rate": 1.681790329848097e-06,
      "loss": 0.0036,
      "reward": 1.828125,
      "reward_std": 0.16887323558330536,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2273
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 427.734375,
      "epoch": 1.8192,
      "grad_norm": 0.4015389990989993,
      "kl": 0.29296875,
      "learning_rate": 1.6798115686809125e-06,
      "loss": 0.0117,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2274
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 410.203125,
      "epoch": 1.8199999999999998,
      "grad_norm": 0.06056705674607063,
      "kl": 0.07861328125,
      "learning_rate": 1.677833383153542e-06,
      "loss": 0.0031,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2275
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 427.5,
      "epoch": 1.8208,
      "grad_norm": 2.205816337885792,
      "kl": 0.07861328125,
      "learning_rate": 1.6758557746543518e-06,
      "loss": 0.0031,
      "reward": 1.859375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2276
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 386.59375,
      "epoch": 1.8216,
      "grad_norm": 1.095409944294399,
      "kl": 0.263671875,
      "learning_rate": 1.673878744571304e-06,
      "loss": 0.0106,
      "reward": 1.90625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.96875,
      "step": 2277
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 395.78125,
      "epoch": 1.8224,
      "grad_norm": 0.07041981281314577,
      "kl": 0.076171875,
      "learning_rate": 1.6719022942919527e-06,
      "loss": 0.003,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 2278
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 406.609375,
      "epoch": 1.8232,
      "grad_norm": 0.4342114951966203,
      "kl": 0.07958984375,
      "learning_rate": 1.6699264252034498e-06,
      "loss": 0.0032,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 2279
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 417.25,
      "epoch": 1.8239999999999998,
      "grad_norm": 0.5092638014697883,
      "kl": 0.244140625,
      "learning_rate": 1.6679511386925337e-06,
      "loss": 0.0098,
      "reward": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 2280
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 432.078125,
      "epoch": 1.8248,
      "grad_norm": 0.06874156878127212,
      "kl": 0.07275390625,
      "learning_rate": 1.6659764361455383e-06,
      "loss": 0.0029,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 2281
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 420.140625,
      "epoch": 1.8256000000000001,
      "grad_norm": 2.3332094846872873,
      "kl": 0.07470703125,
      "learning_rate": 1.6640023189483836e-06,
      "loss": 0.003,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 2282
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 419.3125,
      "epoch": 1.8264,
      "grad_norm": 0.3194931297089578,
      "kl": 0.09326171875,
      "learning_rate": 1.6620287884865831e-06,
      "loss": 0.0037,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2283
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 434.96875,
      "epoch": 1.8272,
      "grad_norm": 0.058001088032383526,
      "kl": 0.0830078125,
      "learning_rate": 1.6600558461452368e-06,
      "loss": 0.0033,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2284
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 400.21875,
      "epoch": 1.8279999999999998,
      "grad_norm": 1.1346504301460851,
      "kl": 0.66015625,
      "learning_rate": 1.65808349330903e-06,
      "loss": 0.0263,
      "reward": 1.546875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 2285
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 416.0,
      "epoch": 1.8288,
      "grad_norm": 0.7323502042973661,
      "kl": 0.09765625,
      "learning_rate": 1.656111731362236e-06,
      "loss": 0.0039,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2286
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 405.046875,
      "epoch": 1.8296000000000001,
      "grad_norm": 0.6425092179459982,
      "kl": 0.08203125,
      "learning_rate": 1.6541405616887138e-06,
      "loss": 0.0033,
      "reward": 1.890625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 2287
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 416.921875,
      "epoch": 1.8304,
      "grad_norm": 0.5851376635857483,
      "kl": 0.08984375,
      "learning_rate": 1.6521699856719065e-06,
      "loss": 0.0036,
      "reward": 1.65625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 2288
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 387.78125,
      "epoch": 1.8312,
      "grad_norm": 0.39461520559472857,
      "kl": 0.07763671875,
      "learning_rate": 1.650200004694839e-06,
      "loss": 0.0031,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 2289
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 456.046875,
      "epoch": 1.8319999999999999,
      "grad_norm": 0.5458663585800854,
      "kl": 0.0703125,
      "learning_rate": 1.6482306201401211e-06,
      "loss": 0.0028,
      "reward": 1.921875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 2290
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 457.375,
      "epoch": 1.8328,
      "grad_norm": 0.35289779031109975,
      "kl": 0.08056640625,
      "learning_rate": 1.6462618333899422e-06,
      "loss": 0.0032,
      "reward": 1.96875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 2291
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 401.78125,
      "epoch": 1.8336000000000001,
      "grad_norm": 0.8117770038248345,
      "kl": 0.25390625,
      "learning_rate": 1.6442936458260723e-06,
      "loss": 0.0101,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.984375,
      "step": 2292
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 478.5625,
      "epoch": 1.8344,
      "grad_norm": 0.6311079921720235,
      "kl": 0.07958984375,
      "learning_rate": 1.6423260588298608e-06,
      "loss": 0.0032,
      "reward": 1.671875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 2293
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 406.40625,
      "epoch": 1.8352,
      "grad_norm": 0.514535236545834,
      "kl": 0.080078125,
      "learning_rate": 1.6403590737822378e-06,
      "loss": 0.0032,
      "reward": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 2294
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 454.40625,
      "epoch": 1.8359999999999999,
      "grad_norm": 0.22239642077532396,
      "kl": 0.1943359375,
      "learning_rate": 1.6383926920637077e-06,
      "loss": 0.0078,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2295
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 451.1875,
      "epoch": 1.8368,
      "grad_norm": 0.6811230340729435,
      "kl": 0.28125,
      "learning_rate": 1.6364269150543533e-06,
      "loss": 0.0113,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.984375,
      "step": 2296
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 392.1875,
      "epoch": 1.8376000000000001,
      "grad_norm": 1.2234684755317045,
      "kl": 0.37890625,
      "learning_rate": 1.6344617441338311e-06,
      "loss": 0.0152,
      "reward": 1.546875,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 2297
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 436.875,
      "epoch": 1.8384,
      "grad_norm": 0.40356023234711014,
      "kl": 0.07861328125,
      "learning_rate": 1.6324971806813766e-06,
      "loss": 0.0031,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2298
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 393.1875,
      "epoch": 1.8392,
      "grad_norm": 1.0542946842458971,
      "kl": 0.083984375,
      "learning_rate": 1.6305332260757937e-06,
      "loss": 0.0034,
      "reward": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 2299
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 429.765625,
      "epoch": 1.8399999999999999,
      "grad_norm": 0.3947925761415726,
      "kl": 0.08203125,
      "learning_rate": 1.6285698816954626e-06,
      "loss": 0.0033,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2300
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 446.265625,
      "epoch": 1.8408,
      "grad_norm": 0.7470857421143552,
      "kl": 0.10302734375,
      "learning_rate": 1.6266071489183327e-06,
      "loss": 0.0041,
      "reward": 1.640625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 2301
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 417.390625,
      "epoch": 1.8416000000000001,
      "grad_norm": 0.6735528871930032,
      "kl": 0.083984375,
      "learning_rate": 1.6246450291219268e-06,
      "loss": 0.0034,
      "reward": 1.875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2302
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 444.84375,
      "epoch": 1.8424,
      "grad_norm": 0.5872276880370481,
      "kl": 0.09228515625,
      "learning_rate": 1.6226835236833356e-06,
      "loss": 0.0037,
      "reward": 1.90625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 2303
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 433.6875,
      "epoch": 1.8432,
      "grad_norm": 1.2067725226290258,
      "kl": 0.10107421875,
      "learning_rate": 1.620722633979219e-06,
      "loss": 0.004,
      "reward": 1.921875,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 2304
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 460.359375,
      "epoch": 1.8439999999999999,
      "grad_norm": 0.7093879480128268,
      "kl": 0.08544921875,
      "learning_rate": 1.6187623613858038e-06,
      "loss": 0.0034,
      "reward": 1.859375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2305
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 435.578125,
      "epoch": 1.8448,
      "grad_norm": 0.7916096166065976,
      "kl": 0.400390625,
      "learning_rate": 1.6168027072788868e-06,
      "loss": 0.016,
      "reward": 1.8125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2306
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 492.09375,
      "epoch": 1.8456000000000001,
      "grad_norm": 0.6229346311295197,
      "kl": 0.07958984375,
      "learning_rate": 1.6148436730338279e-06,
      "loss": 0.0032,
      "reward": 1.9375,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 2307
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 440.515625,
      "epoch": 1.8464,
      "grad_norm": 1.2172980612288264,
      "kl": 0.0830078125,
      "learning_rate": 1.6128852600255518e-06,
      "loss": 0.0033,
      "reward": 1.8125,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 2308
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 489.796875,
      "epoch": 1.8472,
      "grad_norm": 0.8909217088840472,
      "kl": 1.3359375,
      "learning_rate": 1.6109274696285496e-06,
      "loss": 0.0535,
      "reward": 1.9375,
      "reward_std": 0.13719715178012848,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.96875,
      "step": 2309
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 463.203125,
      "epoch": 1.8479999999999999,
      "grad_norm": 0.5301575360099662,
      "kl": 0.09423828125,
      "learning_rate": 1.6089703032168736e-06,
      "loss": 0.0038,
      "reward": 1.875,
      "reward_std": 0.1243029236793518,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.984375,
      "step": 2310
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 479.765625,
      "epoch": 1.8488,
      "grad_norm": 0.44364619351434914,
      "kl": 0.81640625,
      "learning_rate": 1.6070137621641382e-06,
      "loss": 0.0327,
      "reward": 1.765625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 2311
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 457.28125,
      "epoch": 1.8496000000000001,
      "grad_norm": 0.47166131642306697,
      "kl": 0.08837890625,
      "learning_rate": 1.6050578478435184e-06,
      "loss": 0.0035,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2312
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 474.921875,
      "epoch": 1.8504,
      "grad_norm": 0.6161836323243681,
      "kl": 0.2138671875,
      "learning_rate": 1.6031025616277512e-06,
      "loss": 0.0086,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 2313
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 490.984375,
      "epoch": 1.8512,
      "grad_norm": 0.6143641531177049,
      "kl": 0.087890625,
      "learning_rate": 1.6011479048891323e-06,
      "loss": 0.0035,
      "reward": 1.765625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 2314
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 442.25,
      "epoch": 1.8519999999999999,
      "grad_norm": 1.3468600760031937,
      "kl": 0.56640625,
      "learning_rate": 1.5991938789995138e-06,
      "loss": 0.0227,
      "reward": 1.71875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 2315
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 461.890625,
      "epoch": 1.8528,
      "grad_norm": 0.04483782424172691,
      "kl": 0.078125,
      "learning_rate": 1.5972404853303061e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2316
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 396.359375,
      "epoch": 1.8536000000000001,
      "grad_norm": 0.8712801056418593,
      "kl": 0.1123046875,
      "learning_rate": 1.595287725252478e-06,
      "loss": 0.0045,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.984375,
      "step": 2317
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 420.953125,
      "epoch": 1.8544,
      "grad_norm": 1.4722712412340406,
      "kl": 1.5390625,
      "learning_rate": 1.5933356001365502e-06,
      "loss": 0.0616,
      "reward": 1.671875,
      "reward_std": 0.35816800594329834,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.9375,
      "step": 2318
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 503.453125,
      "epoch": 1.8552,
      "grad_norm": 1.788304634704619,
      "kl": 0.0869140625,
      "learning_rate": 1.591384111352599e-06,
      "loss": 0.0035,
      "reward": 1.578125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 2319
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 452.640625,
      "epoch": 1.8559999999999999,
      "grad_norm": 0.9205119308247197,
      "kl": 0.1474609375,
      "learning_rate": 1.5894332602702545e-06,
      "loss": 0.0059,
      "reward": 1.71875,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 2320
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 442.828125,
      "epoch": 1.8568,
      "grad_norm": 0.4724895151392997,
      "kl": 0.12255859375,
      "learning_rate": 1.5874830482587003e-06,
      "loss": 0.0049,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 2321
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 460.4375,
      "epoch": 1.8576000000000001,
      "grad_norm": 0.7016824771040632,
      "kl": 0.88671875,
      "learning_rate": 1.585533476686669e-06,
      "loss": 0.0355,
      "reward": 1.734375,
      "reward_std": 0.2109457403421402,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 2322
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 474.8125,
      "epoch": 1.8584,
      "grad_norm": 0.7669713951662993,
      "kl": 0.111328125,
      "learning_rate": 1.5835845469224447e-06,
      "loss": 0.0045,
      "reward": 1.875,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.984375,
      "step": 2323
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 455.90625,
      "epoch": 1.8592,
      "grad_norm": 0.5967170462706515,
      "kl": 0.0810546875,
      "learning_rate": 1.5816362603338632e-06,
      "loss": 0.0032,
      "reward": 1.734375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 2324
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 502.0,
      "epoch": 1.8599999999999999,
      "grad_norm": 0.4899720910756078,
      "kl": 0.458984375,
      "learning_rate": 1.5796886182883053e-06,
      "loss": 0.0183,
      "reward": 1.578125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 2325
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 501.140625,
      "epoch": 1.8608,
      "grad_norm": 0.8680875979230859,
      "kl": 1.796875,
      "learning_rate": 1.577741622152702e-06,
      "loss": 0.0718,
      "reward": 1.8125,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.953125,
      "step": 2326
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 473.359375,
      "epoch": 1.8616000000000001,
      "grad_norm": 0.6348149800883833,
      "kl": 1.703125,
      "learning_rate": 1.5757952732935288e-06,
      "loss": 0.0684,
      "reward": 1.6875,
      "reward_std": 0.2619796097278595,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.953125,
      "step": 2327
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.5,
      "completion_length": 473.8125,
      "epoch": 1.8624,
      "grad_norm": 0.5647115361224099,
      "kl": 0.2578125,
      "learning_rate": 1.5738495730768104e-06,
      "loss": 0.0103,
      "reward": 1.484375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.984375,
      "step": 2328
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 437.671875,
      "epoch": 1.8632,
      "grad_norm": 1.6814611945657367,
      "kl": 0.181640625,
      "learning_rate": 1.5719045228681127e-06,
      "loss": 0.0072,
      "reward": 1.671875,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 2329
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 405.640625,
      "epoch": 1.8639999999999999,
      "grad_norm": 0.623347105111847,
      "kl": 0.6953125,
      "learning_rate": 1.5699601240325474e-06,
      "loss": 0.0279,
      "reward": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 2330
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 442.453125,
      "epoch": 1.8648,
      "grad_norm": 1.3513951234674806,
      "kl": 1.6328125,
      "learning_rate": 1.5680163779347668e-06,
      "loss": 0.0655,
      "reward": 1.59375,
      "reward_std": 0.2744373679161072,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.90625,
      "step": 2331
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 410.015625,
      "epoch": 1.8656000000000001,
      "grad_norm": 0.6530723952699748,
      "kl": 0.21875,
      "learning_rate": 1.5660732859389687e-06,
      "loss": 0.0087,
      "reward": 1.828125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 2332
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 523.140625,
      "epoch": 1.8664,
      "grad_norm": 1.0787077908420544,
      "kl": 0.2333984375,
      "learning_rate": 1.5641308494088903e-06,
      "loss": 0.0093,
      "reward": 1.828125,
      "reward_std": 0.23925507068634033,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 2333
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 424.53125,
      "epoch": 1.8672,
      "grad_norm": 1.0625795033201053,
      "kl": 0.9140625,
      "learning_rate": 1.5621890697078069e-06,
      "loss": 0.0367,
      "reward": 1.796875,
      "reward_std": 0.2773849368095398,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 2334
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 518.46875,
      "epoch": 1.8679999999999999,
      "grad_norm": 2.898753556284234,
      "kl": 0.341796875,
      "learning_rate": 1.5602479481985333e-06,
      "loss": 0.0137,
      "reward": 1.9375,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.953125,
      "step": 2335
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 375.609375,
      "epoch": 1.8688,
      "grad_norm": 1.1551349822481982,
      "kl": 1.1328125,
      "learning_rate": 1.5583074862434254e-06,
      "loss": 0.0455,
      "reward": 1.46875,
      "reward_std": 0.26409149169921875,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.984375,
      "step": 2336
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 452.53125,
      "epoch": 1.8696000000000002,
      "grad_norm": 1.524298777949189,
      "kl": 1.5859375,
      "learning_rate": 1.5563676852043738e-06,
      "loss": 0.0635,
      "reward": 1.5625,
      "reward_std": 0.3122667670249939,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.90625,
      "step": 2337
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 424.265625,
      "epoch": 1.8704,
      "grad_norm": 1.9314722579043813,
      "kl": 2.0,
      "learning_rate": 1.5544285464428044e-06,
      "loss": 0.0802,
      "reward": 1.703125,
      "reward_std": 0.45481547713279724,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.9375,
      "step": 2338
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 392.40625,
      "epoch": 1.8712,
      "grad_norm": 0.7790863024565302,
      "kl": 0.4375,
      "learning_rate": 1.55249007131968e-06,
      "loss": 0.0175,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2339
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 459.578125,
      "epoch": 1.8719999999999999,
      "grad_norm": 0.302611999932875,
      "kl": 0.10498046875,
      "learning_rate": 1.5505522611954977e-06,
      "loss": 0.0042,
      "reward": 1.890625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 2340
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 472.75,
      "epoch": 1.8728,
      "grad_norm": 1.0436752059242749,
      "kl": 0.341796875,
      "learning_rate": 1.548615117430286e-06,
      "loss": 0.0136,
      "reward": 1.734375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 2341
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 484.453125,
      "epoch": 1.8736000000000002,
      "grad_norm": 0.2955088923753632,
      "kl": 0.6796875,
      "learning_rate": 1.5466786413836077e-06,
      "loss": 0.0272,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.984375,
      "step": 2342
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 438.140625,
      "epoch": 1.8744,
      "grad_norm": 0.6671132830263261,
      "kl": 0.0751953125,
      "learning_rate": 1.5447428344145565e-06,
      "loss": 0.003,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2343
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 418.65625,
      "epoch": 1.8752,
      "grad_norm": 0.12573231959996634,
      "kl": 0.103515625,
      "learning_rate": 1.5428076978817564e-06,
      "loss": 0.0041,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2344
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 403.0625,
      "epoch": 1.876,
      "grad_norm": 1.0971006880127447,
      "kl": 0.42578125,
      "learning_rate": 1.5408732331433596e-06,
      "loss": 0.0171,
      "reward": 1.5625,
      "reward_std": 0.2948937714099884,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.96875,
      "step": 2345
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 432.84375,
      "epoch": 1.8768,
      "grad_norm": 1.2217093029811685,
      "kl": 0.314453125,
      "learning_rate": 1.538939441557048e-06,
      "loss": 0.0126,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.96875,
      "step": 2346
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 403.015625,
      "epoch": 1.8776000000000002,
      "grad_norm": 0.6836710614113861,
      "kl": 1.0,
      "learning_rate": 1.5370063244800326e-06,
      "loss": 0.0401,
      "reward": 1.859375,
      "reward_std": 0.19044627249240875,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 2347
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.375,
      "completion_length": 466.46875,
      "epoch": 1.8784,
      "grad_norm": 0.2853069038669468,
      "kl": 0.1396484375,
      "learning_rate": 1.5350738832690479e-06,
      "loss": 0.0056,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 2348
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 464.375,
      "epoch": 1.8792,
      "grad_norm": 1.0723235378029217,
      "kl": 0.9140625,
      "learning_rate": 1.5331421192803565e-06,
      "loss": 0.0367,
      "reward": 1.765625,
      "reward_std": 0.19044627249240875,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 2349
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 418.28125,
      "epoch": 1.88,
      "grad_norm": 1.3252120392631617,
      "kl": 0.7578125,
      "learning_rate": 1.5312110338697427e-06,
      "loss": 0.0304,
      "reward": 1.78125,
      "reward_std": 0.24608497321605682,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 2350
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 441.625,
      "epoch": 1.8808,
      "grad_norm": 1.1791143111050741,
      "kl": 1.59375,
      "learning_rate": 1.5292806283925192e-06,
      "loss": 0.0637,
      "reward": 1.671875,
      "reward_std": 0.26076778769493103,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.9375,
      "step": 2351
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 438.8125,
      "epoch": 1.8816000000000002,
      "grad_norm": 1.3254748259748228,
      "kl": 1.8671875,
      "learning_rate": 1.5273509042035172e-06,
      "loss": 0.0746,
      "reward": 1.90625,
      "reward_std": 0.1828794628381729,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.96875,
      "step": 2352
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 444.234375,
      "epoch": 1.8824,
      "grad_norm": 0.5973355080036852,
      "kl": 0.60546875,
      "learning_rate": 1.5254218626570927e-06,
      "loss": 0.0243,
      "reward": 1.59375,
      "reward_std": 0.13719715178012848,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.96875,
      "step": 2353
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 392.953125,
      "epoch": 1.8832,
      "grad_norm": 1.1410039809378982,
      "kl": 0.83203125,
      "learning_rate": 1.5234935051071193e-06,
      "loss": 0.0333,
      "reward": 1.75,
      "reward_std": 0.19727617502212524,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 2354
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 385.0,
      "epoch": 1.884,
      "grad_norm": 1.1471713788274727,
      "kl": 0.14453125,
      "learning_rate": 1.521565832906994e-06,
      "loss": 0.0058,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 1.0,
      "step": 2355
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 410.765625,
      "epoch": 1.8848,
      "grad_norm": 0.5609605962424036,
      "kl": 0.0771484375,
      "learning_rate": 1.519638847409632e-06,
      "loss": 0.0031,
      "reward": 1.8125,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 2356
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 407.34375,
      "epoch": 1.8856000000000002,
      "grad_norm": 0.980539815965368,
      "kl": 0.06982421875,
      "learning_rate": 1.5177125499674639e-06,
      "loss": 0.0028,
      "reward": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 2357
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 395.375,
      "epoch": 1.8864,
      "grad_norm": 1.154891923772219,
      "kl": 0.3828125,
      "learning_rate": 1.515786941932441e-06,
      "loss": 0.0153,
      "reward": 1.78125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.984375,
      "step": 2358
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 409.03125,
      "epoch": 1.8872,
      "grad_norm": 0.9968607771230172,
      "kl": 0.52734375,
      "learning_rate": 1.5138620246560295e-06,
      "loss": 0.0211,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 2359
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 401.796875,
      "epoch": 1.888,
      "grad_norm": 0.5690985149571485,
      "kl": 0.0859375,
      "learning_rate": 1.5119377994892095e-06,
      "loss": 0.0034,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 2360
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 457.046875,
      "epoch": 1.8888,
      "grad_norm": 0.8228440234192015,
      "kl": 0.06884765625,
      "learning_rate": 1.5100142677824752e-06,
      "loss": 0.0027,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 2361
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 369.3125,
      "epoch": 1.8896,
      "grad_norm": 0.07799185674390813,
      "kl": 0.0771484375,
      "learning_rate": 1.5080914308858375e-06,
      "loss": 0.0031,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2362
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 393.203125,
      "epoch": 1.8904,
      "grad_norm": 1.940053735681463,
      "kl": 0.47265625,
      "learning_rate": 1.5061692901488161e-06,
      "loss": 0.019,
      "reward": 1.796875,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 2363
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 383.296875,
      "epoch": 1.8912,
      "grad_norm": 0.6430789106343778,
      "kl": 0.208984375,
      "learning_rate": 1.5042478469204437e-06,
      "loss": 0.0084,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 2364
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 492.9375,
      "epoch": 1.892,
      "grad_norm": 1.8130245744964804,
      "kl": 0.69921875,
      "learning_rate": 1.502327102549262e-06,
      "loss": 0.0281,
      "reward": 1.609375,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 2365
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 382.96875,
      "epoch": 1.8928,
      "grad_norm": 0.17595814693903425,
      "kl": 0.1328125,
      "learning_rate": 1.5004070583833252e-06,
      "loss": 0.0053,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2366
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 445.21875,
      "epoch": 1.8936,
      "grad_norm": 0.7845570900182349,
      "kl": 1.265625,
      "learning_rate": 1.4984877157701932e-06,
      "loss": 0.0506,
      "reward": 1.578125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.953125,
      "step": 2367
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 477.453125,
      "epoch": 1.8944,
      "grad_norm": 0.4826090070028168,
      "kl": 0.072265625,
      "learning_rate": 1.4965690760569346e-06,
      "loss": 0.0029,
      "reward": 1.921875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 2368
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 365.5625,
      "epoch": 1.8952,
      "grad_norm": 1.8023395400804547,
      "kl": 0.58984375,
      "learning_rate": 1.4946511405901237e-06,
      "loss": 0.0236,
      "reward": 1.9375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 2369
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 430.734375,
      "epoch": 1.896,
      "grad_norm": 1.1268241412178124,
      "kl": 1.1875,
      "learning_rate": 1.4927339107158437e-06,
      "loss": 0.0471,
      "reward": 1.734375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 2370
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 444.96875,
      "epoch": 1.8968,
      "grad_norm": 1.651657729387728,
      "kl": 3.359375,
      "learning_rate": 1.4908173877796784e-06,
      "loss": 0.1348,
      "reward": 1.625,
      "reward_std": 0.31973996758461,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.921875,
      "step": 2371
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 434.875,
      "epoch": 1.8976,
      "grad_norm": 2.6397654725926225,
      "kl": 4.28125,
      "learning_rate": 1.4889015731267186e-06,
      "loss": 0.1708,
      "reward": 1.84375,
      "reward_std": 0.36278265714645386,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.890625,
      "step": 2372
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 389.8125,
      "epoch": 1.8984,
      "grad_norm": 0.5491844825137755,
      "kl": 0.5859375,
      "learning_rate": 1.486986468101555e-06,
      "loss": 0.0236,
      "reward": 1.765625,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.984375,
      "step": 2373
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 451.4375,
      "epoch": 1.8992,
      "grad_norm": 1.7994233823728274,
      "kl": 1.625,
      "learning_rate": 1.4850720740482842e-06,
      "loss": 0.0652,
      "reward": 1.703125,
      "reward_std": 0.18139132857322693,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.953125,
      "step": 2374
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 486.59375,
      "epoch": 1.9,
      "grad_norm": 1.6773978859741498,
      "kl": 3.859375,
      "learning_rate": 1.4831583923105e-06,
      "loss": 0.1539,
      "reward": 1.671875,
      "reward_std": 0.47389551997184753,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.90625,
      "step": 2375
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 378.1875,
      "epoch": 1.9008,
      "grad_norm": 1.8736716490980116,
      "kl": 1.03125,
      "learning_rate": 1.481245424231298e-06,
      "loss": 0.0413,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 2376
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 451.34375,
      "epoch": 1.9016,
      "grad_norm": 1.013615619433797,
      "kl": 0.578125,
      "learning_rate": 1.4793331711532743e-06,
      "loss": 0.0231,
      "reward": 1.90625,
      "reward_std": 0.1828794628381729,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.953125,
      "step": 2377
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 418.34375,
      "epoch": 1.9024,
      "grad_norm": 1.3228661327007896,
      "kl": 0.5625,
      "learning_rate": 1.4774216344185204e-06,
      "loss": 0.0225,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 2378
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 446.140625,
      "epoch": 1.9032,
      "grad_norm": 1.9025870153471482,
      "kl": 1.984375,
      "learning_rate": 1.4755108153686275e-06,
      "loss": 0.0792,
      "reward": 1.78125,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.9375,
      "step": 2379
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 412.5,
      "epoch": 1.904,
      "grad_norm": 1.7443363795882494,
      "kl": 0.96484375,
      "learning_rate": 1.4736007153446803e-06,
      "loss": 0.0384,
      "reward": 1.921875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 2380
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 386.4375,
      "epoch": 1.9048,
      "grad_norm": 1.185011585277608,
      "kl": 2.25,
      "learning_rate": 1.4716913356872614e-06,
      "loss": 0.0901,
      "reward": 1.828125,
      "reward_std": 0.2475731074810028,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.9375,
      "step": 2381
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 388.140625,
      "epoch": 1.9056,
      "grad_norm": 0.7768321191968288,
      "kl": 0.46484375,
      "learning_rate": 1.4697826777364478e-06,
      "loss": 0.0186,
      "reward": 1.578125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 2382
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 461.53125,
      "epoch": 1.9064,
      "grad_norm": 0.697500500552628,
      "kl": 1.453125,
      "learning_rate": 1.467874742831808e-06,
      "loss": 0.0582,
      "reward": 1.546875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.953125,
      "step": 2383
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 369.421875,
      "epoch": 1.9072,
      "grad_norm": 0.16652836720156833,
      "kl": 0.076171875,
      "learning_rate": 1.4659675323124037e-06,
      "loss": 0.003,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2384
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 391.78125,
      "epoch": 1.908,
      "grad_norm": 3.7073476646067087,
      "kl": 0.6953125,
      "learning_rate": 1.46406104751679e-06,
      "loss": 0.0278,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 2385
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 389.5,
      "epoch": 1.9088,
      "grad_norm": 0.10786571220709351,
      "kl": 0.08642578125,
      "learning_rate": 1.462155289783011e-06,
      "loss": 0.0035,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2386
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 422.171875,
      "epoch": 1.9096,
      "grad_norm": 4.944269786732493,
      "kl": 1.03125,
      "learning_rate": 1.4602502604486e-06,
      "loss": 0.0415,
      "reward": 1.78125,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.953125,
      "step": 2387
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 473.328125,
      "epoch": 1.9104,
      "grad_norm": 1.7538214127820733,
      "kl": 2.953125,
      "learning_rate": 1.45834596085058e-06,
      "loss": 0.1182,
      "reward": 1.71875,
      "reward_std": 0.40139204263687134,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.9375,
      "step": 2388
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 418.25,
      "epoch": 1.9112,
      "grad_norm": 0.8850296233423672,
      "kl": 1.9140625,
      "learning_rate": 1.456442392325463e-06,
      "loss": 0.0766,
      "reward": 1.46875,
      "reward_std": 0.19506090879440308,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.953125,
      "step": 2389
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 402.75,
      "epoch": 1.912,
      "grad_norm": 1.7349139855129643,
      "kl": 2.015625,
      "learning_rate": 1.4545395562092467e-06,
      "loss": 0.0806,
      "reward": 1.5625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.9375,
      "step": 2390
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 452.703125,
      "epoch": 1.9127999999999998,
      "grad_norm": 0.8028662692336452,
      "kl": 3.5625,
      "learning_rate": 1.4526374538374133e-06,
      "loss": 0.1424,
      "reward": 1.5,
      "reward_std": 0.28566449880599976,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.90625,
      "step": 2391
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.375,
      "completion_length": 481.40625,
      "epoch": 1.9136,
      "grad_norm": 3.132424951458745,
      "kl": 4.40625,
      "learning_rate": 1.4507360865449318e-06,
      "loss": 0.1768,
      "reward": 1.4375,
      "reward_std": 0.4297013580799103,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.875,
      "step": 2392
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 563.9375,
      "epoch": 1.9144,
      "grad_norm": 8.129251157032787,
      "kl": 13.0,
      "learning_rate": 1.4488354556662553e-06,
      "loss": 0.5199,
      "reward": 1.484375,
      "reward_std": 0.6906865239143372,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.6875,
      "step": 2393
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 455.484375,
      "epoch": 1.9152,
      "grad_norm": 2.3740502257386185,
      "kl": 5.4375,
      "learning_rate": 1.4469355625353199e-06,
      "loss": 0.2183,
      "reward": 1.46875,
      "reward_std": 0.42286166548728943,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.859375,
      "step": 2394
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 440.90625,
      "epoch": 1.916,
      "grad_norm": 3.8721595956167616,
      "kl": 2.8125,
      "learning_rate": 1.4450364084855433e-06,
      "loss": 0.1127,
      "reward": 1.53125,
      "reward_std": 0.2895520329475403,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.890625,
      "step": 2395
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 414.296875,
      "epoch": 1.9167999999999998,
      "grad_norm": 2.1228233089204775,
      "kl": 0.7265625,
      "learning_rate": 1.4431379948498254e-06,
      "loss": 0.029,
      "reward": 1.71875,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.953125,
      "step": 2396
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 385.703125,
      "epoch": 1.9176,
      "grad_norm": 1.6526787411586574,
      "kl": 1.46875,
      "learning_rate": 1.4412403229605453e-06,
      "loss": 0.0587,
      "reward": 1.65625,
      "reward_std": 0.22558550536632538,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.9375,
      "step": 2397
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 473.8125,
      "epoch": 1.9184,
      "grad_norm": 2.800740644453696,
      "kl": 1.03125,
      "learning_rate": 1.4393433941495638e-06,
      "loss": 0.0411,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 2398
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 396.84375,
      "epoch": 1.9192,
      "grad_norm": 0.7896679883569652,
      "kl": 0.333984375,
      "learning_rate": 1.4374472097482156e-06,
      "loss": 0.0134,
      "reward": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 2399
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 442.875,
      "epoch": 1.92,
      "grad_norm": 1.0336282386280489,
      "kl": 0.9609375,
      "learning_rate": 1.4355517710873184e-06,
      "loss": 0.0385,
      "reward": 1.625,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.96875,
      "step": 2400
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 418.640625,
      "epoch": 1.9207999999999998,
      "grad_norm": 0.23397704646346038,
      "kl": 0.09375,
      "learning_rate": 1.4336570794971643e-06,
      "loss": 0.0038,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2401
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 424.09375,
      "epoch": 1.9216,
      "grad_norm": 1.1937637305859858,
      "kl": 0.6640625,
      "learning_rate": 1.4317631363075186e-06,
      "loss": 0.0267,
      "reward": 1.703125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 2402
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 429.71875,
      "epoch": 1.9224,
      "grad_norm": 0.5965034849870672,
      "kl": 0.458984375,
      "learning_rate": 1.4298699428476236e-06,
      "loss": 0.0184,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 2403
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 474.359375,
      "epoch": 1.9232,
      "grad_norm": 1.405933984259602,
      "kl": 0.78125,
      "learning_rate": 1.427977500446199e-06,
      "loss": 0.0313,
      "reward": 1.53125,
      "reward_std": 0.13719715178012848,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.96875,
      "step": 2404
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 358.390625,
      "epoch": 1.924,
      "grad_norm": 2.1879113892963846,
      "kl": 0.72265625,
      "learning_rate": 1.4260858104314299e-06,
      "loss": 0.0289,
      "reward": 1.8125,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.953125,
      "step": 2405
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 376.21875,
      "epoch": 1.9247999999999998,
      "grad_norm": 0.13799559141652026,
      "kl": 0.0966796875,
      "learning_rate": 1.4241948741309783e-06,
      "loss": 0.0039,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2406
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 451.03125,
      "epoch": 1.9256,
      "grad_norm": 1.7073983939222042,
      "kl": 0.56640625,
      "learning_rate": 1.4223046928719764e-06,
      "loss": 0.0227,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.96875,
      "step": 2407
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 382.21875,
      "epoch": 1.9264000000000001,
      "grad_norm": 1.0241736559368684,
      "kl": 0.09814453125,
      "learning_rate": 1.420415267981026e-06,
      "loss": 0.0039,
      "reward": 1.5625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 2408
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 374.171875,
      "epoch": 1.9272,
      "grad_norm": 0.3417625778417692,
      "kl": 0.095703125,
      "learning_rate": 1.418526600784198e-06,
      "loss": 0.0038,
      "reward": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 2409
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 345.171875,
      "epoch": 1.928,
      "grad_norm": 0.7692397213525569,
      "kl": 0.1572265625,
      "learning_rate": 1.4166386926070322e-06,
      "loss": 0.0063,
      "reward": 1.953125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 2410
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.5,
      "completion_length": 405.09375,
      "epoch": 1.9287999999999998,
      "grad_norm": 0.3322491135522514,
      "kl": 0.1298828125,
      "learning_rate": 1.414751544774535e-06,
      "loss": 0.0052,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 2411
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 469.828125,
      "epoch": 1.9296,
      "grad_norm": 0.5821248791023559,
      "kl": 0.2255859375,
      "learning_rate": 1.412865158611179e-06,
      "loss": 0.009,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2412
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 405.546875,
      "epoch": 1.9304000000000001,
      "grad_norm": 0.7299583132737032,
      "kl": 1.6171875,
      "learning_rate": 1.4109795354409045e-06,
      "loss": 0.0648,
      "reward": 1.75,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.953125,
      "step": 2413
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 389.46875,
      "epoch": 1.9312,
      "grad_norm": 0.7082718237042975,
      "kl": 0.6015625,
      "learning_rate": 1.4090946765871105e-06,
      "loss": 0.0241,
      "reward": 1.921875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.96875,
      "step": 2414
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 389.203125,
      "epoch": 1.932,
      "grad_norm": 1.367174375883604,
      "kl": 0.287109375,
      "learning_rate": 1.4072105833726685e-06,
      "loss": 0.0115,
      "reward": 1.671875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 2415
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 381.0625,
      "epoch": 1.9327999999999999,
      "grad_norm": 0.6774045362409198,
      "kl": 0.09619140625,
      "learning_rate": 1.4053272571199037e-06,
      "loss": 0.0038,
      "reward": 1.765625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 2416
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 400.890625,
      "epoch": 1.9336,
      "grad_norm": 0.6653094918917516,
      "kl": 1.2109375,
      "learning_rate": 1.4034446991506084e-06,
      "loss": 0.0482,
      "reward": 1.78125,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.953125,
      "step": 2417
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 402.6875,
      "epoch": 1.9344000000000001,
      "grad_norm": 0.13353366925534746,
      "kl": 0.08935546875,
      "learning_rate": 1.401562910786034e-06,
      "loss": 0.0036,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 2418
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 398.890625,
      "epoch": 1.9352,
      "grad_norm": 0.7638499083157223,
      "kl": 1.140625,
      "learning_rate": 1.3996818933468926e-06,
      "loss": 0.0457,
      "reward": 1.796875,
      "reward_std": 0.18139132857322693,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.953125,
      "step": 2419
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 433.140625,
      "epoch": 1.936,
      "grad_norm": 0.33290967962973383,
      "kl": 0.353515625,
      "learning_rate": 1.397801648153354e-06,
      "loss": 0.0141,
      "reward": 1.859375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.984375,
      "step": 2420
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 418.21875,
      "epoch": 1.9367999999999999,
      "grad_norm": 0.8091019407578133,
      "kl": 0.94921875,
      "learning_rate": 1.395922176525047e-06,
      "loss": 0.0379,
      "reward": 1.890625,
      "reward_std": 0.15981829166412354,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.96875,
      "step": 2421
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 434.828125,
      "epoch": 1.9376,
      "grad_norm": 1.0066023202487273,
      "kl": 0.6484375,
      "learning_rate": 1.3940434797810567e-06,
      "loss": 0.026,
      "reward": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 2422
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 424.203125,
      "epoch": 1.9384000000000001,
      "grad_norm": 1.0731406534843149,
      "kl": 0.99609375,
      "learning_rate": 1.3921655592399256e-06,
      "loss": 0.0401,
      "reward": 1.796875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 2423
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 422.359375,
      "epoch": 1.9392,
      "grad_norm": 0.9808281341958132,
      "kl": 1.8828125,
      "learning_rate": 1.3902884162196509e-06,
      "loss": 0.0753,
      "reward": 1.859375,
      "reward_std": 0.31546199321746826,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9375,
      "step": 2424
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 390.25,
      "epoch": 1.94,
      "grad_norm": 0.5700271885815287,
      "kl": 0.58984375,
      "learning_rate": 1.388412052037682e-06,
      "loss": 0.0236,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 2425
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 392.171875,
      "epoch": 1.9407999999999999,
      "grad_norm": 1.7155177622592277,
      "kl": 1.234375,
      "learning_rate": 1.3865364680109239e-06,
      "loss": 0.0498,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.96875,
      "step": 2426
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 403.84375,
      "epoch": 1.9416,
      "grad_norm": 0.662619706734705,
      "kl": 1.1171875,
      "learning_rate": 1.384661665455736e-06,
      "loss": 0.0446,
      "reward": 1.796875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 2427
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 402.1875,
      "epoch": 1.9424000000000001,
      "grad_norm": 2.8082600803937443,
      "kl": 1.1328125,
      "learning_rate": 1.3827876456879247e-06,
      "loss": 0.0455,
      "reward": 1.4375,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.953125,
      "step": 2428
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 421.53125,
      "epoch": 1.9432,
      "grad_norm": 0.98101898384185,
      "kl": 0.6484375,
      "learning_rate": 1.3809144100227483e-06,
      "loss": 0.0259,
      "reward": 1.71875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 2429
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 421.71875,
      "epoch": 1.944,
      "grad_norm": 1.1167812076105414,
      "kl": 1.3046875,
      "learning_rate": 1.3790419597749198e-06,
      "loss": 0.0521,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 2430
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 371.671875,
      "epoch": 1.9447999999999999,
      "grad_norm": 0.7169936463537103,
      "kl": 0.77734375,
      "learning_rate": 1.3771702962585928e-06,
      "loss": 0.0311,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 2431
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 527.09375,
      "epoch": 1.9456,
      "grad_norm": 0.7200255970830155,
      "kl": 2.265625,
      "learning_rate": 1.3752994207873743e-06,
      "loss": 0.0902,
      "reward": 1.78125,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.9375,
      "step": 2432
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 434.90625,
      "epoch": 1.9464000000000001,
      "grad_norm": 2.528998755799731,
      "kl": 1.640625,
      "learning_rate": 1.373429334674317e-06,
      "loss": 0.0656,
      "reward": 1.75,
      "reward_std": 0.2314550280570984,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.9375,
      "step": 2433
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 425.203125,
      "epoch": 1.9472,
      "grad_norm": 0.9374989047821527,
      "kl": 1.28125,
      "learning_rate": 1.3715600392319186e-06,
      "loss": 0.0513,
      "reward": 1.78125,
      "reward_std": 0.24684599041938782,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 2434
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 434.71875,
      "epoch": 1.948,
      "grad_norm": 0.3005621124258611,
      "kl": 0.66015625,
      "learning_rate": 1.369691535772123e-06,
      "loss": 0.0265,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.984375,
      "step": 2435
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 455.453125,
      "epoch": 1.9487999999999999,
      "grad_norm": 1.2796207746661543,
      "kl": 1.1796875,
      "learning_rate": 1.3678238256063193e-06,
      "loss": 0.0472,
      "reward": 1.6875,
      "reward_std": 0.213067427277565,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 2436
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 440.140625,
      "epoch": 1.9496,
      "grad_norm": 0.9651874328660741,
      "kl": 2.21875,
      "learning_rate": 1.3659569100453346e-06,
      "loss": 0.0889,
      "reward": 1.734375,
      "reward_std": 0.28778618574142456,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.9375,
      "step": 2437
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 435.359375,
      "epoch": 1.9504000000000001,
      "grad_norm": 2.6329782048013586,
      "kl": 2.125,
      "learning_rate": 1.3640907903994455e-06,
      "loss": 0.085,
      "reward": 1.765625,
      "reward_std": 0.15499483048915863,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.9375,
      "step": 2438
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 411.515625,
      "epoch": 1.9512,
      "grad_norm": 1.4783404096789414,
      "kl": 0.130859375,
      "learning_rate": 1.3622254679783665e-06,
      "loss": 0.0052,
      "reward": 1.734375,
      "reward_std": 0.17782479524612427,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 2439
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 465.5625,
      "epoch": 1.952,
      "grad_norm": 1.6253159465554536,
      "kl": 3.90625,
      "learning_rate": 1.3603609440912508e-06,
      "loss": 0.1559,
      "reward": 1.453125,
      "reward_std": 0.5136494040489197,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.890625,
      "step": 2440
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 469.328125,
      "epoch": 1.9527999999999999,
      "grad_norm": 1.8519862602162838,
      "kl": 2.09375,
      "learning_rate": 1.3584972200466936e-06,
      "loss": 0.0835,
      "reward": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.9375,
      "step": 2441
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 448.578125,
      "epoch": 1.9536,
      "grad_norm": 1.386289677708159,
      "kl": 3.390625,
      "learning_rate": 1.356634297152729e-06,
      "loss": 0.1358,
      "reward": 1.390625,
      "reward_std": 0.3298586905002594,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.90625,
      "step": 2442
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.0,
      "completion_length": 433.734375,
      "epoch": 1.9544000000000001,
      "grad_norm": 1.4488160932843597,
      "kl": 3.21875,
      "learning_rate": 1.3547721767168273e-06,
      "loss": 0.1293,
      "reward": 1.734375,
      "reward_std": 0.5867297649383545,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.890625,
      "step": 2443
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 540.9375,
      "epoch": 1.9552,
      "grad_norm": 0.5355961107458718,
      "kl": 1.984375,
      "learning_rate": 1.3529108600458967e-06,
      "loss": 0.0796,
      "reward": 1.796875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.953125,
      "step": 2444
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 457.359375,
      "epoch": 1.956,
      "grad_norm": 1.4905781930338178,
      "kl": 1.4609375,
      "learning_rate": 1.3510503484462807e-06,
      "loss": 0.0587,
      "reward": 1.75,
      "reward_std": 0.2040124535560608,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 0.953125,
      "step": 2445
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 486.953125,
      "epoch": 1.9567999999999999,
      "grad_norm": 2.6096505744523775,
      "kl": 2.4375,
      "learning_rate": 1.349190643223758e-06,
      "loss": 0.0973,
      "reward": 1.796875,
      "reward_std": 0.42498332262039185,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.921875,
      "step": 2446
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 427.109375,
      "epoch": 1.9576,
      "grad_norm": 8.682296049745519,
      "kl": 1.8671875,
      "learning_rate": 1.347331745683542e-06,
      "loss": 0.0747,
      "reward": 1.828125,
      "reward_std": 0.23925508558750153,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.953125,
      "step": 2447
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.375,
      "completion_length": 420.25,
      "epoch": 1.9584000000000001,
      "grad_norm": 1.0152230889390625,
      "kl": 1.2734375,
      "learning_rate": 1.3454736571302761e-06,
      "loss": 0.051,
      "reward": 1.46875,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.953125,
      "step": 2448
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.375,
      "completion_length": 478.875,
      "epoch": 1.9592,
      "grad_norm": 0.9667083422075894,
      "kl": 1.4375,
      "learning_rate": 1.3436163788680411e-06,
      "loss": 0.0576,
      "reward": 1.546875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 2449
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 425.421875,
      "epoch": 1.96,
      "grad_norm": 1.3667488608028606,
      "kl": 1.8671875,
      "learning_rate": 1.3417599122003464e-06,
      "loss": 0.0745,
      "reward": 1.78125,
      "reward_std": 0.2924008071422577,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.921875,
      "step": 2450
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 393.796875,
      "epoch": 1.9607999999999999,
      "grad_norm": 0.71504983170553,
      "kl": 1.1796875,
      "learning_rate": 1.3399042584301298e-06,
      "loss": 0.0473,
      "reward": 1.703125,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 2451
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 426.921875,
      "epoch": 1.9616,
      "grad_norm": 0.961385853315966,
      "kl": 1.0546875,
      "learning_rate": 1.3380494188597603e-06,
      "loss": 0.042,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 2452
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 413.203125,
      "epoch": 1.9624000000000001,
      "grad_norm": 3.8970906370746277,
      "kl": 1.34375,
      "learning_rate": 1.3361953947910394e-06,
      "loss": 0.0538,
      "reward": 1.828125,
      "reward_std": 0.34564992785453796,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.953125,
      "step": 2453
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 398.75,
      "epoch": 1.9632,
      "grad_norm": 0.33225482099165854,
      "kl": 0.8828125,
      "learning_rate": 1.334342187525189e-06,
      "loss": 0.0352,
      "reward": 1.9375,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.96875,
      "step": 2454
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 458.078125,
      "epoch": 1.964,
      "grad_norm": 9.28529152616632,
      "kl": 1.1171875,
      "learning_rate": 1.3324897983628621e-06,
      "loss": 0.0446,
      "reward": 1.703125,
      "reward_std": 0.21730589866638184,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.953125,
      "step": 2455
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.0,
      "completion_length": 482.25,
      "epoch": 1.9647999999999999,
      "grad_norm": 1.1380537604259222,
      "kl": 0.5546875,
      "learning_rate": 1.330638228604137e-06,
      "loss": 0.0221,
      "reward": 1.734375,
      "reward_std": 0.20189079642295837,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 2456
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 440.40625,
      "epoch": 1.9656,
      "grad_norm": 0.7523422406555214,
      "kl": 0.98046875,
      "learning_rate": 1.3287874795485168e-06,
      "loss": 0.0394,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 2457
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 412.40625,
      "epoch": 1.9664000000000001,
      "grad_norm": 0.6149431433268288,
      "kl": 0.9453125,
      "learning_rate": 1.3269375524949286e-06,
      "loss": 0.0378,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 2458
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 430.15625,
      "epoch": 1.9672,
      "grad_norm": 0.4774950978450209,
      "kl": 0.328125,
      "learning_rate": 1.3250884487417227e-06,
      "loss": 0.0131,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2459
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 453.515625,
      "epoch": 1.968,
      "grad_norm": 0.8734251798841773,
      "kl": 2.015625,
      "learning_rate": 1.3232401695866686e-06,
      "loss": 0.0805,
      "reward": 1.65625,
      "reward_std": 0.2709311842918396,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.9375,
      "step": 2460
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 457.71875,
      "epoch": 1.9687999999999999,
      "grad_norm": 1.3631889780253275,
      "kl": 1.25,
      "learning_rate": 1.321392716326963e-06,
      "loss": 0.05,
      "reward": 1.6875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 2461
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 434.234375,
      "epoch": 1.9696,
      "grad_norm": 0.6394979876062535,
      "kl": 0.828125,
      "learning_rate": 1.3195460902592193e-06,
      "loss": 0.0331,
      "reward": 1.609375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 2462
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.125,
      "completion_length": 444.84375,
      "epoch": 1.9704000000000002,
      "grad_norm": 2.514048156633523,
      "kl": 3.875,
      "learning_rate": 1.3177002926794685e-06,
      "loss": 0.1551,
      "reward": 1.359375,
      "reward_std": 0.5200409889221191,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.859375,
      "step": 2463
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 474.328125,
      "epoch": 1.9712,
      "grad_norm": 1.1340715503269814,
      "kl": 2.609375,
      "learning_rate": 1.3158553248831658e-06,
      "loss": 0.1041,
      "reward": 1.59375,
      "reward_std": 0.3718376159667969,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.9375,
      "step": 2464
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 438.875,
      "epoch": 1.972,
      "grad_norm": 0.6956774003491915,
      "kl": 1.0390625,
      "learning_rate": 1.3140111881651773e-06,
      "loss": 0.0416,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.96875,
      "step": 2465
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 411.171875,
      "epoch": 1.9727999999999999,
      "grad_norm": 0.5235973050143332,
      "kl": 0.83203125,
      "learning_rate": 1.312167883819791e-06,
      "loss": 0.0332,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 2466
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.25,
      "completion_length": 456.796875,
      "epoch": 1.9736,
      "grad_norm": 1.473294108100521,
      "kl": 1.1640625,
      "learning_rate": 1.3103254131407082e-06,
      "loss": 0.0464,
      "reward": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.96875,
      "step": 2467
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.0,
      "completion_length": 427.90625,
      "epoch": 1.9744000000000002,
      "grad_norm": 4.62995735255069,
      "kl": 3.125,
      "learning_rate": 1.308483777421046e-06,
      "loss": 0.1251,
      "reward": 1.59375,
      "reward_std": 0.48066505789756775,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.90625,
      "step": 2468
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 452.9375,
      "epoch": 1.9752,
      "grad_norm": 2.3200179982419966,
      "kl": 1.84375,
      "learning_rate": 1.3066429779533352e-06,
      "loss": 0.0735,
      "reward": 1.671875,
      "reward_std": 0.32764342427253723,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.9375,
      "step": 2469
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 490.140625,
      "epoch": 1.976,
      "grad_norm": 1.138649250709883,
      "kl": 2.28125,
      "learning_rate": 1.3048030160295196e-06,
      "loss": 0.0912,
      "reward": 1.671875,
      "reward_std": 0.27883461117744446,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.9375,
      "step": 2470
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 473.984375,
      "epoch": 1.9768,
      "grad_norm": 3.0332609973450797,
      "kl": 1.2578125,
      "learning_rate": 1.3029638929409555e-06,
      "loss": 0.0504,
      "reward": 1.6875,
      "reward_std": 0.13719715178012848,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.953125,
      "step": 2471
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 442.9375,
      "epoch": 1.9776,
      "grad_norm": 1.7715433838125119,
      "kl": 3.546875,
      "learning_rate": 1.3011256099784103e-06,
      "loss": 0.1423,
      "reward": 1.578125,
      "reward_std": 0.47417181730270386,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.890625,
      "step": 2472
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 442.265625,
      "epoch": 1.9784000000000002,
      "grad_norm": 2.235711684889043,
      "kl": 2.390625,
      "learning_rate": 1.2992881684320627e-06,
      "loss": 0.0958,
      "reward": 1.671875,
      "reward_std": 0.43047669529914856,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.921875,
      "step": 2473
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.0,
      "completion_length": 478.890625,
      "epoch": 1.9792,
      "grad_norm": 1.767685645904705,
      "kl": 4.09375,
      "learning_rate": 1.297451569591498e-06,
      "loss": 0.1643,
      "reward": 1.796875,
      "reward_std": 0.4343159794807434,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.875,
      "step": 2474
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.125,
      "completion_length": 503.921875,
      "epoch": 1.98,
      "grad_norm": 1.558552636436934,
      "kl": 5.6875,
      "learning_rate": 1.2956158147457116e-06,
      "loss": 0.227,
      "reward": 1.5625,
      "reward_std": 0.4267922341823578,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.8125,
      "step": 2475
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 580.984375,
      "epoch": 1.9808,
      "grad_norm": 7.419471445231683,
      "kl": 7.5,
      "learning_rate": 1.2937809051831102e-06,
      "loss": 0.3002,
      "reward": 1.421875,
      "reward_std": 0.4921797215938568,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.765625,
      "step": 2476
    },
    {
      "all_correct": 0.0,
      "all_wrong": 0.125,
      "completion_length": 588.640625,
      "epoch": 1.9816,
      "grad_norm": 7.921893460533328,
      "kl": 11.875,
      "learning_rate": 1.2919468421915008e-06,
      "loss": 0.4759,
      "reward": 1.375,
      "reward_std": 0.7253988981246948,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.671875,
      "step": 2477
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 444.390625,
      "epoch": 1.9824000000000002,
      "grad_norm": 3.1882832646838053,
      "kl": 4.71875,
      "learning_rate": 1.2901136270580994e-06,
      "loss": 0.1892,
      "reward": 1.5625,
      "reward_std": 0.29476165771484375,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.84375,
      "step": 2478
    },
    {
      "all_correct": 0.125,
      "all_wrong": 0.5,
      "completion_length": 437.578125,
      "epoch": 1.9832,
      "grad_norm": 1.2594971316236099,
      "kl": 3.328125,
      "learning_rate": 1.2882812610695305e-06,
      "loss": 0.1334,
      "reward": 1.21875,
      "reward_std": 0.3333997130393982,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 0.875,
      "step": 2479
    },
    {
      "all_correct": 0.25,
      "all_wrong": 0.25,
      "completion_length": 466.5,
      "epoch": 1.984,
      "grad_norm": 1.2952932685895526,
      "kl": 3.0625,
      "learning_rate": 1.2864497455118152e-06,
      "loss": 0.1226,
      "reward": 1.515625,
      "reward_std": 0.4027767777442932,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.875,
      "step": 2480
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 392.984375,
      "epoch": 1.9848,
      "grad_norm": 1.7427594979289083,
      "kl": 0.453125,
      "learning_rate": 1.2846190816703836e-06,
      "loss": 0.0182,
      "reward": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 2481
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 409.3125,
      "epoch": 1.9856,
      "grad_norm": 1.829837754717998,
      "kl": 0.59765625,
      "learning_rate": 1.2827892708300648e-06,
      "loss": 0.0239,
      "reward": 1.9375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.953125,
      "step": 2482
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 422.328125,
      "epoch": 1.9864000000000002,
      "grad_norm": 1.6968579942041917,
      "kl": 1.140625,
      "learning_rate": 1.280960314275092e-06,
      "loss": 0.0457,
      "reward": 1.75,
      "reward_std": 0.2834492623806,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.9375,
      "step": 2483
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 436.828125,
      "epoch": 1.9872,
      "grad_norm": 1.3706405759005194,
      "kl": 0.67578125,
      "learning_rate": 1.279132213289096e-06,
      "loss": 0.0271,
      "reward": 1.8125,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.96875,
      "step": 2484
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 383.765625,
      "epoch": 1.988,
      "grad_norm": 1.2479577260390846,
      "kl": 0.7265625,
      "learning_rate": 1.2773049691551103e-06,
      "loss": 0.0291,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 2485
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 476.171875,
      "epoch": 1.9888,
      "grad_norm": 0.4718047526902604,
      "kl": 0.296875,
      "learning_rate": 1.2754785831555617e-06,
      "loss": 0.0119,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 2486
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.0,
      "completion_length": 426.515625,
      "epoch": 1.9896,
      "grad_norm": 3.873672543262227,
      "kl": 0.81640625,
      "learning_rate": 1.273653056572282e-06,
      "loss": 0.0326,
      "reward": 1.84375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.953125,
      "step": 2487
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.125,
      "completion_length": 385.640625,
      "epoch": 1.9904,
      "grad_norm": 3.808522502892453,
      "kl": 0.80859375,
      "learning_rate": 1.2718283906864939e-06,
      "loss": 0.0324,
      "reward": 1.640625,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.96875,
      "step": 2488
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.375,
      "completion_length": 355.453125,
      "epoch": 1.9912,
      "grad_norm": 0.8173967776247805,
      "kl": 0.4765625,
      "learning_rate": 1.2700045867788184e-06,
      "loss": 0.0191,
      "reward": 1.59375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.984375,
      "step": 2489
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.25,
      "completion_length": 398.59375,
      "epoch": 1.992,
      "grad_norm": 0.6222982221544567,
      "kl": 0.890625,
      "learning_rate": 1.2681816461292715e-06,
      "loss": 0.0356,
      "reward": 1.609375,
      "reward_std": 0.15086673200130463,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 2490
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.0,
      "completion_length": 461.859375,
      "epoch": 1.9928,
      "grad_norm": 0.7749320110635745,
      "kl": 0.373046875,
      "learning_rate": 1.2663595700172631e-06,
      "loss": 0.0149,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.96875,
      "step": 2491
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.0,
      "completion_length": 390.984375,
      "epoch": 1.9936,
      "grad_norm": 0.6220782396058596,
      "kl": 0.4765625,
      "learning_rate": 1.2645383597215965e-06,
      "loss": 0.019,
      "reward": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 2492
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.125,
      "completion_length": 390.734375,
      "epoch": 1.9944,
      "grad_norm": 1.206073841190381,
      "kl": 1.1171875,
      "learning_rate": 1.2627180165204671e-06,
      "loss": 0.0449,
      "reward": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.96875,
      "step": 2493
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 417.6875,
      "epoch": 1.9952,
      "grad_norm": 1.295960181274372,
      "kl": 0.86328125,
      "learning_rate": 1.2608985416914616e-06,
      "loss": 0.0344,
      "reward": 1.734375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 2494
    },
    {
      "all_correct": 0.75,
      "all_wrong": 0.125,
      "completion_length": 465.171875,
      "epoch": 1.996,
      "grad_norm": 0.5028284488457815,
      "kl": 0.87109375,
      "learning_rate": 1.259079936511558e-06,
      "loss": 0.0347,
      "reward": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 2495
    },
    {
      "all_correct": 0.375,
      "all_wrong": 0.25,
      "completion_length": 390.828125,
      "epoch": 1.9968,
      "grad_norm": 0.9870454487172972,
      "kl": 0.384765625,
      "learning_rate": 1.257262202257124e-06,
      "loss": 0.0154,
      "reward": 1.65625,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 2496
    },
    {
      "all_correct": 0.625,
      "all_wrong": 0.25,
      "completion_length": 451.09375,
      "epoch": 1.9976,
      "grad_norm": 0.7070811807566644,
      "kl": 0.271484375,
      "learning_rate": 1.2554453402039124e-06,
      "loss": 0.0108,
      "reward": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 2497
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 377.453125,
      "epoch": 1.9984,
      "grad_norm": 0.1712508918566282,
      "kl": 0.08544921875,
      "learning_rate": 1.2536293516270704e-06,
      "loss": 0.0034,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2498
    },
    {
      "all_correct": 0.875,
      "all_wrong": 0.125,
      "completion_length": 393.109375,
      "epoch": 1.9992,
      "grad_norm": 0.08793411536601524,
      "kl": 0.09326171875,
      "learning_rate": 1.251814237801128e-06,
      "loss": 0.0037,
      "reward": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 2499
    },
    {
      "all_correct": 0.5,
      "all_wrong": 0.125,
      "completion_length": 351.6875,
      "epoch": 2.0,
      "grad_norm": 1.4112992865682874,
      "kl": 1.15625,
      "learning_rate": 1.2500000000000007e-06,
      "loss": 0.0462,
      "reward": 1.78125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.96875,
      "step": 2500
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 3750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}