diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,18787 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.2222222222222222,
+  "eval_steps": 1000,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 246.3125,
+      "epoch": 8.888888888888889e-05,
+      "grad_norm": 60.320960998535156,
+      "learning_rate": 2.5e-07,
+      "loss": -10.229,
+      "reward": 1.7395833730697632,
+      "reward_std": 0.6432403922080994,
+      "rewards/boxed_and_answer_tags_format_reward": 0.65625,
+      "rewards/correctness_reward_func_math": 1.0833333432674408,
+      "step": 1,
+      "zero_std_ratio": 0.0
+    },
+    {
+      "epoch": 0.00017777777777777779,
+      "grad_norm": 71.0031509399414,
+      "learning_rate": 5e-07,
+      "loss": -9.5625,
+      "step": 2
+    },
+    {
+      "epoch": 0.0002666666666666667,
+      "grad_norm": 61.95022964477539,
+      "learning_rate": 7.5e-07,
+      "loss": -16.2291,
+      "step": 3
+    },
+    {
+      "epoch": 0.00035555555555555557,
+      "grad_norm": 67.81867980957031,
+      "learning_rate": 1e-06,
+      "loss": -11.0016,
+      "step": 4
+    },
+    {
+      "epoch": 0.00044444444444444447,
+      "grad_norm": 57.108917236328125,
+      "learning_rate": 1.25e-06,
+      "loss": -6.1658,
+      "step": 5
+    },
+    {
+      "epoch": 0.0005333333333333334,
+      "grad_norm": 72.85011291503906,
+      "learning_rate": 1.5e-06,
+      "loss": -4.0145,
+      "step": 6
+    },
+    {
+      "epoch": 0.0006222222222222223,
+      "grad_norm": 59.103431701660156,
+      "learning_rate": 1.7500000000000002e-06,
+      "loss": -9.9488,
+      "step": 7
+    },
+    {
+      "epoch": 0.0007111111111111111,
+      "grad_norm": 73.94007873535156,
+      "learning_rate": 2e-06,
+      "loss": -9.1809,
+      "step": 8
+    },
+    {
+      "epoch": 0.0008,
+      "grad_norm": 62.28184509277344,
+      "learning_rate": 2.25e-06,
+      "loss": -16.4311,
+      "step": 9
+    },
+    {
+      "epoch": 0.0008888888888888889,
+      "grad_norm": 68.46251678466797,
+      "learning_rate": 2.5e-06,
+      "loss": -11.402,
+      "step": 10
+    },
+    {
+      "epoch": 0.0009777777777777777,
+      "grad_norm": 59.184749603271484,
+      "learning_rate": 2.75e-06,
+      "loss": -6.1384,
+      "step": 11
+    },
+    {
+      "epoch": 0.0010666666666666667,
+      "grad_norm": 71.60365295410156,
+      "learning_rate": 3e-06,
+      "loss": -3.932,
+      "step": 12
+    },
+    {
+      "completion_length": 249.62500762939453,
+      "epoch": 0.0011555555555555555,
+      "grad_norm": 61.79197311401367,
+      "learning_rate": 3e-06,
+      "loss": -2.1748,
+      "reward": 0.9791666865348816,
+      "reward_std": 0.4510806053876877,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6458333134651184,
+      "rewards/correctness_reward_func_math": 0.3333333283662796,
+      "step": 13,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.0012444444444444445,
+      "grad_norm": 55.82655334472656,
+      "learning_rate": 3e-06,
+      "loss": -3.8107,
+      "step": 14
+    },
+    {
+      "epoch": 0.0013333333333333333,
+      "grad_norm": 58.3712158203125,
+      "learning_rate": 3e-06,
+      "loss": 4.085,
+      "step": 15
+    },
+    {
+      "epoch": 0.0014222222222222223,
+      "grad_norm": 73.97306823730469,
+      "learning_rate": 3e-06,
+      "loss": -4.8915,
+      "step": 16
+    },
+    {
+      "epoch": 0.001511111111111111,
+      "grad_norm": 51.51576232910156,
+      "learning_rate": 3e-06,
+      "loss": -0.161,
+      "step": 17
+    },
+    {
+      "epoch": 0.0016,
+      "grad_norm": 72.59639739990234,
+      "learning_rate": 3e-06,
+      "loss": 3.0369,
+      "step": 18
+    },
+    {
+      "epoch": 0.0016888888888888889,
+      "grad_norm": 48.86510467529297,
+      "learning_rate": 3e-06,
+      "loss": -2.3437,
+      "step": 19
+    },
+    {
+      "epoch": 0.0017777777777777779,
+      "grad_norm": 55.3180046081543,
+      "learning_rate": 3e-06,
+      "loss": -4.3585,
+      "step": 20
+    },
+    {
+      "epoch": 0.0018666666666666666,
+      "grad_norm": 54.75101089477539,
+      "learning_rate": 3e-06,
+      "loss": 3.9965,
+      "step": 21
+    },
+    {
+      "epoch": 0.0019555555555555554,
+      "grad_norm": 75.27330017089844,
+      "learning_rate": 3e-06,
+      "loss": -4.9733,
+      "step": 22
+    },
+    {
+      "epoch": 0.0020444444444444447,
+      "grad_norm": 51.991214752197266,
+      "learning_rate": 3e-06,
+      "loss": -0.3033,
+      "step": 23
+    },
+    {
+      "epoch": 0.0021333333333333334,
+      "grad_norm": 54.72827911376953,
+      "learning_rate": 3e-06,
+      "loss": 2.5278,
+      "step": 24
+    },
+    {
+      "completion_length": 226.8125,
+      "epoch": 0.0022222222222222222,
+      "grad_norm": 58.25025939941406,
+      "learning_rate": 3e-06,
+      "loss": -6.7768,
+      "reward": 1.6041666865348816,
+      "reward_std": 0.6311438381671906,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6458333432674408,
+      "rewards/correctness_reward_func_math": 0.9583333432674408,
+      "step": 25,
+      "zero_std_ratio": 0.0
+    },
+    {
+      "epoch": 0.002311111111111111,
+      "grad_norm": 68.62551879882812,
+      "learning_rate": 3e-06,
+      "loss": -4.0906,
+      "step": 26
+    },
+    {
+      "epoch": 0.0024,
+      "grad_norm": 65.11053466796875,
+      "learning_rate": 3e-06,
+      "loss": -4.6172,
+      "step": 27
+    },
+    {
+      "epoch": 0.002488888888888889,
+      "grad_norm": 76.28429412841797,
+      "learning_rate": 3e-06,
+      "loss": -7.9209,
+      "step": 28
+    },
+    {
+      "epoch": 0.002577777777777778,
+      "grad_norm": 62.037696838378906,
+      "learning_rate": 3e-06,
+      "loss": -3.4414,
+      "step": 29
+    },
+    {
+      "epoch": 0.0026666666666666666,
+      "grad_norm": 58.92220687866211,
+      "learning_rate": 3e-06,
+      "loss": -3.2836,
+      "step": 30
+    },
+    {
+      "epoch": 0.0027555555555555554,
+      "grad_norm": 57.03800582885742,
+      "learning_rate": 3e-06,
+      "loss": -7.1747,
+      "step": 31
+    },
+    {
+      "epoch": 0.0028444444444444446,
+      "grad_norm": 71.39422607421875,
+      "learning_rate": 3e-06,
+      "loss": -4.5251,
+      "step": 32
+    },
+    {
+      "epoch": 0.0029333333333333334,
+      "grad_norm": 130.19813537597656,
+      "learning_rate": 3e-06,
+      "loss": -4.5744,
+      "step": 33
+    },
+    {
+      "epoch": 0.003022222222222222,
+      "grad_norm": 76.09828186035156,
+      "learning_rate": 3e-06,
+      "loss": -7.9552,
+      "step": 34
+    },
+    {
+      "epoch": 0.003111111111111111,
+      "grad_norm": 63.77288055419922,
+      "learning_rate": 3e-06,
+      "loss": -3.6391,
+      "step": 35
+    },
+    {
+      "epoch": 0.0032,
+      "grad_norm": 58.53509521484375,
+      "learning_rate": 3e-06,
+      "loss": -3.9415,
+      "step": 36
+    },
+    {
+      "completion_length": 245.14583587646484,
+      "epoch": 0.003288888888888889,
+      "grad_norm": 53.06296920776367,
+      "learning_rate": 3e-06,
+      "loss": 7.1798,
+      "reward": 0.9375000298023224,
+      "reward_std": 0.3340114951133728,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6041666567325592,
+      "rewards/correctness_reward_func_math": 0.3333333246409893,
+      "step": 37,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.0033777777777777777,
+      "grad_norm": 78.04679870605469,
+      "learning_rate": 3e-06,
+      "loss": 13.2393,
+      "step": 38
+    },
+    {
+      "epoch": 0.0034666666666666665,
+      "grad_norm": 64.38521575927734,
+      "learning_rate": 3e-06,
+      "loss": 11.5406,
+      "step": 39
+    },
+    {
+      "epoch": 0.0035555555555555557,
+      "grad_norm": 56.69493865966797,
+      "learning_rate": 3e-06,
+      "loss": 11.0537,
+      "step": 40
+    },
+    {
+      "epoch": 0.0036444444444444445,
+      "grad_norm": 59.67893600463867,
+      "learning_rate": 3e-06,
+      "loss": 12.2084,
+      "step": 41
+    },
+    {
+      "epoch": 0.0037333333333333333,
+      "grad_norm": 44.71684646606445,
+      "learning_rate": 3e-06,
+      "loss": 14.915,
+      "step": 42
+    },
+    {
+      "epoch": 0.003822222222222222,
+      "grad_norm": 53.003570556640625,
+      "learning_rate": 3e-06,
+      "loss": 7.1581,
+      "step": 43
+    },
+    {
+      "epoch": 0.003911111111111111,
+      "grad_norm": 86.505615234375,
+      "learning_rate": 3e-06,
+      "loss": 13.0403,
+      "step": 44
+    },
+    {
+      "epoch": 0.004,
+      "grad_norm": 73.7258529663086,
+      "learning_rate": 3e-06,
+      "loss": 11.1962,
+      "step": 45
+    },
+    {
+      "epoch": 0.004088888888888889,
+      "grad_norm": 92.57136535644531,
+      "learning_rate": 3e-06,
+      "loss": 10.6526,
+      "step": 46
+    },
+    {
+      "epoch": 0.004177777777777778,
+      "grad_norm": 63.43205642700195,
+      "learning_rate": 3e-06,
+      "loss": 11.8212,
+      "step": 47
+    },
+    {
+      "epoch": 0.004266666666666667,
+      "grad_norm": 44.73876953125,
+      "learning_rate": 3e-06,
+      "loss": 14.3455,
+      "step": 48
+    },
+    {
+      "completion_length": 236.37500762939453,
+      "epoch": 0.004355555555555555,
+      "grad_norm": 42.604164123535156,
+      "learning_rate": 3e-06,
+      "loss": -28.8562,
+      "reward": 1.6145833730697632,
+      "reward_std": 0.3440491110086441,
+      "rewards/boxed_and_answer_tags_format_reward": 0.65625,
+      "rewards/correctness_reward_func_math": 0.9583333432674408,
+      "step": 49,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.0044444444444444444,
+      "grad_norm": 55.280696868896484,
+      "learning_rate": 3e-06,
+      "loss": -29.8671,
+      "step": 50
+    },
+    {
+      "epoch": 0.004533333333333334,
+      "grad_norm": 53.84416198730469,
+      "learning_rate": 3e-06,
+      "loss": -28.6148,
+      "step": 51
+    },
+    {
+      "epoch": 0.004622222222222222,
+      "grad_norm": 48.8647575378418,
+      "learning_rate": 3e-06,
+      "loss": -28.0853,
+      "step": 52
+    },
+    {
+      "epoch": 0.004711111111111111,
+      "grad_norm": 65.01343536376953,
+      "learning_rate": 3e-06,
+      "loss": -26.2356,
+      "step": 53
+    },
+    {
+      "epoch": 0.0048,
+      "grad_norm": 64.81402587890625,
+      "learning_rate": 3e-06,
+      "loss": -30.8205,
+      "step": 54
+    },
+    {
+      "epoch": 0.004888888888888889,
+      "grad_norm": 44.85778045654297,
+      "learning_rate": 3e-06,
+      "loss": -28.6691,
+      "step": 55
+    },
+    {
+      "epoch": 0.004977777777777778,
+      "grad_norm": 45.61606216430664,
+      "learning_rate": 3e-06,
+      "loss": -30.0595,
+      "step": 56
+    },
+    {
+      "epoch": 0.005066666666666666,
+      "grad_norm": 49.3116455078125,
+      "learning_rate": 3e-06,
+      "loss": -28.8315,
+      "step": 57
+    },
+    {
+      "epoch": 0.005155555555555556,
+      "grad_norm": 45.42935562133789,
+      "learning_rate": 3e-06,
+      "loss": -28.1493,
+      "step": 58
+    },
+    {
+      "epoch": 0.005244444444444445,
+      "grad_norm": 52.282257080078125,
+      "learning_rate": 3e-06,
+      "loss": -26.8023,
+      "step": 59
+    },
+    {
+      "epoch": 0.005333333333333333,
+      "grad_norm": 61.042945861816406,
+      "learning_rate": 3e-06,
+      "loss": -30.9091,
+      "step": 60
+    },
+    {
+      "completion_length": 250.8541717529297,
+      "epoch": 0.005422222222222222,
+      "grad_norm": 56.66669464111328,
+      "learning_rate": 3e-06,
+      "loss": -4.2644,
+      "reward": 1.0104166865348816,
+      "reward_std": 0.3859569579362869,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5104166716337204,
+      "rewards/correctness_reward_func_math": 0.4999999850988388,
+      "step": 61,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.005511111111111111,
+      "grad_norm": 46.742279052734375,
+      "learning_rate": 3e-06,
+      "loss": 1.921,
+      "step": 62
+    },
+    {
+      "epoch": 0.0056,
+      "grad_norm": 64.74068450927734,
+      "learning_rate": 3e-06,
+      "loss": -1.0677,
+      "step": 63
+    },
+    {
+      "epoch": 0.005688888888888889,
+      "grad_norm": 53.72319412231445,
+      "learning_rate": 3e-06,
+      "loss": 0.7498,
+      "step": 64
+    },
+    {
+      "epoch": 0.0057777777777777775,
+      "grad_norm": 51.9224739074707,
+      "learning_rate": 3e-06,
+      "loss": -1.7073,
+      "step": 65
+    },
+    {
+      "epoch": 0.005866666666666667,
+      "grad_norm": 49.95579528808594,
+      "learning_rate": 3e-06,
+      "loss": -4.7011,
+      "step": 66
+    },
+    {
+      "epoch": 0.005955555555555556,
+      "grad_norm": 54.09262466430664,
+      "learning_rate": 3e-06,
+      "loss": -4.882,
+      "step": 67
+    },
+    {
+      "epoch": 0.006044444444444444,
+      "grad_norm": 51.433746337890625,
+      "learning_rate": 3e-06,
+      "loss": 1.6496,
+      "step": 68
+    },
+    {
+      "epoch": 0.0061333333333333335,
+      "grad_norm": 48.16537094116211,
+      "learning_rate": 3e-06,
+      "loss": -1.5035,
+      "step": 69
+    },
+    {
+      "epoch": 0.006222222222222222,
+      "grad_norm": 55.34268569946289,
+      "learning_rate": 3e-06,
+      "loss": 0.0384,
+      "step": 70
+    },
+    {
+      "epoch": 0.006311111111111111,
+      "grad_norm": 45.631813049316406,
+      "learning_rate": 3e-06,
+      "loss": -1.8713,
+      "step": 71
+    },
+    {
+      "epoch": 0.0064,
+      "grad_norm": 48.471473693847656,
+      "learning_rate": 3e-06,
+      "loss": -4.9618,
+      "step": 72
+    },
+    {
+      "completion_length": 231.14583587646484,
+      "epoch": 0.006488888888888889,
+      "grad_norm": 91.5987777709961,
+      "learning_rate": 3e-06,
+      "loss": 59.5993,
+      "reward": 1.1145833730697632,
+      "reward_std": 0.4806128740310669,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.5,
+      "step": 73,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.006577777777777778,
+      "grad_norm": 83.08058166503906,
+      "learning_rate": 3e-06,
+      "loss": 67.2965,
+      "step": 74
+    },
+    {
+      "epoch": 0.006666666666666667,
+      "grad_norm": 65.38250732421875,
+      "learning_rate": 3e-06,
+      "loss": 60.7344,
+      "step": 75
+    },
+    {
+      "epoch": 0.0067555555555555554,
+      "grad_norm": 66.78120422363281,
+      "learning_rate": 3e-06,
+      "loss": 63.4533,
+      "step": 76
+    },
+    {
+      "epoch": 0.006844444444444445,
+      "grad_norm": 62.675838470458984,
+      "learning_rate": 3e-06,
+      "loss": 54.1729,
+      "step": 77
+    },
+    {
+      "epoch": 0.006933333333333333,
+      "grad_norm": 63.28793716430664,
+      "learning_rate": 3e-06,
+      "loss": 61.2604,
+      "step": 78
+    },
+    {
+      "epoch": 0.007022222222222222,
+      "grad_norm": 75.33735656738281,
+      "learning_rate": 3e-06,
+      "loss": 59.0054,
+      "step": 79
+    },
+    {
+      "epoch": 0.0071111111111111115,
+      "grad_norm": 86.537109375,
+      "learning_rate": 3e-06,
+      "loss": 66.586,
+      "step": 80
+    },
+    {
+      "epoch": 0.0072,
+      "grad_norm": 66.0783462524414,
+      "learning_rate": 3e-06,
+      "loss": 59.9151,
+      "step": 81
+    },
+    {
+      "epoch": 0.007288888888888889,
+      "grad_norm": 66.10869598388672,
+      "learning_rate": 3e-06,
+      "loss": 62.0304,
+      "step": 82
+    },
+    {
+      "epoch": 0.007377777777777777,
+      "grad_norm": 58.026912689208984,
+      "learning_rate": 3e-06,
+      "loss": 53.3198,
+      "step": 83
+    },
+    {
+      "epoch": 0.007466666666666667,
+      "grad_norm": 59.65370559692383,
+      "learning_rate": 3e-06,
+      "loss": 60.237,
+      "step": 84
+    },
+    {
+      "completion_length": 246.70833587646484,
+      "epoch": 0.007555555555555556,
+      "grad_norm": 38.2843017578125,
+      "learning_rate": 3e-06,
+      "loss": -3.4492,
+      "reward": 0.9270833730697632,
+      "reward_std": 0.2587623968720436,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6354166567325592,
+      "rewards/correctness_reward_func_math": 0.2916666679084301,
+      "step": 85,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.007644444444444444,
+      "grad_norm": 42.626834869384766,
+      "learning_rate": 3e-06,
+      "loss": -3.0457,
+      "step": 86
+    },
+    {
+      "epoch": 0.007733333333333333,
+      "grad_norm": 31.817684173583984,
+      "learning_rate": 3e-06,
+      "loss": 0.2054,
+      "step": 87
+    },
+    {
+      "epoch": 0.007822222222222222,
+      "grad_norm": 41.712833404541016,
+      "learning_rate": 3e-06,
+      "loss": 0.6522,
+      "step": 88
+    },
+    {
+      "epoch": 0.007911111111111112,
+      "grad_norm": 33.385929107666016,
+      "learning_rate": 3e-06,
+      "loss": -2.3715,
+      "step": 89
+    },
+    {
+      "epoch": 0.008,
+      "grad_norm": 43.1032829284668,
+      "learning_rate": 3e-06,
+      "loss": 1.5502,
+      "step": 90
+    },
+    {
+      "epoch": 0.008088888888888889,
+      "grad_norm": 36.241458892822266,
+      "learning_rate": 3e-06,
+      "loss": -3.5684,
+      "step": 91
+    },
+    {
+      "epoch": 0.008177777777777779,
+      "grad_norm": 41.06986618041992,
+      "learning_rate": 3e-06,
+      "loss": -3.2263,
+      "step": 92
+    },
+    {
+      "epoch": 0.008266666666666667,
+      "grad_norm": 31.25284767150879,
+      "learning_rate": 3e-06,
+      "loss": -0.3487,
+      "step": 93
+    },
+    {
+      "epoch": 0.008355555555555555,
+      "grad_norm": 36.958518981933594,
+      "learning_rate": 3e-06,
+      "loss": 0.1995,
+      "step": 94
+    },
+    {
+      "epoch": 0.008444444444444444,
+      "grad_norm": 34.949676513671875,
+      "learning_rate": 3e-06,
+      "loss": -2.9378,
+      "step": 95
+    },
+    {
+      "epoch": 0.008533333333333334,
+      "grad_norm": 36.523372650146484,
+      "learning_rate": 3e-06,
+      "loss": 0.9469,
+      "step": 96
+    },
+    {
+      "completion_length": 246.08333587646484,
+      "epoch": 0.008622222222222222,
+      "grad_norm": 59.17626953125,
+      "learning_rate": 3e-06,
+      "loss": 2.6896,
+      "reward": 1.3333333730697632,
+      "reward_std": 0.5695068836212158,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5833333432674408,
+      "rewards/correctness_reward_func_math": 0.7500000149011612,
+      "step": 97,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.00871111111111111,
+      "grad_norm": 71.58135223388672,
+      "learning_rate": 3e-06,
+      "loss": 1.0704,
+      "step": 98
+    },
+    {
+      "epoch": 0.0088,
+      "grad_norm": 65.36974334716797,
+      "learning_rate": 3e-06,
+      "loss": -2.7445,
+      "step": 99
+    },
+    {
+      "epoch": 0.008888888888888889,
+      "grad_norm": 60.50218200683594,
+      "learning_rate": 3e-06,
+      "loss": 3.606,
+      "step": 100
+    },
+    {
+      "epoch": 0.008977777777777777,
+      "grad_norm": 61.99585723876953,
+      "learning_rate": 3e-06,
+      "loss": -1.4435,
+      "step": 101
+    },
+    {
+      "epoch": 0.009066666666666667,
+      "grad_norm": 106.92288970947266,
+      "learning_rate": 3e-06,
+      "loss": -3.972,
+      "step": 102
+    },
+    {
+      "epoch": 0.009155555555555556,
+      "grad_norm": 58.85340118408203,
+      "learning_rate": 3e-06,
+      "loss": 1.5567,
+      "step": 103
+    },
+    {
+      "epoch": 0.009244444444444444,
+      "grad_norm": 70.97467041015625,
+      "learning_rate": 3e-06,
+      "loss": -0.3592,
+      "step": 104
+    },
+    {
+      "epoch": 0.009333333333333334,
+      "grad_norm": 62.310516357421875,
+      "learning_rate": 3e-06,
+      "loss": -4.0897,
+      "step": 105
+    },
+    {
+      "epoch": 0.009422222222222222,
+      "grad_norm": 60.98678207397461,
+      "learning_rate": 3e-06,
+      "loss": 2.9803,
+      "step": 106
+    },
+    {
+      "epoch": 0.00951111111111111,
+      "grad_norm": 60.23484420776367,
+      "learning_rate": 3e-06,
+      "loss": -2.2749,
+      "step": 107
+    },
+    {
+      "epoch": 0.0096,
+      "grad_norm": 58.0914192199707,
+      "learning_rate": 3e-06,
+      "loss": -4.9462,
+      "step": 108
+    },
+    {
+      "completion_length": 249.06250762939453,
+      "epoch": 0.00968888888888889,
+      "grad_norm": 65.09230041503906,
+      "learning_rate": 3e-06,
+      "loss": -11.6324,
+      "reward": 1.125,
+      "reward_std": 0.5275504291057587,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5833333432674408,
+      "rewards/correctness_reward_func_math": 0.5416666716337204,
+      "step": 109,
+      "zero_std_ratio": 0.0
+    },
+    {
+      "epoch": 0.009777777777777778,
+      "grad_norm": 58.50445556640625,
+      "learning_rate": 3e-06,
+      "loss": -12.2664,
+      "step": 110
+    },
+    {
+      "epoch": 0.009866666666666666,
+      "grad_norm": 53.459251403808594,
+      "learning_rate": 3e-06,
+      "loss": -7.2192,
+      "step": 111
+    },
+    {
+      "epoch": 0.009955555555555556,
+      "grad_norm": 60.34041213989258,
+      "learning_rate": 3e-06,
+      "loss": -6.9971,
+      "step": 112
+    },
+    {
+      "epoch": 0.010044444444444444,
+      "grad_norm": 61.72711944580078,
+      "learning_rate": 3e-06,
+      "loss": -0.4686,
+      "step": 113
+    },
+    {
+      "epoch": 0.010133333333333333,
+      "grad_norm": 96.1756591796875,
+      "learning_rate": 3e-06,
+      "loss": -7.4161,
+      "step": 114
+    },
+    {
+      "epoch": 0.010222222222222223,
+      "grad_norm": 61.3508415222168,
+      "learning_rate": 3e-06,
+      "loss": -12.3763,
+      "step": 115
+    },
+    {
+      "epoch": 0.010311111111111111,
+      "grad_norm": 55.424896240234375,
+      "learning_rate": 3e-06,
+      "loss": -12.8949,
+      "step": 116
+    },
+    {
+      "epoch": 0.0104,
+      "grad_norm": 56.08291244506836,
+      "learning_rate": 3e-06,
+      "loss": -7.8472,
+      "step": 117
+    },
+    {
+      "epoch": 0.01048888888888889,
+      "grad_norm": 73.18891906738281,
+      "learning_rate": 3e-06,
+      "loss": -8.0281,
+      "step": 118
+    },
+    {
+      "epoch": 0.010577777777777778,
+      "grad_norm": 64.47604370117188,
+      "learning_rate": 3e-06,
+      "loss": -1.3444,
+      "step": 119
+    },
+    {
+      "epoch": 0.010666666666666666,
+      "grad_norm": 96.0721664428711,
+      "learning_rate": 3e-06,
+      "loss": -8.5737,
+      "step": 120
+    },
+    {
+      "completion_length": 253.0,
+      "epoch": 0.010755555555555556,
+      "grad_norm": 60.78779983520508,
+      "learning_rate": 3e-06,
+      "loss": -1.605,
+      "reward": 1.1145833730697632,
+      "reward_std": 0.4272044152021408,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.4999999850988388,
+      "step": 121,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.010844444444444445,
+      "grad_norm": 49.34260177612305,
+      "learning_rate": 3e-06,
+      "loss": -0.2147,
+      "step": 122
+    },
+    {
+      "epoch": 0.010933333333333333,
+      "grad_norm": 53.38318634033203,
+      "learning_rate": 3e-06,
+      "loss": -7.1697,
+      "step": 123
+    },
+    {
+      "epoch": 0.011022222222222221,
+      "grad_norm": 84.88465881347656,
+      "learning_rate": 3e-06,
+      "loss": 4.7029,
+      "step": 124
+    },
+    {
+      "epoch": 0.011111111111111112,
+      "grad_norm": 50.966583251953125,
+      "learning_rate": 3e-06,
+      "loss": -5.2481,
+      "step": 125
+    },
+    {
+      "epoch": 0.0112,
+      "grad_norm": 64.3619155883789,
+      "learning_rate": 3e-06,
+      "loss": -5.9545,
+      "step": 126
+    },
+    {
+      "epoch": 0.011288888888888888,
+      "grad_norm": 60.359500885009766,
+      "learning_rate": 3e-06,
+      "loss": -1.8244,
+      "step": 127
+    },
+    {
+      "epoch": 0.011377777777777778,
+      "grad_norm": 51.08177947998047,
+      "learning_rate": 3e-06,
+      "loss": -0.8292,
+      "step": 128
+    },
+    {
+      "epoch": 0.011466666666666667,
+      "grad_norm": 53.191165924072266,
+      "learning_rate": 3e-06,
+      "loss": -7.7867,
+      "step": 129
+    },
+    {
+      "epoch": 0.011555555555555555,
+      "grad_norm": 87.42491912841797,
+      "learning_rate": 3e-06,
+      "loss": 3.92,
+      "step": 130
+    },
+    {
+      "epoch": 0.011644444444444445,
+      "grad_norm": 49.99729537963867,
+      "learning_rate": 3e-06,
+      "loss": -5.9859,
+      "step": 131
+    },
+    {
+      "epoch": 0.011733333333333333,
+      "grad_norm": 45.14487075805664,
+      "learning_rate": 3e-06,
+      "loss": -6.6928,
+      "step": 132
+    },
+    {
+      "completion_length": 239.1666717529297,
+      "epoch": 0.011822222222222222,
+      "grad_norm": 68.43509674072266,
+      "learning_rate": 3e-06,
+      "loss": 45.5812,
+      "reward": 1.1666666865348816,
+      "reward_std": 0.47104020416736603,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.5416666679084301,
+      "step": 133,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.011911111111111112,
+      "grad_norm": 62.809059143066406,
+      "learning_rate": 3e-06,
+      "loss": 47.1017,
+      "step": 134
+    },
+    {
+      "epoch": 0.012,
+      "grad_norm": 61.8614387512207,
+      "learning_rate": 3e-06,
+      "loss": 41.2505,
+      "step": 135
+    },
+    {
+      "epoch": 0.012088888888888889,
+      "grad_norm": 65.46350860595703,
+      "learning_rate": 3e-06,
+      "loss": 44.191,
+      "step": 136
+    },
+    {
+      "epoch": 0.012177777777777777,
+      "grad_norm": 59.1669807434082,
+      "learning_rate": 3e-06,
+      "loss": 36.6508,
+      "step": 137
+    },
+    {
+      "epoch": 0.012266666666666667,
+      "grad_norm": 55.44610595703125,
+      "learning_rate": 3e-06,
+      "loss": 41.1041,
+      "step": 138
+    },
+    {
+      "epoch": 0.012355555555555555,
+      "grad_norm": 61.210411071777344,
+      "learning_rate": 3e-06,
+      "loss": 44.6212,
+      "step": 139
+    },
+    {
+      "epoch": 0.012444444444444444,
+      "grad_norm": 64.934326171875,
+      "learning_rate": 3e-06,
+      "loss": 46.5024,
+      "step": 140
+    },
+    {
+      "epoch": 0.012533333333333334,
+      "grad_norm": 67.7354507446289,
+      "learning_rate": 3e-06,
+      "loss": 40.309,
+      "step": 141
+    },
+    {
+      "epoch": 0.012622222222222222,
+      "grad_norm": 69.55413055419922,
+      "learning_rate": 3e-06,
+      "loss": 42.8161,
+      "step": 142
+    },
+    {
+      "epoch": 0.01271111111111111,
+      "grad_norm": 59.040592193603516,
+      "learning_rate": 3e-06,
+      "loss": 35.3869,
+      "step": 143
+    },
+    {
+      "epoch": 0.0128,
+      "grad_norm": 56.21048355102539,
+      "learning_rate": 3e-06,
+      "loss": 39.9197,
+      "step": 144
+    },
+    {
+      "completion_length": 237.37500762939453,
+      "epoch": 0.012888888888888889,
+      "grad_norm": 67.88895416259766,
+      "learning_rate": 3e-06,
+      "loss": -21.4223,
+      "reward": 0.8958333730697632,
+      "reward_std": 0.44294705986976624,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5625000149011612,
+      "rewards/correctness_reward_func_math": 0.3333333358168602,
+      "step": 145,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.012977777777777777,
+      "grad_norm": 55.02178955078125,
+      "learning_rate": 3e-06,
+      "loss": -26.3881,
+      "step": 146
+    },
+    {
+      "epoch": 0.013066666666666667,
+      "grad_norm": 103.78085327148438,
+      "learning_rate": 3e-06,
+      "loss": -21.7028,
+      "step": 147
+    },
+    {
+      "epoch": 0.013155555555555556,
+      "grad_norm": 62.1268196105957,
+      "learning_rate": 3e-06,
+      "loss": -19.048,
+      "step": 148
+    },
+    {
+      "epoch": 0.013244444444444444,
+      "grad_norm": 57.99726486206055,
+      "learning_rate": 3e-06,
+      "loss": -19.346,
+      "step": 149
+    },
+    {
+      "epoch": 0.013333333333333334,
+      "grad_norm": 58.639549255371094,
+      "learning_rate": 3e-06,
+      "loss": -25.0216,
+      "step": 150
+    },
+    {
+      "epoch": 0.013422222222222223,
+      "grad_norm": 75.58393859863281,
+      "learning_rate": 3e-06,
+      "loss": -21.7941,
+      "step": 151
+    },
+    {
+      "epoch": 0.013511111111111111,
+      "grad_norm": 54.83882522583008,
+      "learning_rate": 3e-06,
+      "loss": -27.6056,
+      "step": 152
+    },
+    {
+      "epoch": 0.0136,
+      "grad_norm": 70.61170196533203,
+      "learning_rate": 3e-06,
+      "loss": -21.99,
+      "step": 153
+    },
+    {
+      "epoch": 0.01368888888888889,
+      "grad_norm": 68.1909408569336,
+      "learning_rate": 3e-06,
+      "loss": -20.2119,
+      "step": 154
+    },
+    {
+      "epoch": 0.013777777777777778,
+      "grad_norm": 68.70491027832031,
+      "learning_rate": 3e-06,
+      "loss": -20.2249,
+      "step": 155
+    },
+    {
+      "epoch": 0.013866666666666666,
+      "grad_norm": 55.29183578491211,
+      "learning_rate": 3e-06,
+      "loss": -25.9634,
+      "step": 156
+    },
+    {
+      "completion_length": 250.6666717529297,
+      "epoch": 0.013955555555555556,
+      "grad_norm": 70.28712463378906,
+      "learning_rate": 3e-06,
+      "loss": -13.869,
+      "reward": 1.4583333730697632,
+      "reward_std": 0.6823203265666962,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.8333333283662796,
+      "step": 157,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.014044444444444444,
+      "grad_norm": 60.210201263427734,
+      "learning_rate": 3e-06,
+      "loss": -24.2593,
+      "step": 158
+    },
+    {
+      "epoch": 0.014133333333333333,
+      "grad_norm": 77.62222290039062,
+      "learning_rate": 3e-06,
+      "loss": -11.4696,
+      "step": 159
+    },
+    {
+      "epoch": 0.014222222222222223,
+      "grad_norm": 70.80023193359375,
+      "learning_rate": 3e-06,
+      "loss": -25.8617,
+      "step": 160
+    },
+    {
+      "epoch": 0.014311111111111111,
+      "grad_norm": 64.7750244140625,
+      "learning_rate": 3e-06,
+      "loss": -14.8635,
+      "step": 161
+    },
+    {
+      "epoch": 0.0144,
+      "grad_norm": 77.83097076416016,
+      "learning_rate": 3e-06,
+      "loss": -7.727,
+      "step": 162
+    },
+    {
+      "epoch": 0.01448888888888889,
+      "grad_norm": 79.27497100830078,
+      "learning_rate": 3e-06,
+      "loss": -14.3689,
+      "step": 163
+    },
+    {
+      "epoch": 0.014577777777777778,
+      "grad_norm": 78.7293472290039,
+      "learning_rate": 3e-06,
+      "loss": -24.9423,
+      "step": 164
+    },
+    {
+      "epoch": 0.014666666666666666,
+      "grad_norm": 70.24745178222656,
+      "learning_rate": 3e-06,
+      "loss": -12.4543,
+      "step": 165
+    },
+    {
+      "epoch": 0.014755555555555555,
+      "grad_norm": 75.36212158203125,
+      "learning_rate": 3e-06,
+      "loss": -26.7206,
+      "step": 166
+    },
+    {
+      "epoch": 0.014844444444444445,
+      "grad_norm": 65.05477142333984,
+      "learning_rate": 3e-06,
+      "loss": -16.0018,
+      "step": 167
+    },
+    {
+      "epoch": 0.014933333333333333,
+      "grad_norm": 78.83174133300781,
+      "learning_rate": 3e-06,
+      "loss": -8.9618,
+      "step": 168
+    },
+    {
+      "completion_length": 236.2916717529297,
+      "epoch": 0.015022222222222222,
+      "grad_norm": 82.83967590332031,
+      "learning_rate": 3e-06,
+      "loss": -0.5024,
+      "reward": 1.1145833730697632,
+      "reward_std": 0.3740755543112755,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833134651184,
+      "rewards/correctness_reward_func_math": 0.5,
+      "step": 169,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.015111111111111112,
+      "grad_norm": 51.84051513671875,
+      "learning_rate": 3e-06,
+      "loss": 6.7446,
+      "step": 170
+    },
+    {
+      "epoch": 0.0152,
+      "grad_norm": 63.607723236083984,
+      "learning_rate": 3e-06,
+      "loss": 2.7771,
+      "step": 171
+    },
+    {
+      "epoch": 0.015288888888888888,
+      "grad_norm": 52.88029479980469,
+      "learning_rate": 3e-06,
+      "loss": 8.1945,
+      "step": 172
+    },
+    {
+      "epoch": 0.015377777777777778,
+      "grad_norm": 68.90487670898438,
+      "learning_rate": 3e-06,
+      "loss": 0.8609,
+      "step": 173
+    },
+    {
+      "epoch": 0.015466666666666667,
+      "grad_norm": 57.66716766357422,
+      "learning_rate": 3e-06,
+      "loss": -0.5103,
+      "step": 174
+    },
+    {
+      "epoch": 0.015555555555555555,
+      "grad_norm": 69.48858642578125,
+      "learning_rate": 3e-06,
+      "loss": -0.89,
+      "step": 175
+    },
+    {
+      "epoch": 0.015644444444444443,
+      "grad_norm": 51.13008117675781,
+      "learning_rate": 3e-06,
+      "loss": 5.8779,
+      "step": 176
+    },
+    {
+      "epoch": 0.015733333333333332,
+      "grad_norm": 61.48530578613281,
+      "learning_rate": 3e-06,
+      "loss": 2.0727,
+      "step": 177
+    },
+    {
+      "epoch": 0.015822222222222224,
+      "grad_norm": 55.415924072265625,
+      "learning_rate": 3e-06,
+      "loss": 7.6559,
+      "step": 178
+    },
+    {
+      "epoch": 0.015911111111111112,
+      "grad_norm": 65.15290069580078,
+      "learning_rate": 3e-06,
+      "loss": -0.6101,
+      "step": 179
+    },
+    {
+      "epoch": 0.016,
+      "grad_norm": 52.03913879394531,
+      "learning_rate": 3e-06,
+      "loss": -1.3899,
+      "step": 180
+    },
+    {
+      "completion_length": 250.14583587646484,
+      "epoch": 0.01608888888888889,
+      "grad_norm": 63.963829040527344,
+      "learning_rate": 3e-06,
+      "loss": -8.994,
+      "reward": 0.885416716337204,
+      "reward_std": 0.3302172925323248,
+      "rewards/boxed_and_answer_tags_format_reward": 0.59375,
+      "rewards/correctness_reward_func_math": 0.2916666567325592,
+      "step": 181,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.016177777777777777,
+      "grad_norm": 48.50006866455078,
+      "learning_rate": 3e-06,
+      "loss": -0.9428,
+      "step": 182
+    },
+    {
+      "epoch": 0.016266666666666665,
+      "grad_norm": 58.21607971191406,
+      "learning_rate": 3e-06,
+      "loss": -8.2051,
+      "step": 183
+    },
+    {
+      "epoch": 0.016355555555555557,
+      "grad_norm": 76.80998992919922,
+      "learning_rate": 3e-06,
+      "loss": 1.6849,
+      "step": 184
+    },
+    {
+      "epoch": 0.016444444444444446,
+      "grad_norm": 48.460941314697266,
+      "learning_rate": 3e-06,
+      "loss": -3.4021,
+      "step": 185
+    },
+    {
+      "epoch": 0.016533333333333334,
+      "grad_norm": 55.28091049194336,
+      "learning_rate": 3e-06,
+      "loss": 0.4156,
+      "step": 186
+    },
+    {
+      "epoch": 0.016622222222222222,
+      "grad_norm": 65.21077728271484,
+      "learning_rate": 3e-06,
+      "loss": -9.4278,
+      "step": 187
+    },
+    {
+      "epoch": 0.01671111111111111,
+      "grad_norm": 50.71424865722656,
+      "learning_rate": 3e-06,
+      "loss": -2.0682,
+      "step": 188
+    },
+    {
+      "epoch": 0.0168,
+      "grad_norm": 58.02372360229492,
+      "learning_rate": 3e-06,
+      "loss": -9.1068,
+      "step": 189
+    },
+    {
+      "epoch": 0.016888888888888887,
+      "grad_norm": 61.12031555175781,
+      "learning_rate": 3e-06,
+      "loss": 0.679,
+      "step": 190
+    },
+    {
+      "epoch": 0.01697777777777778,
+      "grad_norm": 51.7930908203125,
+      "learning_rate": 3e-06,
+      "loss": -4.3214,
+      "step": 191
+    },
+    {
+      "epoch": 0.017066666666666667,
+      "grad_norm": 48.15507507324219,
+      "learning_rate": 3e-06,
+      "loss": -0.8545,
+      "step": 192
+    },
+    {
+      "completion_length": 248.1666717529297,
+      "epoch": 0.017155555555555556,
+      "grad_norm": 62.317527770996094,
+      "learning_rate": 3e-06,
+      "loss": 13.8287,
+      "reward": 1.0729167461395264,
+      "reward_std": 0.348264142870903,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5729166865348816,
+      "rewards/correctness_reward_func_math": 0.5,
+      "step": 193,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.017244444444444444,
+      "grad_norm": 43.339691162109375,
+      "learning_rate": 3e-06,
+      "loss": 17.523,
+      "step": 194
+    },
+    {
+      "epoch": 0.017333333333333333,
+      "grad_norm": 48.14270782470703,
+      "learning_rate": 3e-06,
+      "loss": 15.9481,
+      "step": 195
+    },
+    {
+      "epoch": 0.01742222222222222,
+      "grad_norm": 43.32905960083008,
+      "learning_rate": 3e-06,
+      "loss": 14.7259,
+      "step": 196
+    },
+    {
+      "epoch": 0.017511111111111113,
+      "grad_norm": 45.01740264892578,
+      "learning_rate": 3e-06,
+      "loss": 13.8658,
+      "step": 197
+    },
+    {
+      "epoch": 0.0176,
+      "grad_norm": 43.2428092956543,
+      "learning_rate": 3e-06,
+      "loss": 20.4664,
+      "step": 198
+    },
+    {
+      "epoch": 0.01768888888888889,
+      "grad_norm": 56.058616638183594,
+      "learning_rate": 3e-06,
+      "loss": 13.1828,
+      "step": 199
+    },
+    {
+      "epoch": 0.017777777777777778,
+      "grad_norm": 46.968666076660156,
+      "learning_rate": 3e-06,
+      "loss": 16.8591,
+      "step": 200
+    },
+    {
+      "epoch": 0.017866666666666666,
+      "grad_norm": 45.98298263549805,
+      "learning_rate": 3e-06,
+      "loss": 15.5154,
+      "step": 201
+    },
+    {
+      "epoch": 0.017955555555555554,
+      "grad_norm": 43.91643142700195,
+      "learning_rate": 3e-06,
+      "loss": 14.2328,
+      "step": 202
+    },
+    {
+      "epoch": 0.018044444444444443,
+      "grad_norm": 44.83538055419922,
+      "learning_rate": 3e-06,
+      "loss": 13.1929,
+      "step": 203
+    },
+    {
+      "epoch": 0.018133333333333335,
+      "grad_norm": 42.83240509033203,
+      "learning_rate": 3e-06,
+      "loss": 19.8936,
+      "step": 204
+    },
+    {
+      "completion_length": 249.02083587646484,
+      "epoch": 0.018222222222222223,
+      "grad_norm": 119.68338775634766,
+      "learning_rate": 3e-06,
+      "loss": -4.5616,
+      "reward": 0.8229166865348816,
+      "reward_std": 0.28067073225975037,
+      "rewards/boxed_and_answer_tags_format_reward": 0.65625,
+      "rewards/correctness_reward_func_math": 0.1666666716337204,
+      "step": 205,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.01831111111111111,
+      "grad_norm": 40.65678405761719,
+      "learning_rate": 3e-06,
+      "loss": 3.7784,
+      "step": 206
+    },
+    {
+      "epoch": 0.0184,
+      "grad_norm": 51.84949493408203,
+      "learning_rate": 3e-06,
+      "loss": 2.132,
+      "step": 207
+    },
+    {
+      "epoch": 0.018488888888888888,
+      "grad_norm": 40.80442428588867,
+      "learning_rate": 3e-06,
+      "loss": -1.7568,
+      "step": 208
+    },
+    {
+      "epoch": 0.018577777777777776,
+      "grad_norm": 51.88225555419922,
+      "learning_rate": 3e-06,
+      "loss": -2.808,
+      "step": 209
+    },
+    {
+      "epoch": 0.018666666666666668,
+      "grad_norm": 57.230106353759766,
+      "learning_rate": 3e-06,
+      "loss": -1.6958,
+      "step": 210
+    },
+    {
+      "epoch": 0.018755555555555557,
+      "grad_norm": 65.36343383789062,
+      "learning_rate": 3e-06,
+      "loss": -5.016,
+      "step": 211
+    },
+    {
+      "epoch": 0.018844444444444445,
+      "grad_norm": 42.36751937866211,
+      "learning_rate": 3e-06,
+      "loss": 3.2604,
+      "step": 212
+    },
+    {
+      "epoch": 0.018933333333333333,
+      "grad_norm": 54.347625732421875,
+      "learning_rate": 3e-06,
+      "loss": 1.5256,
+      "step": 213
+    },
+    {
+      "epoch": 0.01902222222222222,
+      "grad_norm": 40.971683502197266,
+      "learning_rate": 3e-06,
+      "loss": -2.3704,
+      "step": 214
+    },
+    {
+      "epoch": 0.01911111111111111,
+      "grad_norm": 51.366546630859375,
+      "learning_rate": 3e-06,
+      "loss": -3.7841,
+      "step": 215
+    },
+    {
+      "epoch": 0.0192,
+      "grad_norm": 64.25231170654297,
+      "learning_rate": 3e-06,
+      "loss": -2.6423,
+      "step": 216
+    },
+    {
+      "completion_length": 250.4791717529297,
+      "epoch": 0.01928888888888889,
+      "grad_norm": 60.69169235229492,
+      "learning_rate": 3e-06,
+      "loss": 3.7848,
+      "reward": 1.0833333730697632,
+      "reward_std": 0.5039487332105637,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.4583333283662796,
+      "step": 217,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.01937777777777778,
+      "grad_norm": 65.37804412841797,
+      "learning_rate": 3e-06,
+      "loss": -5.4996,
+      "step": 218
+    },
+    {
+      "epoch": 0.019466666666666667,
+      "grad_norm": 58.69138717651367,
+      "learning_rate": 3e-06,
+      "loss": 0.3025,
+      "step": 219
+    },
+    {
+      "epoch": 0.019555555555555555,
+      "grad_norm": 72.17839813232422,
+      "learning_rate": 3e-06,
+      "loss": -0.8041,
+      "step": 220
+    },
+    {
+      "epoch": 0.019644444444444444,
+      "grad_norm": 69.56704711914062,
+      "learning_rate": 3e-06,
+      "loss": 3.1412,
+      "step": 221
+    },
+    {
+      "epoch": 0.019733333333333332,
+      "grad_norm": 64.57500457763672,
+      "learning_rate": 3e-06,
+      "loss": 6.7244,
+      "step": 222
+    },
+    {
+      "epoch": 0.019822222222222224,
+      "grad_norm": 65.06715393066406,
+      "learning_rate": 3e-06,
+      "loss": 3.0694,
+      "step": 223
+    },
+    {
+      "epoch": 0.019911111111111112,
+      "grad_norm": 72.74304962158203,
+      "learning_rate": 3e-06,
+      "loss": -5.928,
+      "step": 224
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 62.06201934814453,
+      "learning_rate": 3e-06,
+      "loss": -0.2234,
+      "step": 225
+    },
+    {
+      "epoch": 0.02008888888888889,
+      "grad_norm": 74.25010681152344,
+      "learning_rate": 3e-06,
+      "loss": -1.5055,
+      "step": 226
+    },
+    {
+      "epoch": 0.020177777777777777,
+      "grad_norm": 64.32748413085938,
+      "learning_rate": 3e-06,
+      "loss": 2.4819,
+      "step": 227
+    },
+    {
+      "epoch": 0.020266666666666665,
+      "grad_norm": 64.75834655761719,
+      "learning_rate": 3e-06,
+      "loss": 5.504,
+      "step": 228
+    },
+    {
+      "completion_length": 252.9375,
+      "epoch": 0.020355555555555557,
+      "grad_norm": 86.62522888183594,
+      "learning_rate": 3e-06,
+      "loss": 14.2874,
+      "reward": 1.1562500596046448,
+      "reward_std": 0.5227071046829224,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.4166666567325592,
+      "step": 229,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.020444444444444446,
+      "grad_norm": 61.35566329956055,
+      "learning_rate": 3e-06,
+      "loss": 17.7162,
+      "step": 230
+    },
+    {
+      "epoch": 0.020533333333333334,
+      "grad_norm": 61.87510681152344,
+      "learning_rate": 3e-06,
+      "loss": 8.9931,
+      "step": 231
+    },
+    {
+      "epoch": 0.020622222222222222,
+      "grad_norm": 57.673770904541016,
+      "learning_rate": 3e-06,
+      "loss": 14.8206,
+      "step": 232
+    },
+    {
+      "epoch": 0.02071111111111111,
+      "grad_norm": 64.32942199707031,
+      "learning_rate": 3e-06,
+      "loss": 10.9396,
+      "step": 233
+    },
+    {
+      "epoch": 0.0208,
+      "grad_norm": 66.23136138916016,
+      "learning_rate": 3e-06,
+      "loss": 15.6923,
+      "step": 234
+    },
+    {
+      "epoch": 0.020888888888888887,
+      "grad_norm": 74.55809783935547,
+      "learning_rate": 3e-06,
+      "loss": 13.7889,
+      "step": 235
+    },
+    {
+      "epoch": 0.02097777777777778,
+      "grad_norm": 60.680240631103516,
+      "learning_rate": 3e-06,
+      "loss": 17.2098,
+      "step": 236
+    },
+    {
+      "epoch": 0.021066666666666668,
+      "grad_norm": 63.526371002197266,
+      "learning_rate": 3e-06,
+      "loss": 8.3714,
+      "step": 237
+    },
+    {
+      "epoch": 0.021155555555555556,
+      "grad_norm": 60.387813568115234,
+      "learning_rate": 3e-06,
+      "loss": 13.9642,
+      "step": 238
+    },
+    {
+      "epoch": 0.021244444444444444,
+      "grad_norm": 65.6108169555664,
+      "learning_rate": 3e-06,
+      "loss": 10.0042,
+      "step": 239
+    },
+    {
+      "epoch": 0.021333333333333333,
+      "grad_norm": 65.62525177001953,
+      "learning_rate": 3e-06,
+      "loss": 14.6078,
+      "step": 240
+    },
+    {
+      "completion_length": 245.89584350585938,
+      "epoch": 0.02142222222222222,
+      "grad_norm": 99.76964569091797,
+      "learning_rate": 3e-06,
+      "loss": -5.9045,
+      "reward": 1.2916666865348816,
+      "reward_std": 0.23116151243448257,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7083333432674408,
+      "rewards/correctness_reward_func_math": 0.5833333358168602,
+      "step": 241,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.021511111111111113,
+      "grad_norm": 34.929359436035156,
+      "learning_rate": 3e-06,
+      "loss": -5.1064,
+      "step": 242
+    },
+    {
+      "epoch": 0.0216,
+      "grad_norm": 41.406982421875,
+      "learning_rate": 3e-06,
+      "loss": -7.9375,
+      "step": 243
+    },
+    {
+      "epoch": 0.02168888888888889,
+      "grad_norm": 40.73991775512695,
+      "learning_rate": 3e-06,
+      "loss": -4.6122,
+      "step": 244
+    },
+    {
+      "epoch": 0.021777777777777778,
+      "grad_norm": 32.28548812866211,
+      "learning_rate": 3e-06,
+      "loss": -3.2865,
+      "step": 245
+    },
+    {
+      "epoch": 0.021866666666666666,
+      "grad_norm": 37.392860412597656,
+      "learning_rate": 3e-06,
+      "loss": -0.2002,
+      "step": 246
+    },
+    {
+      "epoch": 0.021955555555555555,
+      "grad_norm": 105.95482635498047,
+      "learning_rate": 3e-06,
+      "loss": -5.7761,
+      "step": 247
+    },
+    {
+      "epoch": 0.022044444444444443,
+      "grad_norm": 35.37491226196289,
+      "learning_rate": 3e-06,
+      "loss": -5.3203,
+      "step": 248
+    },
+    {
+      "epoch": 0.022133333333333335,
+      "grad_norm": 37.672000885009766,
+      "learning_rate": 3e-06,
+      "loss": -8.3175,
+      "step": 249
+    },
+    {
+      "epoch": 0.022222222222222223,
+      "grad_norm": 34.36002731323242,
+      "learning_rate": 3e-06,
+      "loss": -4.8629,
+      "step": 250
+    },
+    {
+      "epoch": 0.02231111111111111,
+      "grad_norm": 35.60414123535156,
+      "learning_rate": 3e-06,
+      "loss": -3.8192,
+      "step": 251
+    },
+    {
+      "epoch": 0.0224,
+      "grad_norm": 38.58955764770508,
+      "learning_rate": 3e-06,
+      "loss": -0.897,
+      "step": 252
+    },
+    {
+      "completion_length": 250.1875,
+      "epoch": 0.022488888888888888,
+      "grad_norm": 71.29794311523438,
+      "learning_rate": 3e-06,
+      "loss": 2.5796,
+      "reward": 0.8645833432674408,
+      "reward_std": 0.3201860636472702,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5729166716337204,
+      "rewards/correctness_reward_func_math": 0.2916666641831398,
+      "step": 253,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.022577777777777776,
+      "grad_norm": 62.11003112792969,
+      "learning_rate": 3e-06,
+      "loss": 2.4232,
+      "step": 254
+    },
+    {
+      "epoch": 0.02266666666666667,
+      "grad_norm": 57.18949508666992,
+      "learning_rate": 3e-06,
+      "loss": 5.9388,
+      "step": 255
+    },
+    {
+      "epoch": 0.022755555555555557,
+      "grad_norm": 60.49555206298828,
+      "learning_rate": 3e-06,
+      "loss": 5.5698,
+      "step": 256
+    },
+    {
+      "epoch": 0.022844444444444445,
+      "grad_norm": 134.1082305908203,
+      "learning_rate": 3e-06,
+      "loss": -5.3771,
+      "step": 257
+    },
+    {
+      "epoch": 0.022933333333333333,
+      "grad_norm": 170.15768432617188,
+      "learning_rate": 3e-06,
+      "loss": 6.3811,
+      "step": 258
+    },
+    {
+      "epoch": 0.02302222222222222,
+      "grad_norm": 70.64490509033203,
+      "learning_rate": 3e-06,
+      "loss": 1.7661,
+      "step": 259
+    },
+    {
+      "epoch": 0.02311111111111111,
+      "grad_norm": 71.96417999267578,
+      "learning_rate": 3e-06,
+      "loss": 0.8909,
+      "step": 260
+    },
+    {
+      "epoch": 0.0232,
+      "grad_norm": 58.19865417480469,
+      "learning_rate": 3e-06,
+      "loss": 5.1442,
+      "step": 261
+    },
+    {
+      "epoch": 0.02328888888888889,
+      "grad_norm": 61.813690185546875,
+      "learning_rate": 3e-06,
+      "loss": 4.1458,
+      "step": 262
+    },
+    {
+      "epoch": 0.02337777777777778,
+      "grad_norm": 63.21968460083008,
+      "learning_rate": 3e-06,
+      "loss": -6.5992,
+      "step": 263
+    },
+    {
+      "epoch": 0.023466666666666667,
+      "grad_norm": 130.61351013183594,
+      "learning_rate": 3e-06,
+      "loss": 4.4745,
+      "step": 264
+    },
+    {
+      "completion_length": 254.58333587646484,
+      "epoch": 0.023555555555555555,
+      "grad_norm": 37.4125862121582,
+      "learning_rate": 3e-06,
+      "loss": 6.5013,
+      "reward": 1.0000000298023224,
+      "reward_std": 0.32049281150102615,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7083333134651184,
+      "rewards/correctness_reward_func_math": 0.2916666679084301,
+      "step": 265,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.023644444444444444,
+      "grad_norm": 47.14967346191406,
+      "learning_rate": 3e-06,
+      "loss": 0.9939,
+      "step": 266
+    },
+    {
+      "epoch": 0.023733333333333332,
+      "grad_norm": 52.939048767089844,
+      "learning_rate": 3e-06,
+      "loss": -0.9139,
+      "step": 267
+    },
+    {
+      "epoch": 0.023822222222222224,
+      "grad_norm": 86.99070739746094,
+      "learning_rate": 3e-06,
+      "loss": 8.1755,
+      "step": 268
+    },
+    {
+      "epoch": 0.023911111111111112,
+      "grad_norm": 39.69975280761719,
+      "learning_rate": 3e-06,
+      "loss": 4.7483,
+      "step": 269
+    },
+    {
+      "epoch": 0.024,
+      "grad_norm": 59.52255630493164,
+      "learning_rate": 3e-06,
+      "loss": 1.1483,
+      "step": 270
+    },
+    {
+      "epoch": 0.02408888888888889,
+      "grad_norm": 38.91862106323242,
+      "learning_rate": 3e-06,
+      "loss": 6.09,
+      "step": 271
+    },
+    {
+      "epoch": 0.024177777777777777,
+      "grad_norm": 43.66323471069336,
+      "learning_rate": 3e-06,
+      "loss": 0.4765,
+      "step": 272
+    },
+    {
+      "epoch": 0.024266666666666666,
+      "grad_norm": 44.54389572143555,
+      "learning_rate": 3e-06,
+      "loss": -1.5437,
+      "step": 273
+    },
+    {
+      "epoch": 0.024355555555555554,
+      "grad_norm": 84.41556549072266,
+      "learning_rate": 3e-06,
+      "loss": 7.1028,
+      "step": 274
+    },
+    {
+      "epoch": 0.024444444444444446,
+      "grad_norm": 38.220367431640625,
+      "learning_rate": 3e-06,
+      "loss": 4.0407,
+      "step": 275
+    },
+    {
+      "epoch": 0.024533333333333334,
+      "grad_norm": 45.620452880859375,
+      "learning_rate": 3e-06,
+      "loss": 0.5047,
+      "step": 276
+    },
+    {
+      "completion_length": 243.18750762939453,
+      "epoch": 0.024622222222222222,
+      "grad_norm": 84.81961059570312,
+      "learning_rate": 3e-06,
+      "loss": 5.442,
+      "reward": 1.3333333730697632,
+      "reward_std": 0.4887756109237671,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.7083333432674408,
+      "step": 277,
+      "zero_std_ratio": 0.0
+    },
+    {
+      "epoch": 0.02471111111111111,
+      "grad_norm": 112.86151885986328,
+      "learning_rate": 3e-06,
+      "loss": -1.2528,
+      "step": 278
+    },
+    {
+      "epoch": 0.0248,
+      "grad_norm": 76.52424621582031,
+      "learning_rate": 3e-06,
+      "loss": -4.5325,
+      "step": 279
+    },
+    {
+      "epoch": 0.024888888888888887,
+      "grad_norm": 94.0294189453125,
+      "learning_rate": 3e-06,
+      "loss": -6.6167,
+      "step": 280
+    },
+    {
+      "epoch": 0.02497777777777778,
+      "grad_norm": 78.60155487060547,
+      "learning_rate": 3e-06,
+      "loss": 0.2653,
+      "step": 281
+    },
+    {
+      "epoch": 0.025066666666666668,
+      "grad_norm": 58.42827224731445,
+      "learning_rate": 3e-06,
+      "loss": -5.6591,
+      "step": 282
+    },
+    {
+      "epoch": 0.025155555555555556,
+      "grad_norm": 81.04212188720703,
+      "learning_rate": 3e-06,
+      "loss": 4.2739,
+      "step": 283
+    },
+    {
+      "epoch": 0.025244444444444444,
+      "grad_norm": 67.27478790283203,
+      "learning_rate": 3e-06,
+      "loss": -1.5776,
+      "step": 284
+    },
+    {
+      "epoch": 0.025333333333333333,
+      "grad_norm": 114.41588592529297,
+      "learning_rate": 3e-06,
+      "loss": -5.4532,
+      "step": 285
+    },
+    {
+      "epoch": 0.02542222222222222,
+      "grad_norm": 75.61115264892578,
+      "learning_rate": 3e-06,
+      "loss": -7.4401,
+      "step": 286
+    },
+    {
+      "epoch": 0.02551111111111111,
+      "grad_norm": 236.67214965820312,
+      "learning_rate": 3e-06,
+      "loss": -0.6833,
+      "step": 287
+    },
+    {
+      "epoch": 0.0256,
+      "grad_norm": 59.2407341003418,
+      "learning_rate": 3e-06,
+      "loss": -6.8172,
+      "step": 288
+    },
+    {
+      "completion_length": 250.20834350585938,
+      "epoch": 0.02568888888888889,
+      "grad_norm": 93.99871063232422,
+      "learning_rate": 3e-06,
+      "loss": -12.9867,
+      "reward": 1.4687500596046448,
+      "reward_std": 0.6822589337825775,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.7916666865348816,
+      "step": 289,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.025777777777777778,
+      "grad_norm": 79.78562927246094,
+      "learning_rate": 3e-06,
+      "loss": -10.387,
+      "step": 290
+    },
+    {
+      "epoch": 0.025866666666666666,
+      "grad_norm": 150.55654907226562,
+      "learning_rate": 3e-06,
+      "loss": -11.9684,
+      "step": 291
+    },
+    {
+      "epoch": 0.025955555555555555,
+      "grad_norm": 86.15855407714844,
+      "learning_rate": 3e-06,
+      "loss": -13.3488,
+      "step": 292
+    },
+    {
+      "epoch": 0.026044444444444443,
+      "grad_norm": 82.68080139160156,
+      "learning_rate": 3e-06,
+      "loss": -9.7978,
+      "step": 293
+    },
+    {
+      "epoch": 0.026133333333333335,
+      "grad_norm": 73.47705841064453,
+      "learning_rate": 3e-06,
+      "loss": -15.2661,
+      "step": 294
+    },
+    {
+      "epoch": 0.026222222222222223,
+      "grad_norm": 88.39766693115234,
+      "learning_rate": 3e-06,
+      "loss": -14.0408,
+      "step": 295
+    },
+    {
+      "epoch": 0.02631111111111111,
+      "grad_norm": 81.03710174560547,
+      "learning_rate": 3e-06,
+      "loss": -11.6435,
+      "step": 296
+    },
+    {
+      "epoch": 0.0264,
+      "grad_norm": 97.82394409179688,
+      "learning_rate": 3e-06,
+      "loss": -12.9819,
+      "step": 297
+    },
+    {
+      "epoch": 0.026488888888888888,
+      "grad_norm": 91.29530334472656,
+      "learning_rate": 3e-06,
+      "loss": -14.6218,
+      "step": 298
+    },
+    {
+      "epoch": 0.026577777777777777,
+      "grad_norm": 76.14654541015625,
+      "learning_rate": 3e-06,
+      "loss": -10.3319,
+      "step": 299
+    },
+    {
+      "epoch": 0.02666666666666667,
+      "grad_norm": 76.4620590209961,
+      "learning_rate": 3e-06,
+      "loss": -16.649,
+      "step": 300
+    },
+    {
+      "completion_length": 245.87500762939453,
+      "epoch": 0.026755555555555557,
+      "grad_norm": 66.47940063476562,
+      "learning_rate": 3e-06,
+      "loss": -2.4273,
+      "reward": 1.2916666865348816,
+      "reward_std": 0.4701542556285858,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.5416666716337204,
+      "step": 301,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.026844444444444445,
+      "grad_norm": 57.04201126098633,
+      "learning_rate": 3e-06,
+      "loss": -2.7348,
+      "step": 302
+    },
+    {
+      "epoch": 0.026933333333333333,
+      "grad_norm": 67.42317962646484,
+      "learning_rate": 3e-06,
+      "loss": -0.5419,
+      "step": 303
+    },
+    {
+      "epoch": 0.027022222222222222,
+      "grad_norm": 68.1643295288086,
+      "learning_rate": 3e-06,
+      "loss": -4.7537,
+      "step": 304
+    },
+    {
+      "epoch": 0.02711111111111111,
+      "grad_norm": 66.15480041503906,
+      "learning_rate": 3e-06,
+      "loss": -3.0219,
+      "step": 305
+    },
+    {
+      "epoch": 0.0272,
+      "grad_norm": 72.49027252197266,
+      "learning_rate": 3e-06,
+      "loss": -0.992,
+      "step": 306
+    },
+    {
+      "epoch": 0.02728888888888889,
+      "grad_norm": 63.84511947631836,
+      "learning_rate": 3e-06,
+      "loss": -2.9532,
+      "step": 307
+    },
+    {
+      "epoch": 0.02737777777777778,
+      "grad_norm": 60.41191864013672,
+      "learning_rate": 3e-06,
+      "loss": -3.5425,
+      "step": 308
+    },
+    {
+      "epoch": 0.027466666666666667,
+      "grad_norm": 75.88224029541016,
+      "learning_rate": 3e-06,
+      "loss": -0.626,
+      "step": 309
+    },
+    {
+      "epoch": 0.027555555555555555,
+      "grad_norm": 60.12965774536133,
+      "learning_rate": 3e-06,
+      "loss": -4.971,
+      "step": 310
+    },
+    {
+      "epoch": 0.027644444444444444,
+      "grad_norm": 67.24330139160156,
+      "learning_rate": 3e-06,
+      "loss": -3.5011,
+      "step": 311
+    },
+    {
+      "epoch": 0.027733333333333332,
+      "grad_norm": 66.98039245605469,
+      "learning_rate": 3e-06,
+      "loss": -1.487,
+      "step": 312
+    },
+    {
+      "completion_length": 255.27083587646484,
+      "epoch": 0.027822222222222224,
+      "grad_norm": 36.04975891113281,
+      "learning_rate": 3e-06,
+      "loss": -0.5188,
+      "reward": 1.5312500596046448,
+      "reward_std": 0.1546149756759405,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.9166666567325592,
+      "step": 313,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.027911111111111112,
+      "grad_norm": 35.71009063720703,
+      "learning_rate": 3e-06,
+      "loss": 2.2571,
+      "step": 314
+    },
+    {
+      "epoch": 0.028,
+      "grad_norm": 43.758975982666016,
+      "learning_rate": 3e-06,
+      "loss": 1.5113,
+      "step": 315
+    },
+    {
+      "epoch": 0.02808888888888889,
+      "grad_norm": 42.099124908447266,
+      "learning_rate": 3e-06,
+      "loss": -1.203,
+      "step": 316
+    },
+    {
+      "epoch": 0.028177777777777777,
+      "grad_norm": 46.69057846069336,
+      "learning_rate": 3e-06,
+      "loss": 2.1907,
+      "step": 317
+    },
+    {
+      "epoch": 0.028266666666666666,
+      "grad_norm": 43.00071334838867,
+      "learning_rate": 3e-06,
+      "loss": -0.0413,
+      "step": 318
+    },
+    {
+      "epoch": 0.028355555555555554,
+      "grad_norm": 36.017799377441406,
+      "learning_rate": 3e-06,
+      "loss": -0.9995,
+      "step": 319
+    },
+    {
+      "epoch": 0.028444444444444446,
+      "grad_norm": 35.86075973510742,
+      "learning_rate": 3e-06,
+      "loss": 1.8159,
+      "step": 320
+    },
+    {
+      "epoch": 0.028533333333333334,
+      "grad_norm": 46.41409683227539,
+      "learning_rate": 3e-06,
+      "loss": 0.8693,
+      "step": 321
+    },
+    {
+      "epoch": 0.028622222222222223,
+      "grad_norm": 42.182472229003906,
+      "learning_rate": 3e-06,
+      "loss": -1.9042,
+      "step": 322
+    },
+    {
+      "epoch": 0.02871111111111111,
+      "grad_norm": 47.805999755859375,
+      "learning_rate": 3e-06,
+      "loss": 1.6417,
+      "step": 323
+    },
+    {
+      "epoch": 0.0288,
+      "grad_norm": 45.03670883178711,
+      "learning_rate": 3e-06,
+      "loss": -1.09,
+      "step": 324
+    },
+    {
+      "completion_length": 252.64583587646484,
+      "epoch": 0.028888888888888888,
+      "grad_norm": 58.358917236328125,
+      "learning_rate": 3e-06,
+      "loss": -12.4368,
+      "reward": 1.1666666865348816,
+      "reward_std": 0.37967559695243835,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6666666567325592,
+      "rewards/correctness_reward_func_math": 0.5,
+      "step": 325,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.02897777777777778,
+      "grad_norm": 70.42740631103516,
+      "learning_rate": 3e-06,
+      "loss": -6.5213,
+      "step": 326
+    },
+    {
+      "epoch": 0.029066666666666668,
+      "grad_norm": 71.7884750366211,
+      "learning_rate": 3e-06,
+      "loss": -14.7372,
+      "step": 327
+    },
+    {
+      "epoch": 0.029155555555555556,
+      "grad_norm": 64.89356231689453,
+      "learning_rate": 3e-06,
+      "loss": -2.411,
+      "step": 328
+    },
+    {
+      "epoch": 0.029244444444444444,
+      "grad_norm": 63.557125091552734,
+      "learning_rate": 3e-06,
+      "loss": -6.2777,
+      "step": 329
+    },
+    {
+      "epoch": 0.029333333333333333,
+      "grad_norm": 55.46377182006836,
+      "learning_rate": 3e-06,
+      "loss": -6.9502,
+      "step": 330
+    },
+    {
+      "epoch": 0.02942222222222222,
+      "grad_norm": 67.63842010498047,
+      "learning_rate": 3e-06,
+      "loss": -13.2148,
+      "step": 331
+    },
+    {
+      "epoch": 0.02951111111111111,
+      "grad_norm": 69.31304931640625,
+      "learning_rate": 3e-06,
+      "loss": -7.451,
+      "step": 332
+    },
+    {
+      "epoch": 0.0296,
+      "grad_norm": 72.68626403808594,
+      "learning_rate": 3e-06,
+      "loss": -15.5911,
+      "step": 333
+    },
+    {
+      "epoch": 0.02968888888888889,
+      "grad_norm": 67.20828247070312,
+      "learning_rate": 3e-06,
+      "loss": -3.4952,
+      "step": 334
+    },
+    {
+      "epoch": 0.029777777777777778,
+      "grad_norm": 71.56851959228516,
+      "learning_rate": 3e-06,
+      "loss": -6.8577,
+      "step": 335
+    },
+    {
+      "epoch": 0.029866666666666666,
+      "grad_norm": 55.80412292480469,
+      "learning_rate": 3e-06,
+      "loss": -7.9959,
+      "step": 336
+    },
+    {
+      "completion_length": 254.33333587646484,
+      "epoch": 0.029955555555555555,
+      "grad_norm": 83.18997955322266,
+      "learning_rate": 3e-06,
+      "loss": -4.4206,
+      "reward": 1.3854166865348816,
+      "reward_std": 0.6009446382522583,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6354166567325592,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 337,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.030044444444444443,
+      "grad_norm": 76.95658111572266,
+      "learning_rate": 3e-06,
+      "loss": 0.0507,
+      "step": 338
+    },
+    {
+      "epoch": 0.030133333333333335,
+      "grad_norm": 99.8234634399414,
+      "learning_rate": 3e-06,
+      "loss": -4.7763,
+      "step": 339
+    },
+    {
+      "epoch": 0.030222222222222223,
+      "grad_norm": 89.73624420166016,
+      "learning_rate": 3e-06,
+      "loss": -12.8145,
+      "step": 340
+    },
+    {
+      "epoch": 0.03031111111111111,
+      "grad_norm": 105.25814819335938,
+      "learning_rate": 3e-06,
+      "loss": 1.3688,
+      "step": 341
+    },
+    {
+      "epoch": 0.0304,
+      "grad_norm": 91.62116241455078,
+      "learning_rate": 3e-06,
+      "loss": -7.2119,
+      "step": 342
+    },
+    {
+      "epoch": 0.03048888888888889,
+      "grad_norm": 89.00618743896484,
+      "learning_rate": 3e-06,
+      "loss": -5.8364,
+      "step": 343
+    },
+    {
+      "epoch": 0.030577777777777777,
+      "grad_norm": 75.79231262207031,
+      "learning_rate": 3e-06,
+      "loss": -1.9053,
+      "step": 344
+    },
+    {
+      "epoch": 0.030666666666666665,
+      "grad_norm": 109.15798950195312,
+      "learning_rate": 3e-06,
+      "loss": -7.4347,
+      "step": 345
+    },
+    {
+      "epoch": 0.030755555555555557,
+      "grad_norm": 91.9997787475586,
+      "learning_rate": 3e-06,
+      "loss": -15.0307,
+      "step": 346
+    },
+    {
+      "epoch": 0.030844444444444445,
+      "grad_norm": 113.604248046875,
+      "learning_rate": 3e-06,
+      "loss": -0.8063,
+      "step": 347
+    },
+    {
+      "epoch": 0.030933333333333334,
+      "grad_norm": 90.35537719726562,
+      "learning_rate": 3e-06,
+      "loss": -9.993,
+      "step": 348
+    },
+    {
+      "completion_length": 241.06250762939453,
+      "epoch": 0.031022222222222222,
+      "grad_norm": 79.29890441894531,
+      "learning_rate": 3e-06,
+      "loss": -11.3458,
+      "reward": 1.6979167461395264,
+      "reward_std": 0.5608386099338531,
+      "rewards/boxed_and_answer_tags_format_reward": 0.65625,
+      "rewards/correctness_reward_func_math": 1.0416666567325592,
+      "step": 349,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.03111111111111111,
+      "grad_norm": 85.71048736572266,
+      "learning_rate": 3e-06,
+      "loss": -0.4666,
+      "step": 350
+    },
+    {
+      "epoch": 0.0312,
+      "grad_norm": 74.05301666259766,
+      "learning_rate": 3e-06,
+      "loss": -17.7307,
+      "step": 351
+    },
+    {
+      "epoch": 0.03128888888888889,
+      "grad_norm": 77.7562026977539,
+      "learning_rate": 3e-06,
+      "loss": -10.9744,
+      "step": 352
+    },
+    {
+      "epoch": 0.031377777777777775,
+      "grad_norm": 81.5293197631836,
+      "learning_rate": 3e-06,
+      "loss": -7.753,
+      "step": 353
+    },
+    {
+      "epoch": 0.031466666666666664,
+      "grad_norm": 74.96295166015625,
+      "learning_rate": 3e-06,
+      "loss": -4.9168,
+      "step": 354
+    },
+    {
+      "epoch": 0.03155555555555556,
+      "grad_norm": 77.19646453857422,
+      "learning_rate": 3e-06,
+      "loss": -12.4852,
+      "step": 355
+    },
+    {
+      "epoch": 0.03164444444444445,
+      "grad_norm": 88.24592590332031,
+      "learning_rate": 3e-06,
+      "loss": -2.2546,
+      "step": 356
+    },
+    {
+      "epoch": 0.031733333333333336,
+      "grad_norm": 69.32099151611328,
+      "learning_rate": 3e-06,
+      "loss": -18.7936,
+      "step": 357
+    },
+    {
+      "epoch": 0.031822222222222224,
+      "grad_norm": 74.59849548339844,
+      "learning_rate": 3e-06,
+      "loss": -12.5008,
+      "step": 358
+    },
+    {
+      "epoch": 0.03191111111111111,
+      "grad_norm": 89.21590423583984,
+      "learning_rate": 3e-06,
+      "loss": -9.1562,
+      "step": 359
+    },
+    {
+      "epoch": 0.032,
+      "grad_norm": 70.9638671875,
+      "learning_rate": 3e-06,
+      "loss": -7.0573,
+      "step": 360
+    },
+    {
+      "completion_length": 250.83334350585938,
+      "epoch": 0.03208888888888889,
+      "grad_norm": 72.94464111328125,
+      "learning_rate": 3e-06,
+      "loss": -14.781,
+      "reward": 0.947916716337204,
+      "reward_std": 0.45044803619384766,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6979166567325592,
+      "rewards/correctness_reward_func_math": 0.2499999962747097,
+      "step": 361,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.03217777777777778,
+      "grad_norm": 91.14832305908203,
+      "learning_rate": 3e-06,
+      "loss": -15.6654,
+      "step": 362
+    },
+    {
+      "epoch": 0.032266666666666666,
+      "grad_norm": 74.7421875,
+      "learning_rate": 3e-06,
+      "loss": -18.7276,
+      "step": 363
+    },
+    {
+      "epoch": 0.032355555555555554,
+      "grad_norm": 76.62783813476562,
+      "learning_rate": 3e-06,
+      "loss": -20.5042,
+      "step": 364
+    },
+    {
+      "epoch": 0.03244444444444444,
+      "grad_norm": 77.55496978759766,
+      "learning_rate": 3e-06,
+      "loss": -23.5561,
+      "step": 365
+    },
+    {
+      "epoch": 0.03253333333333333,
+      "grad_norm": 87.73894500732422,
+      "learning_rate": 3e-06,
+      "loss": -13.7666,
+      "step": 366
+    },
+    {
+      "epoch": 0.03262222222222222,
+      "grad_norm": 85.25169372558594,
+      "learning_rate": 3e-06,
+      "loss": -15.9124,
+      "step": 367
+    },
+    {
+      "epoch": 0.032711111111111114,
+      "grad_norm": 82.08868408203125,
+      "learning_rate": 3e-06,
+      "loss": -17.5144,
+      "step": 368
+    },
+    {
+      "epoch": 0.0328,
+      "grad_norm": 88.5888900756836,
+      "learning_rate": 3e-06,
+      "loss": -20.7047,
+      "step": 369
+    },
+    {
+      "epoch": 0.03288888888888889,
+      "grad_norm": 87.28410339355469,
+      "learning_rate": 3e-06,
+      "loss": -22.9003,
+      "step": 370
+    },
+    {
+      "epoch": 0.03297777777777778,
+      "grad_norm": 75.89826965332031,
+      "learning_rate": 3e-06,
+      "loss": -25.7767,
+      "step": 371
+    },
+    {
+      "epoch": 0.03306666666666667,
+      "grad_norm": 79.7787094116211,
+      "learning_rate": 3e-06,
+      "loss": -16.1167,
+      "step": 372
+    },
+    {
+      "completion_length": 251.9791717529297,
+      "epoch": 0.033155555555555556,
+      "grad_norm": 122.57412719726562,
+      "learning_rate": 3e-06,
+      "loss": -7.3842,
+      "reward": 1.1666666865348816,
+      "reward_std": 0.7582502365112305,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5833333134651184,
+      "rewards/correctness_reward_func_math": 0.5833333283662796,
+      "step": 373,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.033244444444444445,
+      "grad_norm": 103.27584838867188,
+      "learning_rate": 3e-06,
+      "loss": -2.3769,
+      "step": 374
+    },
+    {
+      "epoch": 0.03333333333333333,
+      "grad_norm": 111.76622772216797,
+      "learning_rate": 3e-06,
+      "loss": 5.4646,
+      "step": 375
+    },
+    {
+      "epoch": 0.03342222222222222,
+      "grad_norm": 110.36207580566406,
+      "learning_rate": 3e-06,
+      "loss": -9.271,
+      "step": 376
+    },
+    {
+      "epoch": 0.03351111111111111,
+      "grad_norm": 102.65152740478516,
+      "learning_rate": 3e-06,
+      "loss": -0.8725,
+      "step": 377
+    },
+    {
+      "epoch": 0.0336,
+      "grad_norm": 107.27348327636719,
+      "learning_rate": 3e-06,
+      "loss": -4.3279,
+      "step": 378
+    },
+    {
+      "epoch": 0.033688888888888886,
+      "grad_norm": 118.1567153930664,
+      "learning_rate": 3e-06,
+      "loss": -8.1813,
+      "step": 379
+    },
+    {
+      "epoch": 0.033777777777777775,
+      "grad_norm": 98.9560317993164,
+      "learning_rate": 3e-06,
+      "loss": -3.3073,
+      "step": 380
+    },
+    {
+      "epoch": 0.03386666666666667,
+      "grad_norm": 119.64665222167969,
+      "learning_rate": 3e-06,
+      "loss": 4.1117,
+      "step": 381
+    },
+    {
+      "epoch": 0.03395555555555556,
+      "grad_norm": 118.64970397949219,
+      "learning_rate": 3e-06,
+      "loss": -10.3727,
+      "step": 382
+    },
+    {
+      "epoch": 0.03404444444444445,
+      "grad_norm": 117.53937530517578,
+      "learning_rate": 3e-06,
+      "loss": -3.6092,
+      "step": 383
+    },
+    {
+      "epoch": 0.034133333333333335,
+      "grad_norm": 108.86544799804688,
+      "learning_rate": 3e-06,
+      "loss": -6.4896,
+      "step": 384
+    },
+    {
+      "completion_length": 242.27084350585938,
+      "epoch": 0.03422222222222222,
+      "grad_norm": 37.28844451904297,
+      "learning_rate": 3e-06,
+      "loss": 35.3752,
+      "reward": 1.0208333432674408,
+      "reward_std": 0.12909945845603943,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.3333333358168602,
+      "step": 385,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.03431111111111111,
+      "grad_norm": 34.82659149169922,
+      "learning_rate": 3e-06,
+      "loss": 32.1003,
+      "step": 386
+    },
+    {
+      "epoch": 0.0344,
+      "grad_norm": 34.34743881225586,
+      "learning_rate": 3e-06,
+      "loss": 31.1165,
+      "step": 387
+    },
+    {
+      "epoch": 0.03448888888888889,
+      "grad_norm": 44.72328186035156,
+      "learning_rate": 3e-06,
+      "loss": 32.7756,
+      "step": 388
+    },
+    {
+      "epoch": 0.03457777777777778,
+      "grad_norm": 42.72700119018555,
+      "learning_rate": 3e-06,
+      "loss": 32.2398,
+      "step": 389
+    },
+    {
+      "epoch": 0.034666666666666665,
+      "grad_norm": 47.69383239746094,
+      "learning_rate": 3e-06,
+      "loss": 33.0683,
+      "step": 390
+    },
+    {
+      "epoch": 0.03475555555555555,
+      "grad_norm": 39.66519546508789,
+      "learning_rate": 3e-06,
+      "loss": 34.5053,
+      "step": 391
+    },
+    {
+      "epoch": 0.03484444444444444,
+      "grad_norm": 39.71942138671875,
+      "learning_rate": 3e-06,
+      "loss": 31.0092,
+      "step": 392
+    },
+    {
+      "epoch": 0.03493333333333333,
+      "grad_norm": 36.60993576049805,
+      "learning_rate": 3e-06,
+      "loss": 30.3034,
+      "step": 393
+    },
+    {
+      "epoch": 0.035022222222222225,
+      "grad_norm": 47.912837982177734,
+      "learning_rate": 3e-06,
+      "loss": 31.4023,
+      "step": 394
+    },
+    {
+      "epoch": 0.035111111111111114,
+      "grad_norm": 42.3475341796875,
+      "learning_rate": 3e-06,
+      "loss": 31.2326,
+      "step": 395
+    },
+    {
+      "epoch": 0.0352,
+      "grad_norm": 40.417381286621094,
+      "learning_rate": 3e-06,
+      "loss": 31.5571,
+      "step": 396
+    },
+    {
+      "completion_length": 251.56250762939453,
+      "epoch": 0.03528888888888889,
+      "grad_norm": 41.37530517578125,
+      "learning_rate": 3e-06,
+      "loss": 9.2621,
+      "reward": 1.6458333730697632,
+      "reward_std": 0.1489431317895651,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7291666865348816,
+      "rewards/correctness_reward_func_math": 0.9166666865348816,
+      "step": 397,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.03537777777777778,
+      "grad_norm": 42.853084564208984,
+      "learning_rate": 3e-06,
+      "loss": 8.8572,
+      "step": 398
+    },
+    {
+      "epoch": 0.03546666666666667,
+      "grad_norm": 41.449344635009766,
+      "learning_rate": 3e-06,
+      "loss": 5.6158,
+      "step": 399
+    },
+    {
+      "epoch": 0.035555555555555556,
+      "grad_norm": 58.31279754638672,
+      "learning_rate": 3e-06,
+      "loss": 7.968,
+      "step": 400
+    },
+    {
+      "epoch": 0.035644444444444444,
+      "grad_norm": 48.664459228515625,
+      "learning_rate": 3e-06,
+      "loss": 11.1793,
+      "step": 401
+    },
+    {
+      "epoch": 0.03573333333333333,
+      "grad_norm": 45.02378845214844,
+      "learning_rate": 3e-06,
+      "loss": 7.6242,
+      "step": 402
+    },
+    {
+      "epoch": 0.03582222222222222,
+      "grad_norm": 43.53935241699219,
+      "learning_rate": 3e-06,
+      "loss": 8.0172,
+      "step": 403
+    },
+    {
+      "epoch": 0.03591111111111111,
+      "grad_norm": 42.496604919433594,
+      "learning_rate": 3e-06,
+      "loss": 7.5088,
+      "step": 404
+    },
+    {
+      "epoch": 0.036,
+      "grad_norm": 44.294986724853516,
+      "learning_rate": 3e-06,
+      "loss": 3.9932,
+      "step": 405
+    },
+    {
+      "epoch": 0.036088888888888886,
+      "grad_norm": 73.07268524169922,
+      "learning_rate": 3e-06,
+      "loss": 6.5222,
+      "step": 406
+    },
+    {
+      "epoch": 0.03617777777777778,
+      "grad_norm": 44.31553649902344,
+      "learning_rate": 3e-06,
+      "loss": 9.2936,
+      "step": 407
+    },
+    {
+      "epoch": 0.03626666666666667,
+      "grad_norm": 48.42079162597656,
+      "learning_rate": 3e-06,
+      "loss": 5.8115,
+      "step": 408
+    },
+    {
+      "completion_length": 253.7291717529297,
+      "epoch": 0.03635555555555556,
+      "grad_norm": 55.15653991699219,
+      "learning_rate": 3e-06,
+      "loss": -4.6673,
+      "reward": 0.90625,
+      "reward_std": 0.15461495518684387,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.1666666716337204,
+      "step": 409,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.036444444444444446,
+      "grad_norm": 54.93301773071289,
+      "learning_rate": 3e-06,
+      "loss": 1.7365,
+      "step": 410
+    },
+    {
+      "epoch": 0.036533333333333334,
+      "grad_norm": 50.56829071044922,
+      "learning_rate": 3e-06,
+      "loss": 2.3972,
+      "step": 411
+    },
+    {
+      "epoch": 0.03662222222222222,
+      "grad_norm": 50.894187927246094,
+      "learning_rate": 3e-06,
+      "loss": 0.7298,
+      "step": 412
+    },
+    {
+      "epoch": 0.03671111111111111,
+      "grad_norm": 64.99378204345703,
+      "learning_rate": 3e-06,
+      "loss": 1.6822,
+      "step": 413
+    },
+    {
+      "epoch": 0.0368,
+      "grad_norm": 53.45103454589844,
+      "learning_rate": 3e-06,
+      "loss": 2.1252,
+      "step": 414
+    },
+    {
+      "epoch": 0.03688888888888889,
+      "grad_norm": 58.880393981933594,
+      "learning_rate": 3e-06,
+      "loss": -4.8912,
+      "step": 415
+    },
+    {
+      "epoch": 0.036977777777777776,
+      "grad_norm": 52.00230407714844,
+      "learning_rate": 3e-06,
+      "loss": 0.9087,
+      "step": 416
+    },
+    {
+      "epoch": 0.037066666666666664,
+      "grad_norm": 54.192508697509766,
+      "learning_rate": 3e-06,
+      "loss": 1.5202,
+      "step": 417
+    },
+    {
+      "epoch": 0.03715555555555555,
+      "grad_norm": 52.10379409790039,
+      "learning_rate": 3e-06,
+      "loss": -0.0026,
+      "step": 418
+    },
+    {
+      "epoch": 0.03724444444444444,
+      "grad_norm": 56.19913864135742,
+      "learning_rate": 3e-06,
+      "loss": 0.9843,
+      "step": 419
+    },
+    {
+      "epoch": 0.037333333333333336,
+      "grad_norm": 51.573699951171875,
+      "learning_rate": 3e-06,
+      "loss": 0.9967,
+      "step": 420
+    },
+    {
+      "completion_length": 241.95833587646484,
+      "epoch": 0.037422222222222225,
+      "grad_norm": 65.98304748535156,
+      "learning_rate": 3e-06,
+      "loss": 1.8524,
+      "reward": 1.0937500596046448,
+      "reward_std": 0.3994170129299164,
+      "rewards/boxed_and_answer_tags_format_reward": 0.59375,
+      "rewards/correctness_reward_func_math": 0.5,
+      "step": 421,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.03751111111111111,
+      "grad_norm": 68.6063232421875,
+      "learning_rate": 3e-06,
+      "loss": -2.8346,
+      "step": 422
+    },
+    {
+      "epoch": 0.0376,
+      "grad_norm": 61.650146484375,
+      "learning_rate": 3e-06,
+      "loss": -2.9892,
+      "step": 423
+    },
+    {
+      "epoch": 0.03768888888888889,
+      "grad_norm": 66.45751953125,
+      "learning_rate": 3e-06,
+      "loss": -3.349,
+      "step": 424
+    },
+    {
+      "epoch": 0.03777777777777778,
+      "grad_norm": 66.84425354003906,
+      "learning_rate": 3e-06,
+      "loss": 2.716,
+      "step": 425
+    },
+    {
+      "epoch": 0.037866666666666667,
+      "grad_norm": 61.292354583740234,
+      "learning_rate": 3e-06,
+      "loss": 1.6447,
+      "step": 426
+    },
+    {
+      "epoch": 0.037955555555555555,
+      "grad_norm": 66.02394104003906,
+      "learning_rate": 3e-06,
+      "loss": 1.1078,
+      "step": 427
+    },
+    {
+      "epoch": 0.03804444444444444,
+      "grad_norm": 62.10107421875,
+      "learning_rate": 3e-06,
+      "loss": -3.9971,
+      "step": 428
+    },
+    {
+      "epoch": 0.03813333333333333,
+      "grad_norm": 85.67212677001953,
+      "learning_rate": 3e-06,
+      "loss": -3.8248,
+      "step": 429
+    },
+    {
+      "epoch": 0.03822222222222222,
+      "grad_norm": 60.61140060424805,
+      "learning_rate": 3e-06,
+      "loss": -4.433,
+      "step": 430
+    },
+    {
+      "epoch": 0.03831111111111111,
+      "grad_norm": 62.81836700439453,
+      "learning_rate": 3e-06,
+      "loss": 1.751,
+      "step": 431
+    },
+    {
+      "epoch": 0.0384,
+      "grad_norm": 60.05256652832031,
+      "learning_rate": 3e-06,
+      "loss": 0.8432,
+      "step": 432
+    },
+    {
+      "completion_length": 247.27083587646484,
+      "epoch": 0.03848888888888889,
+      "grad_norm": 83.67351531982422,
+      "learning_rate": 3e-06,
+      "loss": 6.9951,
+      "reward": 1.3229166865348816,
+      "reward_std": 0.3936077058315277,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.5833333544433117,
+      "step": 433,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.03857777777777778,
+      "grad_norm": 71.3005599975586,
+      "learning_rate": 3e-06,
+      "loss": 4.71,
+      "step": 434
+    },
+    {
+      "epoch": 0.03866666666666667,
+      "grad_norm": 81.9188003540039,
+      "learning_rate": 3e-06,
+      "loss": 8.7788,
+      "step": 435
+    },
+    {
+      "epoch": 0.03875555555555556,
+      "grad_norm": 79.42195129394531,
+      "learning_rate": 3e-06,
+      "loss": 11.129,
+      "step": 436
+    },
+    {
+      "epoch": 0.038844444444444445,
+      "grad_norm": 84.69261169433594,
+      "learning_rate": 3e-06,
+      "loss": 10.7206,
+      "step": 437
+    },
+    {
+      "epoch": 0.038933333333333334,
+      "grad_norm": 74.52008819580078,
+      "learning_rate": 3e-06,
+      "loss": 12.7332,
+      "step": 438
+    },
+    {
+      "epoch": 0.03902222222222222,
+      "grad_norm": 93.8567123413086,
+      "learning_rate": 3e-06,
+      "loss": 6.2331,
+      "step": 439
+    },
+    {
+      "epoch": 0.03911111111111111,
+      "grad_norm": 79.21229553222656,
+      "learning_rate": 3e-06,
+      "loss": 3.7102,
+      "step": 440
+    },
+    {
+      "epoch": 0.0392,
+      "grad_norm": 74.53849792480469,
+      "learning_rate": 3e-06,
+      "loss": 8.0091,
+      "step": 441
+    },
+    {
+      "epoch": 0.03928888888888889,
+      "grad_norm": 79.04943084716797,
+      "learning_rate": 3e-06,
+      "loss": 9.4256,
+      "step": 442
+    },
+    {
+      "epoch": 0.039377777777777775,
+      "grad_norm": 81.54142761230469,
+      "learning_rate": 3e-06,
+      "loss": 9.0032,
+      "step": 443
+    },
+    {
+      "epoch": 0.039466666666666664,
+      "grad_norm": 83.44096374511719,
+      "learning_rate": 3e-06,
+      "loss": 11.701,
+      "step": 444
+    },
+    {
+      "completion_length": 249.2291717529297,
+      "epoch": 0.03955555555555555,
+      "grad_norm": 78.02682495117188,
+      "learning_rate": 3e-06,
+      "loss": 1.3882,
+      "reward": 1.1458333730697632,
+      "reward_std": 0.33968228101730347,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6041666865348816,
+      "rewards/correctness_reward_func_math": 0.5416666679084301,
+      "step": 445,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.03964444444444445,
+      "grad_norm": 77.9029312133789,
+      "learning_rate": 3e-06,
+      "loss": -10.5754,
+      "step": 446
+    },
+    {
+      "epoch": 0.039733333333333336,
+      "grad_norm": 78.25541687011719,
+      "learning_rate": 3e-06,
+      "loss": -4.1056,
+      "step": 447
+    },
+    {
+      "epoch": 0.039822222222222224,
+      "grad_norm": 100.47134399414062,
+      "learning_rate": 3e-06,
+      "loss": -10.7897,
+      "step": 448
+    },
+    {
+      "epoch": 0.03991111111111111,
+      "grad_norm": 74.78359985351562,
+      "learning_rate": 3e-06,
+      "loss": -3.3144,
+      "step": 449
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 96.55880737304688,
+      "learning_rate": 3e-06,
+      "loss": 2.8763,
+      "step": 450
+    },
+    {
+      "epoch": 0.04008888888888889,
+      "grad_norm": 87.14449310302734,
+      "learning_rate": 3e-06,
+      "loss": 0.2308,
+      "step": 451
+    },
+    {
+      "epoch": 0.04017777777777778,
+      "grad_norm": 97.77748107910156,
+      "learning_rate": 3e-06,
+      "loss": -11.2666,
+      "step": 452
+    },
+    {
+      "epoch": 0.040266666666666666,
+      "grad_norm": 73.44164276123047,
+      "learning_rate": 3e-06,
+      "loss": -5.678,
+      "step": 453
+    },
+    {
+      "epoch": 0.040355555555555554,
+      "grad_norm": 100.09737396240234,
+      "learning_rate": 3e-06,
+      "loss": -11.4908,
+      "step": 454
+    },
+    {
+      "epoch": 0.04044444444444444,
+      "grad_norm": 74.52122497558594,
+      "learning_rate": 3e-06,
+      "loss": -4.6635,
+      "step": 455
+    },
+    {
+      "epoch": 0.04053333333333333,
+      "grad_norm": 71.74639892578125,
+      "learning_rate": 3e-06,
+      "loss": 1.7931,
+      "step": 456
+    },
+    {
+      "completion_length": 252.52084350585938,
+      "epoch": 0.04062222222222222,
+      "grad_norm": 102.89392852783203,
+      "learning_rate": 3e-06,
+      "loss": 12.935,
+      "reward": 1.0625000596046448,
+      "reward_std": 0.43528568744659424,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.3750000111758709,
+      "step": 457,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.040711111111111115,
+      "grad_norm": 105.67884063720703,
+      "learning_rate": 3e-06,
+      "loss": 2.3682,
+      "step": 458
+    },
+    {
+      "epoch": 0.0408,
+      "grad_norm": 90.33670806884766,
+      "learning_rate": 3e-06,
+      "loss": 14.3722,
+      "step": 459
+    },
+    {
+      "epoch": 0.04088888888888889,
+      "grad_norm": 109.0367431640625,
+      "learning_rate": 3e-06,
+      "loss": 11.6758,
+      "step": 460
+    },
+    {
+      "epoch": 0.04097777777777778,
+      "grad_norm": 108.55673217773438,
+      "learning_rate": 3e-06,
+      "loss": 13.7948,
+      "step": 461
+    },
+    {
+      "epoch": 0.04106666666666667,
+      "grad_norm": 91.73406982421875,
+      "learning_rate": 3e-06,
+      "loss": 7.7516,
+      "step": 462
+    },
+    {
+      "epoch": 0.041155555555555556,
+      "grad_norm": 94.59785461425781,
+      "learning_rate": 3e-06,
+      "loss": 11.5977,
+      "step": 463
+    },
+    {
+      "epoch": 0.041244444444444445,
+      "grad_norm": 78.94522857666016,
+      "learning_rate": 3e-06,
+      "loss": 1.9425,
+      "step": 464
+    },
+    {
+      "epoch": 0.04133333333333333,
+      "grad_norm": 93.91060638427734,
+      "learning_rate": 3e-06,
+      "loss": 12.3454,
+      "step": 465
+    },
+    {
+      "epoch": 0.04142222222222222,
+      "grad_norm": 98.78986358642578,
+      "learning_rate": 3e-06,
+      "loss": 10.3247,
+      "step": 466
+    },
+    {
+      "epoch": 0.04151111111111111,
+      "grad_norm": 111.63731384277344,
+      "learning_rate": 3e-06,
+      "loss": 12.2581,
+      "step": 467
+    },
+    {
+      "epoch": 0.0416,
+      "grad_norm": 88.63348388671875,
+      "learning_rate": 3e-06,
+      "loss": 5.7115,
+      "step": 468
+    },
+    {
+      "completion_length": 235.58333587646484,
+      "epoch": 0.041688888888888886,
+      "grad_norm": 70.78716278076172,
+      "learning_rate": 3e-06,
+      "loss": -40.1334,
+      "reward": 1.4687500596046448,
+      "reward_std": 0.38577648997306824,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.7916666865348816,
+      "step": 469,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.041777777777777775,
+      "grad_norm": 93.38511657714844,
+      "learning_rate": 3e-06,
+      "loss": -36.9438,
+      "step": 470
+    },
+    {
+      "epoch": 0.04186666666666667,
+      "grad_norm": 84.67546081542969,
+      "learning_rate": 3e-06,
+      "loss": -43.221,
+      "step": 471
+    },
+    {
+      "epoch": 0.04195555555555556,
+      "grad_norm": 99.22493743896484,
+      "learning_rate": 3e-06,
+      "loss": -43.1284,
+      "step": 472
+    },
+    {
+      "epoch": 0.04204444444444445,
+      "grad_norm": 77.45098876953125,
+      "learning_rate": 3e-06,
+      "loss": -40.9152,
+      "step": 473
+    },
+    {
+      "epoch": 0.042133333333333335,
+      "grad_norm": 70.95470428466797,
+      "learning_rate": 3e-06,
+      "loss": -32.8293,
+      "step": 474
+    },
+    {
+      "epoch": 0.042222222222222223,
+      "grad_norm": 77.2640609741211,
+      "learning_rate": 3e-06,
+      "loss": -41.1009,
+      "step": 475
+    },
+    {
+      "epoch": 0.04231111111111111,
+      "grad_norm": 84.97554779052734,
+      "learning_rate": 3e-06,
+      "loss": -37.7295,
+      "step": 476
+    },
+    {
+      "epoch": 0.0424,
+      "grad_norm": 67.63621520996094,
+      "learning_rate": 3e-06,
+      "loss": -44.2423,
+      "step": 477
+    },
+    {
+      "epoch": 0.04248888888888889,
+      "grad_norm": 99.13645935058594,
+      "learning_rate": 3e-06,
+      "loss": -45.4078,
+      "step": 478
+    },
+    {
+      "epoch": 0.04257777777777778,
+      "grad_norm": 71.1223373413086,
+      "learning_rate": 3e-06,
+      "loss": -43.0538,
+      "step": 479
+    },
+    {
+      "epoch": 0.042666666666666665,
+      "grad_norm": 81.17517852783203,
+      "learning_rate": 3e-06,
+      "loss": -34.5254,
+      "step": 480
+    },
+    {
+      "completion_length": 250.5625,
+      "epoch": 0.042755555555555554,
+      "grad_norm": 111.48869323730469,
+      "learning_rate": 3e-06,
+      "loss": 20.9161,
+      "reward": 1.2291666865348816,
+      "reward_std": 0.26686520874500275,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6041666865348816,
+      "rewards/correctness_reward_func_math": 0.6250000223517418,
+      "step": 481,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.04284444444444444,
+      "grad_norm": 110.74134063720703,
+      "learning_rate": 3e-06,
+      "loss": 21.5354,
+      "step": 482
+    },
+    {
+      "epoch": 0.04293333333333333,
+      "grad_norm": 121.27849578857422,
+      "learning_rate": 3e-06,
+      "loss": 22.1143,
+      "step": 483
+    },
+    {
+      "epoch": 0.043022222222222226,
+      "grad_norm": 111.5245590209961,
+      "learning_rate": 3e-06,
+      "loss": 27.6408,
+      "step": 484
+    },
+    {
+      "epoch": 0.043111111111111114,
+      "grad_norm": 100.0729751586914,
+      "learning_rate": 3e-06,
+      "loss": 24.0997,
+      "step": 485
+    },
+    {
+      "epoch": 0.0432,
+      "grad_norm": 123.4867935180664,
+      "learning_rate": 3e-06,
+      "loss": 19.3324,
+      "step": 486
+    },
+    {
+      "epoch": 0.04328888888888889,
+      "grad_norm": 107.24234008789062,
+      "learning_rate": 3e-06,
+      "loss": 18.5364,
+      "step": 487
+    },
+    {
+      "epoch": 0.04337777777777778,
+      "grad_norm": 113.5108413696289,
+      "learning_rate": 3e-06,
+      "loss": 18.1962,
+      "step": 488
+    },
+    {
+      "epoch": 0.04346666666666667,
+      "grad_norm": 136.6710968017578,
+      "learning_rate": 3e-06,
+      "loss": 17.9099,
+      "step": 489
+    },
+    {
+      "epoch": 0.043555555555555556,
+      "grad_norm": 110.78118896484375,
+      "learning_rate": 3e-06,
+      "loss": 24.1538,
+      "step": 490
+    },
+    {
+      "epoch": 0.043644444444444444,
+      "grad_norm": 87.0732192993164,
+      "learning_rate": 3e-06,
+      "loss": 20.6286,
+      "step": 491
+    },
+    {
+      "epoch": 0.04373333333333333,
+      "grad_norm": 111.51911163330078,
+      "learning_rate": 3e-06,
+      "loss": 15.8689,
+      "step": 492
+    },
+    {
+      "completion_length": 232.1666717529297,
+      "epoch": 0.04382222222222222,
+      "grad_norm": 58.495811462402344,
+      "learning_rate": 3e-06,
+      "loss": 1.5791,
+      "reward": 1.1145833730697632,
+      "reward_std": 0.20219221711158752,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.5,
+      "step": 493,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.04391111111111111,
+      "grad_norm": 66.24677276611328,
+      "learning_rate": 3e-06,
+      "loss": 8.869,
+      "step": 494
+    },
+    {
+      "epoch": 0.044,
+      "grad_norm": 75.63920593261719,
+      "learning_rate": 3e-06,
+      "loss": 6.6083,
+      "step": 495
+    },
+    {
+      "epoch": 0.044088888888888886,
+      "grad_norm": 41.957889556884766,
+      "learning_rate": 3e-06,
+      "loss": 6.373,
+      "step": 496
+    },
+    {
+      "epoch": 0.04417777777777778,
+      "grad_norm": 56.327693939208984,
+      "learning_rate": 3e-06,
+      "loss": 8.5285,
+      "step": 497
+    },
+    {
+      "epoch": 0.04426666666666667,
+      "grad_norm": 56.58005905151367,
+      "learning_rate": 3e-06,
+      "loss": 8.228,
+      "step": 498
+    },
+    {
+      "epoch": 0.04435555555555556,
+      "grad_norm": 56.65522766113281,
+      "learning_rate": 3e-06,
+      "loss": 1.1127,
+      "step": 499
+    },
+    {
+      "epoch": 0.044444444444444446,
+      "grad_norm": 52.785221099853516,
+      "learning_rate": 3e-06,
+      "loss": 8.324,
+      "step": 500
+    },
+    {
+      "epoch": 0.044533333333333334,
+      "grad_norm": 52.37721252441406,
+      "learning_rate": 3e-06,
+      "loss": 5.6566,
+      "step": 501
+    },
+    {
+      "epoch": 0.04462222222222222,
+      "grad_norm": 50.5732307434082,
+      "learning_rate": 3e-06,
+      "loss": 5.266,
+      "step": 502
+    },
+    {
+      "epoch": 0.04471111111111111,
+      "grad_norm": 59.614261627197266,
+      "learning_rate": 3e-06,
+      "loss": 7.5254,
+      "step": 503
+    },
+    {
+      "epoch": 0.0448,
+      "grad_norm": 57.78561782836914,
+      "learning_rate": 3e-06,
+      "loss": 6.7155,
+      "step": 504
+    },
+    {
+      "completion_length": 249.5416717529297,
+      "epoch": 0.04488888888888889,
+      "grad_norm": 75.29253387451172,
+      "learning_rate": 3e-06,
+      "loss": 1.5945,
+      "reward": 1.1979166865348816,
+      "reward_std": 0.3113893121480942,
+      "rewards/boxed_and_answer_tags_format_reward": 0.65625,
+      "rewards/correctness_reward_func_math": 0.5416666567325592,
+      "step": 505,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.044977777777777776,
+      "grad_norm": 71.35801696777344,
+      "learning_rate": 3e-06,
+      "loss": 6.5836,
+      "step": 506
+    },
+    {
+      "epoch": 0.045066666666666665,
+      "grad_norm": 68.37297058105469,
+      "learning_rate": 3e-06,
+      "loss": 2.1054,
+      "step": 507
+    },
+    {
+      "epoch": 0.04515555555555555,
+      "grad_norm": 67.4723892211914,
+      "learning_rate": 3e-06,
+      "loss": 2.3974,
+      "step": 508
+    },
+    {
+      "epoch": 0.04524444444444444,
+      "grad_norm": 84.1152114868164,
+      "learning_rate": 3e-06,
+      "loss": 0.1293,
+      "step": 509
+    },
+    {
+      "epoch": 0.04533333333333334,
+      "grad_norm": 95.73898315429688,
+      "learning_rate": 3e-06,
+      "loss": -3.5888,
+      "step": 510
+    },
+    {
+      "epoch": 0.045422222222222225,
+      "grad_norm": 73.4489974975586,
+      "learning_rate": 3e-06,
+      "loss": 0.6248,
+      "step": 511
+    },
+    {
+      "epoch": 0.04551111111111111,
+      "grad_norm": 67.3970947265625,
+      "learning_rate": 3e-06,
+      "loss": 5.0044,
+      "step": 512
+    },
+    {
+      "epoch": 0.0456,
+      "grad_norm": 68.55184936523438,
+      "learning_rate": 3e-06,
+      "loss": 1.048,
+      "step": 513
+    },
+    {
+      "epoch": 0.04568888888888889,
+      "grad_norm": 72.8236312866211,
+      "learning_rate": 3e-06,
+      "loss": 0.9001,
+      "step": 514
+    },
+    {
+      "epoch": 0.04577777777777778,
+      "grad_norm": 72.06463623046875,
+      "learning_rate": 3e-06,
+      "loss": -1.2144,
+      "step": 515
+    },
+    {
+      "epoch": 0.04586666666666667,
+      "grad_norm": 87.04244995117188,
+      "learning_rate": 3e-06,
+      "loss": -5.3325,
+      "step": 516
+    },
+    {
+      "completion_length": 235.27083587646484,
+      "epoch": 0.045955555555555555,
+      "grad_norm": 76.81670379638672,
+      "learning_rate": 3e-06,
+      "loss": -18.2768,
+      "reward": 1.2395833730697632,
+      "reward_std": 0.4973409175872803,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.625,
+      "step": 517,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.04604444444444444,
+      "grad_norm": 129.93211364746094,
+      "learning_rate": 3e-06,
+      "loss": -20.3578,
+      "step": 518
+    },
+    {
+      "epoch": 0.04613333333333333,
+      "grad_norm": 87.40878295898438,
+      "learning_rate": 3e-06,
+      "loss": -19.4647,
+      "step": 519
+    },
+    {
+      "epoch": 0.04622222222222222,
+      "grad_norm": 88.27845764160156,
+      "learning_rate": 3e-06,
+      "loss": -23.9652,
+      "step": 520
+    },
+    {
+      "epoch": 0.04631111111111111,
+      "grad_norm": 74.01776123046875,
+      "learning_rate": 3e-06,
+      "loss": -23.0054,
+      "step": 521
+    },
+    {
+      "epoch": 0.0464,
+      "grad_norm": 86.09662628173828,
+      "learning_rate": 3e-06,
+      "loss": -16.2859,
+      "step": 522
+    },
+    {
+      "epoch": 0.04648888888888889,
+      "grad_norm": 86.07221221923828,
+      "learning_rate": 3e-06,
+      "loss": -19.6548,
+      "step": 523
+    },
+    {
+      "epoch": 0.04657777777777778,
+      "grad_norm": 81.51930236816406,
+      "learning_rate": 3e-06,
+      "loss": -22.0114,
+      "step": 524
+    },
+    {
+      "epoch": 0.04666666666666667,
+      "grad_norm": 85.83316802978516,
+      "learning_rate": 3e-06,
+      "loss": -21.3756,
+      "step": 525
+    },
+    {
+      "epoch": 0.04675555555555556,
+      "grad_norm": 80.6872787475586,
+      "learning_rate": 3e-06,
+      "loss": -25.6275,
+      "step": 526
+    },
+    {
+      "epoch": 0.046844444444444445,
+      "grad_norm": 82.79136657714844,
+      "learning_rate": 3e-06,
+      "loss": -24.846,
+      "step": 527
+    },
+    {
+      "epoch": 0.046933333333333334,
+      "grad_norm": 91.39631652832031,
+      "learning_rate": 3e-06,
+      "loss": -17.8527,
+      "step": 528
+    },
+    {
+      "completion_length": 251.77084350585938,
+      "epoch": 0.04702222222222222,
+      "grad_norm": 58.933895111083984,
+      "learning_rate": 3e-06,
+      "loss": 1.5821,
+      "reward": 1.7083333730697632,
+      "reward_std": 0.29204893112182617,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7083333432674408,
+      "rewards/correctness_reward_func_math": 1.0,
+      "step": 529,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.04711111111111111,
+      "grad_norm": 70.82673645019531,
+      "learning_rate": 3e-06,
+      "loss": 4.666,
+      "step": 530
+    },
+    {
+      "epoch": 0.0472,
+      "grad_norm": 62.45388412475586,
+      "learning_rate": 3e-06,
+      "loss": -0.8319,
+      "step": 531
+    },
+    {
+      "epoch": 0.04728888888888889,
+      "grad_norm": 68.45257568359375,
+      "learning_rate": 3e-06,
+      "loss": -4.2921,
+      "step": 532
+    },
+    {
+      "epoch": 0.047377777777777776,
+      "grad_norm": 63.96629333496094,
+      "learning_rate": 3e-06,
+      "loss": 7.1791,
+      "step": 533
+    },
+    {
+      "epoch": 0.047466666666666664,
+      "grad_norm": 71.47022247314453,
+      "learning_rate": 3e-06,
+      "loss": -0.0559,
+      "step": 534
+    },
+    {
+      "epoch": 0.04755555555555555,
+      "grad_norm": 69.21123504638672,
+      "learning_rate": 3e-06,
+      "loss": 1.1193,
+      "step": 535
+    },
+    {
+      "epoch": 0.04764444444444445,
+      "grad_norm": 54.05726623535156,
+      "learning_rate": 3e-06,
+      "loss": 3.7663,
+      "step": 536
+    },
+    {
+      "epoch": 0.047733333333333336,
+      "grad_norm": 57.079166412353516,
+      "learning_rate": 3e-06,
+      "loss": -1.8682,
+      "step": 537
+    },
+    {
+      "epoch": 0.047822222222222224,
+      "grad_norm": 89.69593811035156,
+      "learning_rate": 3e-06,
+      "loss": -5.0232,
+      "step": 538
+    },
+    {
+      "epoch": 0.04791111111111111,
+      "grad_norm": 68.15176391601562,
+      "learning_rate": 3e-06,
+      "loss": 6.4281,
+      "step": 539
+    },
+    {
+      "epoch": 0.048,
+      "grad_norm": 71.53436279296875,
+      "learning_rate": 3e-06,
+      "loss": -0.8678,
+      "step": 540
+    },
+    {
+      "completion_length": 248.4791717529297,
+      "epoch": 0.04808888888888889,
+      "grad_norm": 111.93199157714844,
+      "learning_rate": 3e-06,
+      "loss": -3.1821,
+      "reward": 1.3125,
+      "reward_std": 0.7091469466686249,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.625,
+      "step": 541,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.04817777777777778,
+      "grad_norm": 104.72472381591797,
+      "learning_rate": 3e-06,
+      "loss": -9.114,
+      "step": 542
+    },
+    {
+      "epoch": 0.048266666666666666,
+      "grad_norm": 137.42185974121094,
+      "learning_rate": 3e-06,
+      "loss": -12.9595,
+      "step": 543
+    },
+    {
+      "epoch": 0.048355555555555554,
+      "grad_norm": 115.6964340209961,
+      "learning_rate": 3e-06,
+      "loss": -16.5078,
+      "step": 544
+    },
+    {
+      "epoch": 0.04844444444444444,
+      "grad_norm": 107.43921661376953,
+      "learning_rate": 3e-06,
+      "loss": -12.0856,
+      "step": 545
+    },
+    {
+      "epoch": 0.04853333333333333,
+      "grad_norm": 139.91366577148438,
+      "learning_rate": 3e-06,
+      "loss": -8.5451,
+      "step": 546
+    },
+    {
+      "epoch": 0.04862222222222222,
+      "grad_norm": 107.25052642822266,
+      "learning_rate": 3e-06,
+      "loss": -4.7342,
+      "step": 547
+    },
+    {
+      "epoch": 0.04871111111111111,
+      "grad_norm": 104.6925048828125,
+      "learning_rate": 3e-06,
+      "loss": -10.6582,
+      "step": 548
+    },
+    {
+      "epoch": 0.0488,
+      "grad_norm": 108.13795471191406,
+      "learning_rate": 3e-06,
+      "loss": -14.883,
+      "step": 549
+    },
+    {
+      "epoch": 0.04888888888888889,
+      "grad_norm": 108.62395477294922,
+      "learning_rate": 3e-06,
+      "loss": -19.0199,
+      "step": 550
+    },
+    {
+      "epoch": 0.04897777777777778,
+      "grad_norm": 103.06570434570312,
+      "learning_rate": 3e-06,
+      "loss": -14.306,
+      "step": 551
+    },
+    {
+      "epoch": 0.04906666666666667,
+      "grad_norm": 125.94219970703125,
+      "learning_rate": 3e-06,
+      "loss": -10.2489,
+      "step": 552
+    },
+    {
+      "completion_length": 221.43750762939453,
+      "epoch": 0.049155555555555557,
+      "grad_norm": 90.84999084472656,
+      "learning_rate": 3e-06,
+      "loss": 12.0477,
+      "reward": 2.0104166865348816,
+      "reward_std": 0.406316339969635,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 1.3333333730697632,
+      "step": 553,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.049244444444444445,
+      "grad_norm": 81.99716186523438,
+      "learning_rate": 3e-06,
+      "loss": 11.4046,
+      "step": 554
+    },
+    {
+      "epoch": 0.04933333333333333,
+      "grad_norm": 89.69168090820312,
+      "learning_rate": 3e-06,
+      "loss": 15.5262,
+      "step": 555
+    },
+    {
+      "epoch": 0.04942222222222222,
+      "grad_norm": 93.98981475830078,
+      "learning_rate": 3e-06,
+      "loss": 15.7367,
+      "step": 556
+    },
+    {
+      "epoch": 0.04951111111111111,
+      "grad_norm": 86.68983459472656,
+      "learning_rate": 3e-06,
+      "loss": 5.511,
+      "step": 557
+    },
+    {
+      "epoch": 0.0496,
+      "grad_norm": 78.73108673095703,
+      "learning_rate": 3e-06,
+      "loss": 13.5853,
+      "step": 558
+    },
+    {
+      "epoch": 0.04968888888888889,
+      "grad_norm": 88.60321044921875,
+      "learning_rate": 3e-06,
+      "loss": 10.8131,
+      "step": 559
+    },
+    {
+      "epoch": 0.049777777777777775,
+      "grad_norm": 78.82019805908203,
+      "learning_rate": 3e-06,
+      "loss": 10.9253,
+      "step": 560
+    },
+    {
+      "epoch": 0.04986666666666666,
+      "grad_norm": 91.8647232055664,
+      "learning_rate": 3e-06,
+      "loss": 14.6415,
+      "step": 561
+    },
+    {
+      "epoch": 0.04995555555555556,
+      "grad_norm": 106.29496765136719,
+      "learning_rate": 3e-06,
+      "loss": 14.4603,
+      "step": 562
+    },
+    {
+      "epoch": 0.05004444444444445,
+      "grad_norm": 72.984130859375,
+      "learning_rate": 3e-06,
+      "loss": 4.1028,
+      "step": 563
+    },
+    {
+      "epoch": 0.050133333333333335,
+      "grad_norm": 88.46710205078125,
+      "learning_rate": 3e-06,
+      "loss": 11.4185,
+      "step": 564
+    },
+    {
+      "completion_length": 238.89584350585938,
+      "epoch": 0.050222222222222224,
+      "grad_norm": 91.9970703125,
+      "learning_rate": 3e-06,
+      "loss": -6.0724,
+      "reward": 1.0729166865348816,
+      "reward_std": 0.5935818552970886,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.4583333283662796,
+      "step": 565,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.05031111111111111,
+      "grad_norm": 101.35919952392578,
+      "learning_rate": 3e-06,
+      "loss": -6.1114,
+      "step": 566
+    },
+    {
+      "epoch": 0.0504,
+      "grad_norm": 100.29364776611328,
+      "learning_rate": 3e-06,
+      "loss": 3.7467,
+      "step": 567
+    },
+    {
+      "epoch": 0.05048888888888889,
+      "grad_norm": 93.71849822998047,
+      "learning_rate": 3e-06,
+      "loss": 7.819,
+      "step": 568
+    },
+    {
+      "epoch": 0.05057777777777778,
+      "grad_norm": 84.14008331298828,
+      "learning_rate": 3e-06,
+      "loss": -3.9591,
+      "step": 569
+    },
+    {
+      "epoch": 0.050666666666666665,
+      "grad_norm": 91.14092254638672,
+      "learning_rate": 3e-06,
+      "loss": -7.111,
+      "step": 570
+    },
+    {
+      "epoch": 0.050755555555555554,
+      "grad_norm": 93.15682220458984,
+      "learning_rate": 3e-06,
+      "loss": -7.5642,
+      "step": 571
+    },
+    {
+      "epoch": 0.05084444444444444,
+      "grad_norm": 113.89299011230469,
+      "learning_rate": 3e-06,
+      "loss": -6.5357,
+      "step": 572
+    },
+    {
+      "epoch": 0.05093333333333333,
+      "grad_norm": 91.64227294921875,
+      "learning_rate": 3e-06,
+      "loss": 2.2068,
+      "step": 573
+    },
+    {
+      "epoch": 0.05102222222222222,
+      "grad_norm": 108.78038024902344,
+      "learning_rate": 3e-06,
+      "loss": 5.8992,
+      "step": 574
+    },
+    {
+      "epoch": 0.051111111111111114,
+      "grad_norm": 83.73683166503906,
+      "learning_rate": 3e-06,
+      "loss": -5.9239,
+      "step": 575
+    },
+    {
+      "epoch": 0.0512,
+      "grad_norm": 143.7006378173828,
+      "learning_rate": 3e-06,
+      "loss": -8.4902,
+      "step": 576
+    },
+    {
+      "completion_length": 243.7291717529297,
+      "epoch": 0.05128888888888889,
+      "grad_norm": 82.69302368164062,
+      "learning_rate": 3e-06,
+      "loss": 26.9064,
+      "reward": 1.4583333730697632,
+      "reward_std": 0.4701542556285858,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.708333358168602,
+      "step": 577,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.05137777777777778,
+      "grad_norm": 77.71623992919922,
+      "learning_rate": 3e-06,
+      "loss": 24.0495,
+      "step": 578
+    },
+    {
+      "epoch": 0.05146666666666667,
+      "grad_norm": 80.28130340576172,
+      "learning_rate": 3e-06,
+      "loss": 19.1044,
+      "step": 579
+    },
+    {
+      "epoch": 0.051555555555555556,
+      "grad_norm": 86.22237396240234,
+      "learning_rate": 3e-06,
+      "loss": 22.9667,
+      "step": 580
+    },
+    {
+      "epoch": 0.051644444444444444,
+      "grad_norm": 94.3071060180664,
+      "learning_rate": 3e-06,
+      "loss": 17.6854,
+      "step": 581
+    },
+    {
+      "epoch": 0.05173333333333333,
+      "grad_norm": 86.01050567626953,
+      "learning_rate": 3e-06,
+      "loss": 28.4794,
+      "step": 582
+    },
+    {
+      "epoch": 0.05182222222222222,
+      "grad_norm": 87.72801971435547,
+      "learning_rate": 3e-06,
+      "loss": 25.3194,
+      "step": 583
+    },
+    {
+      "epoch": 0.05191111111111111,
+      "grad_norm": 74.66322326660156,
+      "learning_rate": 3e-06,
+      "loss": 22.9833,
+      "step": 584
+    },
+    {
+      "epoch": 0.052,
+      "grad_norm": 90.72804260253906,
+      "learning_rate": 3e-06,
+      "loss": 17.4641,
+      "step": 585
+    },
+    {
+      "epoch": 0.052088888888888886,
+      "grad_norm": 87.39917755126953,
+      "learning_rate": 3e-06,
+      "loss": 21.6816,
+      "step": 586
+    },
+    {
+      "epoch": 0.052177777777777774,
+      "grad_norm": 85.64997100830078,
+      "learning_rate": 3e-06,
+      "loss": 15.9135,
+      "step": 587
+    },
+    {
+      "epoch": 0.05226666666666667,
+      "grad_norm": 92.35039520263672,
+      "learning_rate": 3e-06,
+      "loss": 26.4856,
+      "step": 588
+    },
+    {
+      "completion_length": 248.3541717529297,
+      "epoch": 0.05235555555555556,
+      "grad_norm": 93.8902359008789,
+      "learning_rate": 3e-06,
+      "loss": 14.9976,
+      "reward": 1.3541666865348816,
+      "reward_std": 0.5839263796806335,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7291666865348816,
+      "rewards/correctness_reward_func_math": 0.6250000149011612,
+      "step": 589,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.052444444444444446,
+      "grad_norm": 98.34622192382812,
+      "learning_rate": 3e-06,
+      "loss": 12.6228,
+      "step": 590
+    },
+    {
+      "epoch": 0.052533333333333335,
+      "grad_norm": 108.93590545654297,
+      "learning_rate": 3e-06,
+      "loss": 9.3556,
+      "step": 591
+    },
+    {
+      "epoch": 0.05262222222222222,
+      "grad_norm": 114.49545288085938,
+      "learning_rate": 3e-06,
+      "loss": 23.8665,
+      "step": 592
+    },
+    {
+      "epoch": 0.05271111111111111,
+      "grad_norm": 102.37223052978516,
+      "learning_rate": 3e-06,
+      "loss": 7.3813,
+      "step": 593
+    },
+    {
+      "epoch": 0.0528,
+      "grad_norm": 98.17306518554688,
+      "learning_rate": 3e-06,
+      "loss": 13.3691,
+      "step": 594
+    },
+    {
+      "epoch": 0.05288888888888889,
+      "grad_norm": 95.84387969970703,
+      "learning_rate": 3e-06,
+      "loss": 13.8857,
+      "step": 595
+    },
+    {
+      "epoch": 0.052977777777777776,
+      "grad_norm": 101.44741821289062,
+      "learning_rate": 3e-06,
+      "loss": 11.0637,
+      "step": 596
+    },
+    {
+      "epoch": 0.053066666666666665,
+      "grad_norm": 96.87228393554688,
+      "learning_rate": 3e-06,
+      "loss": 7.1168,
+      "step": 597
+    },
+    {
+      "epoch": 0.05315555555555555,
+      "grad_norm": 129.4097442626953,
+      "learning_rate": 3e-06,
+      "loss": 22.4722,
+      "step": 598
+    },
+    {
+      "epoch": 0.05324444444444444,
+      "grad_norm": 100.4094467163086,
+      "learning_rate": 3e-06,
+      "loss": 4.9249,
+      "step": 599
+    },
+    {
+      "epoch": 0.05333333333333334,
+      "grad_norm": 97.2879638671875,
+      "learning_rate": 3e-06,
+      "loss": 10.7864,
+      "step": 600
+    },
+    {
+      "completion_length": 249.31250762939453,
+      "epoch": 0.053422222222222225,
+      "grad_norm": 81.29678344726562,
+      "learning_rate": 3e-06,
+      "loss": -1.4361,
+      "reward": 1.6458333730697632,
+      "reward_std": 0.43528565764427185,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.9583333432674408,
+      "step": 601,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.05351111111111111,
+      "grad_norm": 71.18397521972656,
+      "learning_rate": 3e-06,
+      "loss": -9.7138,
+      "step": 602
+    },
+    {
+      "epoch": 0.0536,
+      "grad_norm": 79.09825897216797,
+      "learning_rate": 3e-06,
+      "loss": -7.5626,
+      "step": 603
+    },
+    {
+      "epoch": 0.05368888888888889,
+      "grad_norm": 88.4588394165039,
+      "learning_rate": 3e-06,
+      "loss": -8.3433,
+      "step": 604
+    },
+    {
+      "epoch": 0.05377777777777778,
+      "grad_norm": 75.85662841796875,
+      "learning_rate": 3e-06,
+      "loss": -3.126,
+      "step": 605
+    },
+    {
+      "epoch": 0.05386666666666667,
+      "grad_norm": 76.75032806396484,
+      "learning_rate": 3e-06,
+      "loss": -4.1707,
+      "step": 606
+    },
+    {
+      "epoch": 0.053955555555555555,
+      "grad_norm": 96.06957244873047,
+      "learning_rate": 3e-06,
+      "loss": -2.3824,
+      "step": 607
+    },
+    {
+      "epoch": 0.054044444444444444,
+      "grad_norm": 108.4106674194336,
+      "learning_rate": 3e-06,
+      "loss": -10.8022,
+      "step": 608
+    },
+    {
+      "epoch": 0.05413333333333333,
+      "grad_norm": 82.68360900878906,
+      "learning_rate": 3e-06,
+      "loss": -9.1987,
+      "step": 609
+    },
+    {
+      "epoch": 0.05422222222222222,
+      "grad_norm": 92.35367584228516,
+      "learning_rate": 3e-06,
+      "loss": -9.379,
+      "step": 610
+    },
+    {
+      "epoch": 0.05431111111111111,
+      "grad_norm": 78.61454010009766,
+      "learning_rate": 3e-06,
+      "loss": -4.4421,
+      "step": 611
+    },
+    {
+      "epoch": 0.0544,
+      "grad_norm": 83.68685150146484,
+      "learning_rate": 3e-06,
+      "loss": -5.8651,
+      "step": 612
+    },
+    {
+      "completion_length": 252.8541717529297,
+      "epoch": 0.05448888888888889,
+      "grad_norm": 156.99725341796875,
+      "learning_rate": 3e-06,
+      "loss": -38.1088,
+      "reward": 1.2395833730697632,
+      "reward_std": 0.3936076909303665,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.5,
+      "step": 613,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.05457777777777778,
+      "grad_norm": 82.39892578125,
+      "learning_rate": 3e-06,
+      "loss": -24.9067,
+      "step": 614
+    },
+    {
+      "epoch": 0.05466666666666667,
+      "grad_norm": 76.3405532836914,
+      "learning_rate": 3e-06,
+      "loss": -21.2904,
+      "step": 615
+    },
+    {
+      "epoch": 0.05475555555555556,
+      "grad_norm": 114.32887268066406,
+      "learning_rate": 3e-06,
+      "loss": -41.3273,
+      "step": 616
+    },
+    {
+      "epoch": 0.054844444444444446,
+      "grad_norm": 90.10194396972656,
+      "learning_rate": 3e-06,
+      "loss": -19.8759,
+      "step": 617
+    },
+    {
+      "epoch": 0.054933333333333334,
+      "grad_norm": 101.73013305664062,
+      "learning_rate": 3e-06,
+      "loss": -31.9803,
+      "step": 618
+    },
+    {
+      "epoch": 0.05502222222222222,
+      "grad_norm": 107.18656921386719,
+      "learning_rate": 3e-06,
+      "loss": -38.8443,
+      "step": 619
+    },
+    {
+      "epoch": 0.05511111111111111,
+      "grad_norm": 85.78763580322266,
+      "learning_rate": 3e-06,
+      "loss": -26.4852,
+      "step": 620
+    },
+    {
+      "epoch": 0.0552,
+      "grad_norm": 78.22523498535156,
+      "learning_rate": 3e-06,
+      "loss": -22.9265,
+      "step": 621
+    },
+    {
+      "epoch": 0.05528888888888889,
+      "grad_norm": 120.62594604492188,
+      "learning_rate": 3e-06,
+      "loss": -44.6351,
+      "step": 622
+    },
+    {
+      "epoch": 0.055377777777777776,
+      "grad_norm": 93.32075500488281,
+      "learning_rate": 3e-06,
+      "loss": -21.4564,
+      "step": 623
+    },
+    {
+      "epoch": 0.055466666666666664,
+      "grad_norm": 97.2727279663086,
+      "learning_rate": 3e-06,
+      "loss": -34.0292,
+      "step": 624
+    },
+    {
+      "completion_length": 247.5,
+      "epoch": 0.05555555555555555,
+      "grad_norm": 98.01384735107422,
+      "learning_rate": 3e-06,
+      "loss": -25.8639,
+      "reward": 1.9895833730697632,
+      "reward_std": 0.6326004266738892,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 1.25,
+      "step": 625,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.05564444444444445,
+      "grad_norm": 110.7737045288086,
+      "learning_rate": 3e-06,
+      "loss": -26.4317,
+      "step": 626
+    },
+    {
+      "epoch": 0.055733333333333336,
+      "grad_norm": 100.14824676513672,
+      "learning_rate": 3e-06,
+      "loss": -17.6826,
+      "step": 627
+    },
+    {
+      "epoch": 0.055822222222222224,
+      "grad_norm": 95.32125854492188,
+      "learning_rate": 3e-06,
+      "loss": -13.7046,
+      "step": 628
+    },
+    {
+      "epoch": 0.05591111111111111,
+      "grad_norm": 153.66207885742188,
+      "learning_rate": 3e-06,
+      "loss": -33.0736,
+      "step": 629
+    },
+    {
+      "epoch": 0.056,
+      "grad_norm": 98.00647735595703,
+      "learning_rate": 3e-06,
+      "loss": -23.4488,
+      "step": 630
+    },
+    {
+      "epoch": 0.05608888888888889,
+      "grad_norm": 101.93690490722656,
+      "learning_rate": 3e-06,
+      "loss": -27.788,
+      "step": 631
+    },
+    {
+      "epoch": 0.05617777777777778,
+      "grad_norm": 109.6976089477539,
+      "learning_rate": 3e-06,
+      "loss": -27.8496,
+      "step": 632
+    },
+    {
+      "epoch": 0.056266666666666666,
+      "grad_norm": 94.93986511230469,
+      "learning_rate": 3e-06,
+      "loss": -18.741,
+      "step": 633
+    },
+    {
+      "epoch": 0.056355555555555555,
+      "grad_norm": 99.8827133178711,
+      "learning_rate": 3e-06,
+      "loss": -15.4788,
+      "step": 634
+    },
+    {
+      "epoch": 0.05644444444444444,
+      "grad_norm": 157.86849975585938,
+      "learning_rate": 3e-06,
+      "loss": -35.8463,
+      "step": 635
+    },
+    {
+      "epoch": 0.05653333333333333,
+      "grad_norm": 109.21646118164062,
+      "learning_rate": 3e-06,
+      "loss": -25.4238,
+      "step": 636
+    },
+    {
+      "completion_length": 247.5416717529297,
+      "epoch": 0.05662222222222222,
+      "grad_norm": 107.59577941894531,
+      "learning_rate": 3e-06,
+      "loss": 2.5406,
+      "reward": 0.71875,
+      "reward_std": 0.11004260182380676,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.0416666679084301,
+      "step": 637,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.05671111111111111,
+      "grad_norm": 92.30776977539062,
+      "learning_rate": 3e-06,
+      "loss": 3.1641,
+      "step": 638
+    },
+    {
+      "epoch": 0.0568,
+      "grad_norm": 41.486106872558594,
+      "learning_rate": 3e-06,
+      "loss": 4.5319,
+      "step": 639
+    },
+    {
+      "epoch": 0.05688888888888889,
+      "grad_norm": 50.142147064208984,
+      "learning_rate": 3e-06,
+      "loss": 0.4783,
+      "step": 640
+    },
+    {
+      "epoch": 0.05697777777777778,
+      "grad_norm": 41.67461395263672,
+      "learning_rate": 3e-06,
+      "loss": 0.8438,
+      "step": 641
+    },
+    {
+      "epoch": 0.05706666666666667,
+      "grad_norm": 51.555999755859375,
+      "learning_rate": 3e-06,
+      "loss": 0.3726,
+      "step": 642
+    },
+    {
+      "epoch": 0.05715555555555556,
+      "grad_norm": 56.56801223754883,
+      "learning_rate": 3e-06,
+      "loss": 2.3059,
+      "step": 643
+    },
+    {
+      "epoch": 0.057244444444444445,
+      "grad_norm": 57.52075958251953,
+      "learning_rate": 3e-06,
+      "loss": 2.0942,
+      "step": 644
+    },
+    {
+      "epoch": 0.05733333333333333,
+      "grad_norm": 51.82474136352539,
+      "learning_rate": 3e-06,
+      "loss": 3.4151,
+      "step": 645
+    },
+    {
+      "epoch": 0.05742222222222222,
+      "grad_norm": 45.45164108276367,
+      "learning_rate": 3e-06,
+      "loss": -0.6204,
+      "step": 646
+    },
+    {
+      "epoch": 0.05751111111111111,
+      "grad_norm": 43.34312057495117,
+      "learning_rate": 3e-06,
+      "loss": -0.6824,
+      "step": 647
+    },
+    {
+      "epoch": 0.0576,
+      "grad_norm": 88.9341812133789,
+      "learning_rate": 3e-06,
+      "loss": -0.345,
+      "step": 648
+    },
+    {
+      "completion_length": 246.81250762939453,
+      "epoch": 0.05768888888888889,
+      "grad_norm": 91.70360565185547,
+      "learning_rate": 3e-06,
+      "loss": -21.4258,
+      "reward": 1.8541666865348816,
+      "reward_std": 0.3332235887646675,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 1.1666666567325592,
+      "step": 649,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.057777777777777775,
+      "grad_norm": 117.84858703613281,
+      "learning_rate": 3e-06,
+      "loss": -24.4443,
+      "step": 650
+    },
+    {
+      "epoch": 0.057866666666666663,
+      "grad_norm": 94.53849792480469,
+      "learning_rate": 3e-06,
+      "loss": -21.691,
+      "step": 651
+    },
+    {
+      "epoch": 0.05795555555555556,
+      "grad_norm": 97.39710998535156,
+      "learning_rate": 3e-06,
+      "loss": -24.1024,
+      "step": 652
+    },
+    {
+      "epoch": 0.05804444444444445,
+      "grad_norm": 90.82528686523438,
+      "learning_rate": 3e-06,
+      "loss": -23.0164,
+      "step": 653
+    },
+    {
+      "epoch": 0.058133333333333335,
+      "grad_norm": 77.49068450927734,
+      "learning_rate": 3e-06,
+      "loss": -22.6643,
+      "step": 654
+    },
+    {
+      "epoch": 0.058222222222222224,
+      "grad_norm": 87.06861114501953,
+      "learning_rate": 3e-06,
+      "loss": -22.1806,
+      "step": 655
+    },
+    {
+      "epoch": 0.05831111111111111,
+      "grad_norm": 116.24286651611328,
+      "learning_rate": 3e-06,
+      "loss": -25.0616,
+      "step": 656
+    },
+    {
+      "epoch": 0.0584,
+      "grad_norm": 80.90653228759766,
+      "learning_rate": 3e-06,
+      "loss": -23.1949,
+      "step": 657
+    },
+    {
+      "epoch": 0.05848888888888889,
+      "grad_norm": 122.82756042480469,
+      "learning_rate": 3e-06,
+      "loss": -25.4261,
+      "step": 658
+    },
+    {
+      "epoch": 0.05857777777777778,
+      "grad_norm": 87.51302337646484,
+      "learning_rate": 3e-06,
+      "loss": -24.0548,
+      "step": 659
+    },
+    {
+      "epoch": 0.058666666666666666,
+      "grad_norm": 85.40345764160156,
+      "learning_rate": 3e-06,
+      "loss": -23.4326,
+      "step": 660
+    },
+    {
+      "completion_length": 250.9375,
+      "epoch": 0.058755555555555554,
+      "grad_norm": 40.29450607299805,
+      "learning_rate": 3e-06,
+      "loss": 9.1824,
+      "reward": 0.8541666865348816,
+      "reward_std": 0.10206206887960434,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5625,
+      "rewards/correctness_reward_func_math": 0.2916666567325592,
+      "step": 661,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.05884444444444444,
+      "grad_norm": 53.10469436645508,
+      "learning_rate": 3e-06,
+      "loss": 5.2732,
+      "step": 662
+    },
+    {
+      "epoch": 0.05893333333333333,
+      "grad_norm": 53.52404022216797,
+      "learning_rate": 3e-06,
+      "loss": 9.7305,
+      "step": 663
+    },
+    {
+      "epoch": 0.05902222222222222,
+      "grad_norm": 43.00156021118164,
+      "learning_rate": 3e-06,
+      "loss": 10.5135,
+      "step": 664
+    },
+    {
+      "epoch": 0.059111111111111114,
+      "grad_norm": 51.94622802734375,
+      "learning_rate": 3e-06,
+      "loss": 8.7434,
+      "step": 665
+    },
+    {
+      "epoch": 0.0592,
+      "grad_norm": 45.08756637573242,
+      "learning_rate": 3e-06,
+      "loss": 10.9107,
+      "step": 666
+    },
+    {
+      "epoch": 0.05928888888888889,
+      "grad_norm": 38.80097961425781,
+      "learning_rate": 3e-06,
+      "loss": 9.0445,
+      "step": 667
+    },
+    {
+      "epoch": 0.05937777777777778,
+      "grad_norm": 50.45252227783203,
+      "learning_rate": 3e-06,
+      "loss": 4.2955,
+      "step": 668
+    },
+    {
+      "epoch": 0.05946666666666667,
+      "grad_norm": 42.282501220703125,
+      "learning_rate": 3e-06,
+      "loss": 8.7618,
+      "step": 669
+    },
+    {
+      "epoch": 0.059555555555555556,
+      "grad_norm": 40.22513961791992,
+      "learning_rate": 3e-06,
+      "loss": 9.5891,
+      "step": 670
+    },
+    {
+      "epoch": 0.059644444444444444,
+      "grad_norm": 55.79698181152344,
+      "learning_rate": 3e-06,
+      "loss": 7.7516,
+      "step": 671
+    },
+    {
+      "epoch": 0.05973333333333333,
+      "grad_norm": 41.462181091308594,
+      "learning_rate": 3e-06,
+      "loss": 10.17,
+      "step": 672
+    },
+    {
+      "completion_length": 238.4375,
+      "epoch": 0.05982222222222222,
+      "grad_norm": 95.73463439941406,
+      "learning_rate": 3e-06,
+      "loss": 14.012,
+      "reward": 1.4583333730697632,
+      "reward_std": 0.3410547822713852,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.7083333134651184,
+      "step": 673,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.05991111111111111,
+      "grad_norm": 92.55806732177734,
+      "learning_rate": 3e-06,
+      "loss": 7.0593,
+      "step": 674
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 90.97936248779297,
+      "learning_rate": 3e-06,
+      "loss": 6.1863,
+      "step": 675
+    },
+    {
+      "epoch": 0.060088888888888886,
+      "grad_norm": 82.84268951416016,
+      "learning_rate": 3e-06,
+      "loss": -3.7791,
+      "step": 676
+    },
+    {
+      "epoch": 0.060177777777777774,
+      "grad_norm": 93.55880737304688,
+      "learning_rate": 3e-06,
+      "loss": 1.1706,
+      "step": 677
+    },
+    {
+      "epoch": 0.06026666666666667,
+      "grad_norm": 143.54983520507812,
+      "learning_rate": 3e-06,
+      "loss": 5.2435,
+      "step": 678
+    },
+    {
+      "epoch": 0.06035555555555556,
+      "grad_norm": 103.67829132080078,
+      "learning_rate": 3e-06,
+      "loss": 12.2804,
+      "step": 679
+    },
+    {
+      "epoch": 0.060444444444444446,
+      "grad_norm": 94.47793579101562,
+      "learning_rate": 3e-06,
+      "loss": 5.8252,
+      "step": 680
+    },
+    {
+      "epoch": 0.060533333333333335,
+      "grad_norm": 87.38697814941406,
+      "learning_rate": 3e-06,
+      "loss": 4.5993,
+      "step": 681
+    },
+    {
+      "epoch": 0.06062222222222222,
+      "grad_norm": 79.12389373779297,
+      "learning_rate": 3e-06,
+      "loss": -4.8813,
+      "step": 682
+    },
+    {
+      "epoch": 0.06071111111111111,
+      "grad_norm": 97.0263900756836,
+      "learning_rate": 3e-06,
+      "loss": -0.474,
+      "step": 683
+    },
+    {
+      "epoch": 0.0608,
+      "grad_norm": 183.30641174316406,
+      "learning_rate": 3e-06,
+      "loss": 2.9729,
+      "step": 684
+    },
+    {
+      "completion_length": 255.4166717529297,
+      "epoch": 0.06088888888888889,
+      "grad_norm": 119.54466247558594,
+      "learning_rate": 3e-06,
+      "loss": -13.0228,
+      "reward": 1.4583333730697632,
+      "reward_std": 0.505022794008255,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.708333358168602,
+      "step": 685,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.06097777777777778,
+      "grad_norm": 121.260986328125,
+      "learning_rate": 3e-06,
+      "loss": -7.3608,
+      "step": 686
+    },
+    {
+      "epoch": 0.061066666666666665,
+      "grad_norm": 107.9063720703125,
+      "learning_rate": 3e-06,
+      "loss": -2.885,
+      "step": 687
+    },
+    {
+      "epoch": 0.06115555555555555,
+      "grad_norm": 122.71879577636719,
+      "learning_rate": 3e-06,
+      "loss": -9.4339,
+      "step": 688
+    },
+    {
+      "epoch": 0.06124444444444444,
+      "grad_norm": 98.11092376708984,
+      "learning_rate": 3e-06,
+      "loss": -7.9372,
+      "step": 689
+    },
+    {
+      "epoch": 0.06133333333333333,
+      "grad_norm": 106.66928100585938,
+      "learning_rate": 3e-06,
+      "loss": 2.2675,
+      "step": 690
+    },
+    {
+      "epoch": 0.061422222222222225,
+      "grad_norm": 117.10845947265625,
+      "learning_rate": 3e-06,
+      "loss": -13.9389,
+      "step": 691
+    },
+    {
+      "epoch": 0.061511111111111114,
+      "grad_norm": 123.24708557128906,
+      "learning_rate": 3e-06,
+      "loss": -9.064,
+      "step": 692
+    },
+    {
+      "epoch": 0.0616,
+      "grad_norm": 108.9686050415039,
+      "learning_rate": 3e-06,
+      "loss": -4.3708,
+      "step": 693
+    },
+    {
+      "epoch": 0.06168888888888889,
+      "grad_norm": 120.98512268066406,
+      "learning_rate": 3e-06,
+      "loss": -10.9314,
+      "step": 694
+    },
+    {
+      "epoch": 0.06177777777777778,
+      "grad_norm": 96.25732421875,
+      "learning_rate": 3e-06,
+      "loss": -9.4401,
+      "step": 695
+    },
+    {
+      "epoch": 0.06186666666666667,
+      "grad_norm": 107.97279357910156,
+      "learning_rate": 3e-06,
+      "loss": 0.1328,
+      "step": 696
+    },
+    {
+      "completion_length": 242.4166717529297,
+      "epoch": 0.061955555555555555,
+      "grad_norm": 77.24815368652344,
+      "learning_rate": 3e-06,
+      "loss": 21.7633,
+      "reward": 1.6875,
+      "reward_std": 0.3680921494960785,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 1.0,
+      "step": 697,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.062044444444444444,
+      "grad_norm": 90.61959075927734,
+      "learning_rate": 3e-06,
+      "loss": 21.3004,
+      "step": 698
+    },
+    {
+      "epoch": 0.06213333333333333,
+      "grad_norm": 79.43978881835938,
+      "learning_rate": 3e-06,
+      "loss": 22.885,
+      "step": 699
+    },
+    {
+      "epoch": 0.06222222222222222,
+      "grad_norm": 88.06346130371094,
+      "learning_rate": 3e-06,
+      "loss": 16.6794,
+      "step": 700
+    },
+    {
+      "epoch": 0.06231111111111111,
+      "grad_norm": 215.39535522460938,
+      "learning_rate": 3e-06,
+      "loss": 18.8777,
+      "step": 701
+    },
+    {
+      "epoch": 0.0624,
+      "grad_norm": 92.26004791259766,
+      "learning_rate": 3e-06,
+      "loss": 11.0055,
+      "step": 702
+    },
+    {
+      "epoch": 0.062488888888888885,
+      "grad_norm": 77.40907287597656,
+      "learning_rate": 3e-06,
+      "loss": 20.7747,
+      "step": 703
+    },
+    {
+      "epoch": 0.06257777777777777,
+      "grad_norm": 100.23749542236328,
+      "learning_rate": 3e-06,
+      "loss": 20.3858,
+      "step": 704
+    },
+    {
+      "epoch": 0.06266666666666666,
+      "grad_norm": 75.75386047363281,
+      "learning_rate": 3e-06,
+      "loss": 22.1331,
+      "step": 705
+    },
+    {
+      "epoch": 0.06275555555555555,
+      "grad_norm": 88.5040054321289,
+      "learning_rate": 3e-06,
+      "loss": 15.6211,
+      "step": 706
+    },
+    {
+      "epoch": 0.06284444444444444,
+      "grad_norm": 83.19754028320312,
+      "learning_rate": 3e-06,
+      "loss": 17.9444,
+      "step": 707
+    },
+    {
+      "epoch": 0.06293333333333333,
+      "grad_norm": 88.55027770996094,
+      "learning_rate": 3e-06,
+      "loss": 9.7909,
+      "step": 708
+    },
+    {
+      "completion_length": 253.8541717529297,
+      "epoch": 0.06302222222222222,
+      "grad_norm": 106.40866088867188,
+      "learning_rate": 3e-06,
+      "loss": -1.6061,
+      "reward": 1.2708333730697632,
+      "reward_std": 0.38547582924366,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.5833333432674408,
+      "step": 709,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.06311111111111112,
+      "grad_norm": 123.38434600830078,
+      "learning_rate": 3e-06,
+      "loss": -9.3301,
+      "step": 710
+    },
+    {
+      "epoch": 0.0632,
+      "grad_norm": 101.67454528808594,
+      "learning_rate": 3e-06,
+      "loss": -7.1273,
+      "step": 711
+    },
+    {
+      "epoch": 0.0632888888888889,
+      "grad_norm": 112.82793426513672,
+      "learning_rate": 3e-06,
+      "loss": -9.8195,
+      "step": 712
+    },
+    {
+      "epoch": 0.06337777777777778,
+      "grad_norm": 108.99236297607422,
+      "learning_rate": 3e-06,
+      "loss": -2.3741,
+      "step": 713
+    },
+    {
+      "epoch": 0.06346666666666667,
+      "grad_norm": 106.9615478515625,
+      "learning_rate": 3e-06,
+      "loss": -7.792,
+      "step": 714
+    },
+    {
+      "epoch": 0.06355555555555556,
+      "grad_norm": 94.91867065429688,
+      "learning_rate": 3e-06,
+      "loss": -3.0819,
+      "step": 715
+    },
+    {
+      "epoch": 0.06364444444444445,
+      "grad_norm": 114.82243347167969,
+      "learning_rate": 3e-06,
+      "loss": -11.4017,
+      "step": 716
+    },
+    {
+      "epoch": 0.06373333333333334,
+      "grad_norm": 104.33937072753906,
+      "learning_rate": 3e-06,
+      "loss": -9.7339,
+      "step": 717
+    },
+    {
+      "epoch": 0.06382222222222222,
+      "grad_norm": 136.31576538085938,
+      "learning_rate": 3e-06,
+      "loss": -12.1719,
+      "step": 718
+    },
+    {
+      "epoch": 0.06391111111111111,
+      "grad_norm": 99.78816223144531,
+      "learning_rate": 3e-06,
+      "loss": -4.7219,
+      "step": 719
+    },
+    {
+      "epoch": 0.064,
+      "grad_norm": 120.37998962402344,
+      "learning_rate": 3e-06,
+      "loss": -10.5969,
+      "step": 720
+    },
+    {
+      "completion_length": 244.0416717529297,
+      "epoch": 0.06408888888888889,
+      "grad_norm": 101.32681274414062,
+      "learning_rate": 3e-06,
+      "loss": 2.1527,
+      "reward": 1.0,
+      "reward_std": 0.4417443424463272,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6666666865348816,
+      "rewards/correctness_reward_func_math": 0.3333333246409893,
+      "step": 721,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.06417777777777778,
+      "grad_norm": 113.38105773925781,
+      "learning_rate": 3e-06,
+      "loss": 11.181,
+      "step": 722
+    },
+    {
+      "epoch": 0.06426666666666667,
+      "grad_norm": 114.85003662109375,
+      "learning_rate": 3e-06,
+      "loss": -2.1413,
+      "step": 723
+    },
+    {
+      "epoch": 0.06435555555555555,
+      "grad_norm": 133.62515258789062,
+      "learning_rate": 3e-06,
+      "loss": 9.3709,
+      "step": 724
+    },
+    {
+      "epoch": 0.06444444444444444,
+      "grad_norm": 113.68856048583984,
+      "learning_rate": 3e-06,
+      "loss": -3.0918,
+      "step": 725
+    },
+    {
+      "epoch": 0.06453333333333333,
+      "grad_norm": 161.93836975097656,
+      "learning_rate": 3e-06,
+      "loss": -1.9418,
+      "step": 726
+    },
+    {
+      "epoch": 0.06462222222222222,
+      "grad_norm": 100.46546173095703,
+      "learning_rate": 3e-06,
+      "loss": 0.5341,
+      "step": 727
+    },
+    {
+      "epoch": 0.06471111111111111,
+      "grad_norm": 141.3654022216797,
+      "learning_rate": 3e-06,
+      "loss": 9.9107,
+      "step": 728
+    },
+    {
+      "epoch": 0.0648,
+      "grad_norm": 131.8665771484375,
+      "learning_rate": 3e-06,
+      "loss": -3.674,
+      "step": 729
+    },
+    {
+      "epoch": 0.06488888888888888,
+      "grad_norm": 128.03195190429688,
+      "learning_rate": 3e-06,
+      "loss": 6.8735,
+      "step": 730
+    },
+    {
+      "epoch": 0.06497777777777777,
+      "grad_norm": 117.97486114501953,
+      "learning_rate": 3e-06,
+      "loss": -5.1859,
+      "step": 731
+    },
+    {
+      "epoch": 0.06506666666666666,
+      "grad_norm": 158.5392303466797,
+      "learning_rate": 3e-06,
+      "loss": -4.222,
+      "step": 732
+    },
+    {
+      "completion_length": 249.08334350585938,
+      "epoch": 0.06515555555555555,
+      "grad_norm": 122.15642547607422,
+      "learning_rate": 3e-06,
+      "loss": 5.173,
+      "reward": 1.1770833730697632,
+      "reward_std": 0.2915456295013428,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6354166567325592,
+      "rewards/correctness_reward_func_math": 0.5416666567325592,
+      "step": 733,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.06524444444444444,
+      "grad_norm": 122.5665054321289,
+      "learning_rate": 3e-06,
+      "loss": -10.2692,
+      "step": 734
+    },
+    {
+      "epoch": 0.06533333333333333,
+      "grad_norm": 95.55500030517578,
+      "learning_rate": 3e-06,
+      "loss": -4.3816,
+      "step": 735
+    },
+    {
+      "epoch": 0.06542222222222223,
+      "grad_norm": 111.13971710205078,
+      "learning_rate": 3e-06,
+      "loss": -4.5502,
+      "step": 736
+    },
+    {
+      "epoch": 0.06551111111111112,
+      "grad_norm": 98.50959014892578,
+      "learning_rate": 3e-06,
+      "loss": -5.0919,
+      "step": 737
+    },
+    {
+      "epoch": 0.0656,
+      "grad_norm": 82.98762512207031,
+      "learning_rate": 3e-06,
+      "loss": -7.9156,
+      "step": 738
+    },
+    {
+      "epoch": 0.0656888888888889,
+      "grad_norm": 114.31904602050781,
+      "learning_rate": 3e-06,
+      "loss": 4.2146,
+      "step": 739
+    },
+    {
+      "epoch": 0.06577777777777778,
+      "grad_norm": 109.786376953125,
+      "learning_rate": 3e-06,
+      "loss": -11.7776,
+      "step": 740
+    },
+    {
+      "epoch": 0.06586666666666667,
+      "grad_norm": 95.00721740722656,
+      "learning_rate": 3e-06,
+      "loss": -5.7331,
+      "step": 741
+    },
+    {
+      "epoch": 0.06595555555555556,
+      "grad_norm": 87.01516723632812,
+      "learning_rate": 3e-06,
+      "loss": -5.0034,
+      "step": 742
+    },
+    {
+      "epoch": 0.06604444444444445,
+      "grad_norm": 101.2181167602539,
+      "learning_rate": 3e-06,
+      "loss": -6.7069,
+      "step": 743
+    },
+    {
+      "epoch": 0.06613333333333334,
+      "grad_norm": 79.3046875,
+      "learning_rate": 3e-06,
+      "loss": -8.6759,
+      "step": 744
+    },
+    {
+      "completion_length": 252.93750762939453,
+      "epoch": 0.06622222222222222,
+      "grad_norm": 101.21440887451172,
+      "learning_rate": 3e-06,
+      "loss": 2.4358,
+      "reward": 1.354166716337204,
+      "reward_std": 0.4736091196537018,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6458333432674408,
+      "rewards/correctness_reward_func_math": 0.7083333358168602,
+      "step": 745,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.06631111111111111,
+      "grad_norm": 133.58985900878906,
+      "learning_rate": 3e-06,
+      "loss": 0.2243,
+      "step": 746
+    },
+    {
+      "epoch": 0.0664,
+      "grad_norm": 137.97776794433594,
+      "learning_rate": 3e-06,
+      "loss": 3.7561,
+      "step": 747
+    },
+    {
+      "epoch": 0.06648888888888889,
+      "grad_norm": 97.12255859375,
+      "learning_rate": 3e-06,
+      "loss": 2.8029,
+      "step": 748
+    },
+    {
+      "epoch": 0.06657777777777778,
+      "grad_norm": 133.5359344482422,
+      "learning_rate": 3e-06,
+      "loss": 5.9185,
+      "step": 749
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 165.00294494628906,
+      "learning_rate": 3e-06,
+      "loss": 5.6118,
+      "step": 750
+    },
+    {
+      "epoch": 0.06675555555555555,
+      "grad_norm": 147.9979705810547,
+      "learning_rate": 3e-06,
+      "loss": 1.9697,
+      "step": 751
+    },
+    {
+      "epoch": 0.06684444444444444,
+      "grad_norm": 119.22462463378906,
+      "learning_rate": 3e-06,
+      "loss": -0.882,
+      "step": 752
+    },
+    {
+      "epoch": 0.06693333333333333,
+      "grad_norm": 144.49305725097656,
+      "learning_rate": 3e-06,
+      "loss": 2.487,
+      "step": 753
+    },
+    {
+      "epoch": 0.06702222222222222,
+      "grad_norm": 96.12986755371094,
+      "learning_rate": 3e-06,
+      "loss": 1.0482,
+      "step": 754
+    },
+    {
+      "epoch": 0.06711111111111111,
+      "grad_norm": 117.53173065185547,
+      "learning_rate": 3e-06,
+      "loss": 4.1922,
+      "step": 755
+    },
+    {
+      "epoch": 0.0672,
+      "grad_norm": 99.70003509521484,
+      "learning_rate": 3e-06,
+      "loss": 4.014,
+      "step": 756
+    },
+    {
+      "completion_length": 252.9166717529297,
+      "epoch": 0.06728888888888888,
+      "grad_norm": 107.6641616821289,
+      "learning_rate": 3e-06,
+      "loss": -6.2275,
+      "reward": 1.7812500596046448,
+      "reward_std": 0.38577648997306824,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 1.0416666567325592,
+      "step": 757,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.06737777777777777,
+      "grad_norm": 87.34847259521484,
+      "learning_rate": 3e-06,
+      "loss": -3.2976,
+      "step": 758
+    },
+    {
+      "epoch": 0.06746666666666666,
+      "grad_norm": 114.33875274658203,
+      "learning_rate": 3e-06,
+      "loss": -4.307,
+      "step": 759
+    },
+    {
+      "epoch": 0.06755555555555555,
+      "grad_norm": 106.53048706054688,
+      "learning_rate": 3e-06,
+      "loss": -1.13,
+      "step": 760
+    },
+    {
+      "epoch": 0.06764444444444444,
+      "grad_norm": 120.27633666992188,
+      "learning_rate": 3e-06,
+      "loss": -3.2187,
+      "step": 761
+    },
+    {
+      "epoch": 0.06773333333333334,
+      "grad_norm": 96.025390625,
+      "learning_rate": 3e-06,
+      "loss": 0.3329,
+      "step": 762
+    },
+    {
+      "epoch": 0.06782222222222223,
+      "grad_norm": 115.00994110107422,
+      "learning_rate": 3e-06,
+      "loss": -7.6432,
+      "step": 763
+    },
+    {
+      "epoch": 0.06791111111111112,
+      "grad_norm": 87.76792907714844,
+      "learning_rate": 3e-06,
+      "loss": -4.6143,
+      "step": 764
+    },
+    {
+      "epoch": 0.068,
+      "grad_norm": 113.94709777832031,
+      "learning_rate": 3e-06,
+      "loss": -5.7334,
+      "step": 765
+    },
+    {
+      "epoch": 0.0680888888888889,
+      "grad_norm": 111.59996032714844,
+      "learning_rate": 3e-06,
+      "loss": -2.659,
+      "step": 766
+    },
+    {
+      "epoch": 0.06817777777777778,
+      "grad_norm": 129.63861083984375,
+      "learning_rate": 3e-06,
+      "loss": -5.0582,
+      "step": 767
+    },
+    {
+      "epoch": 0.06826666666666667,
+      "grad_norm": 93.2347412109375,
+      "learning_rate": 3e-06,
+      "loss": -1.0268,
+      "step": 768
+    },
+    {
+      "completion_length": 251.52083587646484,
+      "epoch": 0.06835555555555556,
+      "grad_norm": 184.61167907714844,
+      "learning_rate": 3e-06,
+      "loss": 5.0221,
+      "reward": 1.4166666865348816,
+      "reward_std": 0.6262910515069962,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.6666666567325592,
+      "step": 769,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.06844444444444445,
+      "grad_norm": 137.45404052734375,
+      "learning_rate": 3e-06,
+      "loss": 7.8168,
+      "step": 770
+    },
+    {
+      "epoch": 0.06853333333333333,
+      "grad_norm": 202.94410705566406,
+      "learning_rate": 3e-06,
+      "loss": 9.1514,
+      "step": 771
+    },
+    {
+      "epoch": 0.06862222222222222,
+      "grad_norm": 161.6815948486328,
+      "learning_rate": 3e-06,
+      "loss": 8.282,
+      "step": 772
+    },
+    {
+      "epoch": 0.06871111111111111,
+      "grad_norm": 123.37694549560547,
+      "learning_rate": 3e-06,
+      "loss": 3.5652,
+      "step": 773
+    },
+    {
+      "epoch": 0.0688,
+      "grad_norm": 146.2295379638672,
+      "learning_rate": 3e-06,
+      "loss": 9.5204,
+      "step": 774
+    },
+    {
+      "epoch": 0.06888888888888889,
+      "grad_norm": 145.88613891601562,
+      "learning_rate": 3e-06,
+      "loss": 4.6675,
+      "step": 775
+    },
+    {
+      "epoch": 0.06897777777777778,
+      "grad_norm": 135.99313354492188,
+      "learning_rate": 3e-06,
+      "loss": 6.1798,
+      "step": 776
+    },
+    {
+      "epoch": 0.06906666666666667,
+      "grad_norm": 134.66729736328125,
+      "learning_rate": 3e-06,
+      "loss": 6.9994,
+      "step": 777
+    },
+    {
+      "epoch": 0.06915555555555555,
+      "grad_norm": 173.39735412597656,
+      "learning_rate": 3e-06,
+      "loss": 7.3314,
+      "step": 778
+    },
+    {
+      "epoch": 0.06924444444444444,
+      "grad_norm": 136.74331665039062,
+      "learning_rate": 3e-06,
+      "loss": 1.5412,
+      "step": 779
+    },
+    {
+      "epoch": 0.06933333333333333,
+      "grad_norm": 142.06529235839844,
+      "learning_rate": 3e-06,
+      "loss": 7.5,
+      "step": 780
+    },
+    {
+      "completion_length": 236.14583587646484,
+      "epoch": 0.06942222222222222,
+      "grad_norm": 92.94062805175781,
+      "learning_rate": 3e-06,
+      "loss": 9.4998,
+      "reward": 0.8854166865348816,
+      "reward_std": 0.3922351598739624,
+      "rewards/boxed_and_answer_tags_format_reward": 0.71875,
+      "rewards/correctness_reward_func_math": 0.1666666679084301,
+      "step": 781,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.0695111111111111,
+      "grad_norm": 84.2685546875,
+      "learning_rate": 3e-06,
+      "loss": 20.3539,
+      "step": 782
+    },
+    {
+      "epoch": 0.0696,
+      "grad_norm": 79.0595932006836,
+      "learning_rate": 3e-06,
+      "loss": 16.2649,
+      "step": 783
+    },
+    {
+      "epoch": 0.06968888888888888,
+      "grad_norm": 151.0120086669922,
+      "learning_rate": 3e-06,
+      "loss": 13.9466,
+      "step": 784
+    },
+    {
+      "epoch": 0.06977777777777777,
+      "grad_norm": 84.83601379394531,
+      "learning_rate": 3e-06,
+      "loss": 13.7201,
+      "step": 785
+    },
+    {
+      "epoch": 0.06986666666666666,
+      "grad_norm": 149.62045288085938,
+      "learning_rate": 3e-06,
+      "loss": 13.032,
+      "step": 786
+    },
+    {
+      "epoch": 0.06995555555555556,
+      "grad_norm": 95.56779479980469,
+      "learning_rate": 3e-06,
+      "loss": 8.1661,
+      "step": 787
+    },
+    {
+      "epoch": 0.07004444444444445,
+      "grad_norm": 89.89787292480469,
+      "learning_rate": 3e-06,
+      "loss": 19.6015,
+      "step": 788
+    },
+    {
+      "epoch": 0.07013333333333334,
+      "grad_norm": 78.87327575683594,
+      "learning_rate": 3e-06,
+      "loss": 15.0965,
+      "step": 789
+    },
+    {
+      "epoch": 0.07022222222222223,
+      "grad_norm": 88.94290161132812,
+      "learning_rate": 3e-06,
+      "loss": 12.3248,
+      "step": 790
+    },
+    {
+      "epoch": 0.07031111111111112,
+      "grad_norm": 84.62553405761719,
+      "learning_rate": 3e-06,
+      "loss": 12.571,
+      "step": 791
+    },
+    {
+      "epoch": 0.0704,
+      "grad_norm": 102.37629699707031,
+      "learning_rate": 3e-06,
+      "loss": 11.5955,
+      "step": 792
+    },
+    {
+      "completion_length": 245.95834350585938,
+      "epoch": 0.07048888888888889,
+      "grad_norm": 87.68032836914062,
+      "learning_rate": 3e-06,
+      "loss": -2.896,
+      "reward": 0.979166716337204,
+      "reward_std": 0.3332235962152481,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5625,
+      "rewards/correctness_reward_func_math": 0.416666679084301,
+      "step": 793,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.07057777777777778,
+      "grad_norm": 102.35977935791016,
+      "learning_rate": 3e-06,
+      "loss": -14.1397,
+      "step": 794
+    },
+    {
+      "epoch": 0.07066666666666667,
+      "grad_norm": 121.52274322509766,
+      "learning_rate": 3e-06,
+      "loss": -20.6379,
+      "step": 795
+    },
+    {
+      "epoch": 0.07075555555555556,
+      "grad_norm": 123.77271270751953,
+      "learning_rate": 3e-06,
+      "loss": -15.954,
+      "step": 796
+    },
+    {
+      "epoch": 0.07084444444444445,
+      "grad_norm": 115.74909210205078,
+      "learning_rate": 3e-06,
+      "loss": -19.257,
+      "step": 797
+    },
+    {
+      "epoch": 0.07093333333333333,
+      "grad_norm": 171.18011474609375,
+      "learning_rate": 3e-06,
+      "loss": -15.1334,
+      "step": 798
+    },
+    {
+      "epoch": 0.07102222222222222,
+      "grad_norm": 92.46512603759766,
+      "learning_rate": 3e-06,
+      "loss": -3.7242,
+      "step": 799
+    },
+    {
+      "epoch": 0.07111111111111111,
+      "grad_norm": 104.15264129638672,
+      "learning_rate": 3e-06,
+      "loss": -14.9942,
+      "step": 800
+    },
+    {
+      "epoch": 0.0712,
+      "grad_norm": 120.8930892944336,
+      "learning_rate": 3e-06,
+      "loss": -22.0188,
+      "step": 801
+    },
+    {
+      "epoch": 0.07128888888888889,
+      "grad_norm": 112.13275146484375,
+      "learning_rate": 3e-06,
+      "loss": -17.6709,
+      "step": 802
+    },
+    {
+      "epoch": 0.07137777777777778,
+      "grad_norm": 136.23388671875,
+      "learning_rate": 3e-06,
+      "loss": -20.6627,
+      "step": 803
+    },
+    {
+      "epoch": 0.07146666666666666,
+      "grad_norm": 124.1773681640625,
+      "learning_rate": 3e-06,
+      "loss": -17.0962,
+      "step": 804
+    },
+    {
+      "completion_length": 232.89583587646484,
+      "epoch": 0.07155555555555555,
+      "grad_norm": 91.88973236083984,
+      "learning_rate": 3e-06,
+      "loss": -7.8492,
+      "reward": 1.7291666865348816,
+      "reward_std": 0.47015421837568283,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 1.0416666567325592,
+      "step": 805,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.07164444444444444,
+      "grad_norm": 107.12287139892578,
+      "learning_rate": 3e-06,
+      "loss": -23.4332,
+      "step": 806
+    },
+    {
+      "epoch": 0.07173333333333333,
+      "grad_norm": 104.5981674194336,
+      "learning_rate": 3e-06,
+      "loss": -11.8278,
+      "step": 807
+    },
+    {
+      "epoch": 0.07182222222222222,
+      "grad_norm": 102.6692886352539,
+      "learning_rate": 3e-06,
+      "loss": -20.165,
+      "step": 808
+    },
+    {
+      "epoch": 0.0719111111111111,
+      "grad_norm": 88.46080017089844,
+      "learning_rate": 3e-06,
+      "loss": -10.8484,
+      "step": 809
+    },
+    {
+      "epoch": 0.072,
+      "grad_norm": 124.10685729980469,
+      "learning_rate": 3e-06,
+      "loss": -20.237,
+      "step": 810
+    },
+    {
+      "epoch": 0.07208888888888888,
+      "grad_norm": 95.68196868896484,
+      "learning_rate": 3e-06,
+      "loss": -9.2521,
+      "step": 811
+    },
+    {
+      "epoch": 0.07217777777777777,
+      "grad_norm": 95.62832641601562,
+      "learning_rate": 3e-06,
+      "loss": -25.6333,
+      "step": 812
+    },
+    {
+      "epoch": 0.07226666666666667,
+      "grad_norm": 104.2110366821289,
+      "learning_rate": 3e-06,
+      "loss": -14.2533,
+      "step": 813
+    },
+    {
+      "epoch": 0.07235555555555556,
+      "grad_norm": 114.39372253417969,
+      "learning_rate": 3e-06,
+      "loss": -22.2888,
+      "step": 814
+    },
+    {
+      "epoch": 0.07244444444444445,
+      "grad_norm": 112.07288360595703,
+      "learning_rate": 3e-06,
+      "loss": -12.7028,
+      "step": 815
+    },
+    {
+      "epoch": 0.07253333333333334,
+      "grad_norm": 138.86337280273438,
+      "learning_rate": 3e-06,
+      "loss": -23.0688,
+      "step": 816
+    },
+    {
+      "completion_length": 249.3541717529297,
+      "epoch": 0.07262222222222223,
+      "grad_norm": 106.32572937011719,
+      "learning_rate": 3e-06,
+      "loss": 1.5611,
+      "reward": 1.291666716337204,
+      "reward_std": 0.37455084919929504,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6666666865348816,
+      "rewards/correctness_reward_func_math": 0.625,
+      "step": 817,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.07271111111111112,
+      "grad_norm": 119.1928482055664,
+      "learning_rate": 3e-06,
+      "loss": 6.5715,
+      "step": 818
+    },
+    {
+      "epoch": 0.0728,
+      "grad_norm": 89.53767395019531,
+      "learning_rate": 3e-06,
+      "loss": 9.9694,
+      "step": 819
+    },
+    {
+      "epoch": 0.07288888888888889,
+      "grad_norm": 97.2778091430664,
+      "learning_rate": 3e-06,
+      "loss": 8.1343,
+      "step": 820
+    },
+    {
+      "epoch": 0.07297777777777778,
+      "grad_norm": 99.85116577148438,
+      "learning_rate": 3e-06,
+      "loss": 3.7739,
+      "step": 821
+    },
+    {
+      "epoch": 0.07306666666666667,
+      "grad_norm": 107.6137924194336,
+      "learning_rate": 3e-06,
+      "loss": -0.6852,
+      "step": 822
+    },
+    {
+      "epoch": 0.07315555555555556,
+      "grad_norm": 113.21214294433594,
+      "learning_rate": 3e-06,
+      "loss": 1.024,
+      "step": 823
+    },
+    {
+      "epoch": 0.07324444444444445,
+      "grad_norm": 130.27040100097656,
+      "learning_rate": 3e-06,
+      "loss": 4.4904,
+      "step": 824
+    },
+    {
+      "epoch": 0.07333333333333333,
+      "grad_norm": 87.94723510742188,
+      "learning_rate": 3e-06,
+      "loss": 7.7117,
+      "step": 825
+    },
+    {
+      "epoch": 0.07342222222222222,
+      "grad_norm": 102.87310791015625,
+      "learning_rate": 3e-06,
+      "loss": 5.3932,
+      "step": 826
+    },
+    {
+      "epoch": 0.07351111111111111,
+      "grad_norm": 97.53043365478516,
+      "learning_rate": 3e-06,
+      "loss": 1.1637,
+      "step": 827
+    },
+    {
+      "epoch": 0.0736,
+      "grad_norm": 126.4931640625,
+      "learning_rate": 3e-06,
+      "loss": -3.7025,
+      "step": 828
+    },
+    {
+      "completion_length": 240.75,
+      "epoch": 0.07368888888888889,
+      "grad_norm": 88.22557067871094,
+      "learning_rate": 3e-06,
+      "loss": 22.5183,
+      "reward": 1.1354166865348816,
+      "reward_std": 0.35377833247184753,
+      "rewards/boxed_and_answer_tags_format_reward": 0.59375,
+      "rewards/correctness_reward_func_math": 0.5416666567325592,
+      "step": 829,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.07377777777777778,
+      "grad_norm": 86.47975158691406,
+      "learning_rate": 3e-06,
+      "loss": 24.5943,
+      "step": 830
+    },
+    {
+      "epoch": 0.07386666666666666,
+      "grad_norm": 93.10693359375,
+      "learning_rate": 3e-06,
+      "loss": 27.2039,
+      "step": 831
+    },
+    {
+      "epoch": 0.07395555555555555,
+      "grad_norm": 96.74742889404297,
+      "learning_rate": 3e-06,
+      "loss": 32.9558,
+      "step": 832
+    },
+    {
+      "epoch": 0.07404444444444444,
+      "grad_norm": 97.9085693359375,
+      "learning_rate": 3e-06,
+      "loss": 25.7304,
+      "step": 833
+    },
+    {
+      "epoch": 0.07413333333333333,
+      "grad_norm": 95.80497741699219,
+      "learning_rate": 3e-06,
+      "loss": 28.8338,
+      "step": 834
+    },
+    {
+      "epoch": 0.07422222222222222,
+      "grad_norm": 82.9104995727539,
+      "learning_rate": 3e-06,
+      "loss": 20.9569,
+      "step": 835
+    },
+    {
+      "epoch": 0.0743111111111111,
+      "grad_norm": 332.6025695800781,
+      "learning_rate": 3e-06,
+      "loss": 22.7693,
+      "step": 836
+    },
+    {
+      "epoch": 0.0744,
+      "grad_norm": 97.00851440429688,
+      "learning_rate": 3e-06,
+      "loss": 24.6048,
+      "step": 837
+    },
+    {
+      "epoch": 0.07448888888888888,
+      "grad_norm": 90.94817352294922,
+      "learning_rate": 3e-06,
+      "loss": 30.2657,
+      "step": 838
+    },
+    {
+      "epoch": 0.07457777777777778,
+      "grad_norm": 91.87737274169922,
+      "learning_rate": 3e-06,
+      "loss": 23.8124,
+      "step": 839
+    },
+    {
+      "epoch": 0.07466666666666667,
+      "grad_norm": 100.71826934814453,
+      "learning_rate": 3e-06,
+      "loss": 26.5346,
+      "step": 840
+    },
+    {
+      "completion_length": 244.3125,
+      "epoch": 0.07475555555555556,
+      "grad_norm": 454.38482666015625,
+      "learning_rate": 3e-06,
+      "loss": -16.9865,
+      "reward": 1.2812500596046448,
+      "reward_std": 0.3782803416252136,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.6666666492819786,
+      "step": 841,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.07484444444444445,
+      "grad_norm": 102.08020782470703,
+      "learning_rate": 3e-06,
+      "loss": -18.5601,
+      "step": 842
+    },
+    {
+      "epoch": 0.07493333333333334,
+      "grad_norm": 101.7997055053711,
+      "learning_rate": 3e-06,
+      "loss": -23.9473,
+      "step": 843
+    },
+    {
+      "epoch": 0.07502222222222223,
+      "grad_norm": 100.2668685913086,
+      "learning_rate": 3e-06,
+      "loss": -26.2402,
+      "step": 844
+    },
+    {
+      "epoch": 0.07511111111111111,
+      "grad_norm": 119.95198059082031,
+      "learning_rate": 3e-06,
+      "loss": -22.5011,
+      "step": 845
+    },
+    {
+      "epoch": 0.0752,
+      "grad_norm": 101.83861541748047,
+      "learning_rate": 3e-06,
+      "loss": -14.7265,
+      "step": 846
+    },
+    {
+      "epoch": 0.07528888888888889,
+      "grad_norm": 106.50312042236328,
+      "learning_rate": 3e-06,
+      "loss": -18.8046,
+      "step": 847
+    },
+    {
+      "epoch": 0.07537777777777778,
+      "grad_norm": 114.58135986328125,
+      "learning_rate": 3e-06,
+      "loss": -20.037,
+      "step": 848
+    },
+    {
+      "epoch": 0.07546666666666667,
+      "grad_norm": 121.03673553466797,
+      "learning_rate": 3e-06,
+      "loss": -25.817,
+      "step": 849
+    },
+    {
+      "epoch": 0.07555555555555556,
+      "grad_norm": 97.15817260742188,
+      "learning_rate": 3e-06,
+      "loss": -28.7689,
+      "step": 850
+    },
+    {
+      "epoch": 0.07564444444444444,
+      "grad_norm": 101.58541107177734,
+      "learning_rate": 3e-06,
+      "loss": -24.6109,
+      "step": 851
+    },
+    {
+      "epoch": 0.07573333333333333,
+      "grad_norm": 100.84017181396484,
+      "learning_rate": 3e-06,
+      "loss": -16.8195,
+      "step": 852
+    },
+    {
+      "completion_length": 247.4166717529297,
+      "epoch": 0.07582222222222222,
+      "grad_norm": 60.974788665771484,
+      "learning_rate": 3e-06,
+      "loss": 1.297,
+      "reward": 1.2708333432674408,
+      "reward_std": 0.12909945845603943,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.5833333358168602,
+      "step": 853,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.07591111111111111,
+      "grad_norm": 58.498374938964844,
+      "learning_rate": 3e-06,
+      "loss": -4.5823,
+      "step": 854
+    },
+    {
+      "epoch": 0.076,
+      "grad_norm": 51.417320251464844,
+      "learning_rate": 3e-06,
+      "loss": -2.8582,
+      "step": 855
+    },
+    {
+      "epoch": 0.07608888888888889,
+      "grad_norm": 52.74491882324219,
+      "learning_rate": 3e-06,
+      "loss": -3.5933,
+      "step": 856
+    },
+    {
+      "epoch": 0.07617777777777777,
+      "grad_norm": 54.02571487426758,
+      "learning_rate": 3e-06,
+      "loss": -0.1777,
+      "step": 857
+    },
+    {
+      "epoch": 0.07626666666666666,
+      "grad_norm": 44.29707717895508,
+      "learning_rate": 3e-06,
+      "loss": -0.0581,
+      "step": 858
+    },
+    {
+      "epoch": 0.07635555555555555,
+      "grad_norm": 63.4464111328125,
+      "learning_rate": 3e-06,
+      "loss": 0.981,
+      "step": 859
+    },
+    {
+      "epoch": 0.07644444444444444,
+      "grad_norm": 58.98569869995117,
+      "learning_rate": 3e-06,
+      "loss": -5.3316,
+      "step": 860
+    },
+    {
+      "epoch": 0.07653333333333333,
+      "grad_norm": 54.73743438720703,
+      "learning_rate": 3e-06,
+      "loss": -3.799,
+      "step": 861
+    },
+    {
+      "epoch": 0.07662222222222222,
+      "grad_norm": 55.15388107299805,
+      "learning_rate": 3e-06,
+      "loss": -4.4757,
+      "step": 862
+    },
+    {
+      "epoch": 0.0767111111111111,
+      "grad_norm": 61.510887145996094,
+      "learning_rate": 3e-06,
+      "loss": -0.6125,
+      "step": 863
+    },
+    {
+      "epoch": 0.0768,
+      "grad_norm": 46.19833755493164,
+      "learning_rate": 3e-06,
+      "loss": -0.7855,
+      "step": 864
+    },
+    {
+      "completion_length": 251.3125,
+      "epoch": 0.0768888888888889,
+      "grad_norm": 111.38273620605469,
+      "learning_rate": 3e-06,
+      "loss": 14.4732,
+      "reward": 1.5208333730697632,
+      "reward_std": 0.4937378317117691,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7291666567325592,
+      "rewards/correctness_reward_func_math": 0.7916666567325592,
+      "step": 865,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.07697777777777778,
+      "grad_norm": 112.39920043945312,
+      "learning_rate": 3e-06,
+      "loss": 3.7967,
+      "step": 866
+    },
+    {
+      "epoch": 0.07706666666666667,
+      "grad_norm": 106.71125793457031,
+      "learning_rate": 3e-06,
+      "loss": 1.1063,
+      "step": 867
+    },
+    {
+      "epoch": 0.07715555555555556,
+      "grad_norm": 129.4515838623047,
+      "learning_rate": 3e-06,
+      "loss": -2.7262,
+      "step": 868
+    },
+    {
+      "epoch": 0.07724444444444445,
+      "grad_norm": 109.67815399169922,
+      "learning_rate": 3e-06,
+      "loss": 0.1256,
+      "step": 869
+    },
+    {
+      "epoch": 0.07733333333333334,
+      "grad_norm": 116.57035827636719,
+      "learning_rate": 3e-06,
+      "loss": -2.946,
+      "step": 870
+    },
+    {
+      "epoch": 0.07742222222222223,
+      "grad_norm": 122.09374237060547,
+      "learning_rate": 3e-06,
+      "loss": 13.0142,
+      "step": 871
+    },
+    {
+      "epoch": 0.07751111111111111,
+      "grad_norm": 108.0525894165039,
+      "learning_rate": 3e-06,
+      "loss": 2.4968,
+      "step": 872
+    },
+    {
+      "epoch": 0.0776,
+      "grad_norm": 108.2818374633789,
+      "learning_rate": 3e-06,
+      "loss": -0.8286,
+      "step": 873
+    },
+    {
+      "epoch": 0.07768888888888889,
+      "grad_norm": 139.2396697998047,
+      "learning_rate": 3e-06,
+      "loss": -5.0471,
+      "step": 874
+    },
+    {
+      "epoch": 0.07777777777777778,
+      "grad_norm": 114.9443588256836,
+      "learning_rate": 3e-06,
+      "loss": -1.3382,
+      "step": 875
+    },
+    {
+      "epoch": 0.07786666666666667,
+      "grad_norm": 128.51272583007812,
+      "learning_rate": 3e-06,
+      "loss": -4.8307,
+      "step": 876
+    },
+    {
+      "completion_length": 252.50000762939453,
+      "epoch": 0.07795555555555556,
+      "grad_norm": 126.4562759399414,
+      "learning_rate": 3e-06,
+      "loss": 0.751,
+      "reward": 1.0729166865348816,
+      "reward_std": 0.22548970580101013,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.4583333283662796,
+      "step": 877,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.07804444444444444,
+      "grad_norm": 82.09607696533203,
+      "learning_rate": 3e-06,
+      "loss": 0.195,
+      "step": 878
+    },
+    {
+      "epoch": 0.07813333333333333,
+      "grad_norm": 74.75113677978516,
+      "learning_rate": 3e-06,
+      "loss": -1.5817,
+      "step": 879
+    },
+    {
+      "epoch": 0.07822222222222222,
+      "grad_norm": 115.73063659667969,
+      "learning_rate": 3e-06,
+      "loss": -8.3706,
+      "step": 880
+    },
+    {
+      "epoch": 0.07831111111111111,
+      "grad_norm": 70.39916229248047,
+      "learning_rate": 3e-06,
+      "loss": -3.2191,
+      "step": 881
+    },
+    {
+      "epoch": 0.0784,
+      "grad_norm": 103.28494262695312,
+      "learning_rate": 3e-06,
+      "loss": 1.0954,
+      "step": 882
+    },
+    {
+      "epoch": 0.07848888888888889,
+      "grad_norm": 100.23104858398438,
+      "learning_rate": 3e-06,
+      "loss": 0.0428,
+      "step": 883
+    },
+    {
+      "epoch": 0.07857777777777777,
+      "grad_norm": 90.33434295654297,
+      "learning_rate": 3e-06,
+      "loss": -0.8422,
+      "step": 884
+    },
+    {
+      "epoch": 0.07866666666666666,
+      "grad_norm": 99.41636657714844,
+      "learning_rate": 3e-06,
+      "loss": -2.558,
+      "step": 885
+    },
+    {
+      "epoch": 0.07875555555555555,
+      "grad_norm": 99.72211456298828,
+      "learning_rate": 3e-06,
+      "loss": -8.9057,
+      "step": 886
+    },
+    {
+      "epoch": 0.07884444444444444,
+      "grad_norm": 73.4344711303711,
+      "learning_rate": 3e-06,
+      "loss": -4.09,
+      "step": 887
+    },
+    {
+      "epoch": 0.07893333333333333,
+      "grad_norm": 98.25971221923828,
+      "learning_rate": 3e-06,
+      "loss": -0.4152,
+      "step": 888
+    },
+    {
+      "completion_length": 249.2916717529297,
+      "epoch": 0.07902222222222222,
+      "grad_norm": 165.06918334960938,
+      "learning_rate": 3e-06,
+      "loss": 25.8684,
+      "reward": 1.4270833730697632,
+      "reward_std": 0.5354157984256744,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 889,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.0791111111111111,
+      "grad_norm": 132.98858642578125,
+      "learning_rate": 3e-06,
+      "loss": 19.4902,
+      "step": 890
+    },
+    {
+      "epoch": 0.0792,
+      "grad_norm": 122.3958969116211,
+      "learning_rate": 3e-06,
+      "loss": 32.6103,
+      "step": 891
+    },
+    {
+      "epoch": 0.0792888888888889,
+      "grad_norm": 118.38285064697266,
+      "learning_rate": 3e-06,
+      "loss": 32.6835,
+      "step": 892
+    },
+    {
+      "epoch": 0.07937777777777778,
+      "grad_norm": 126.4738540649414,
+      "learning_rate": 3e-06,
+      "loss": 30.5181,
+      "step": 893
+    },
+    {
+      "epoch": 0.07946666666666667,
+      "grad_norm": 128.20831298828125,
+      "learning_rate": 3e-06,
+      "loss": 33.4332,
+      "step": 894
+    },
+    {
+      "epoch": 0.07955555555555556,
+      "grad_norm": 152.9354248046875,
+      "learning_rate": 3e-06,
+      "loss": 23.7035,
+      "step": 895
+    },
+    {
+      "epoch": 0.07964444444444445,
+      "grad_norm": 130.27053833007812,
+      "learning_rate": 3e-06,
+      "loss": 16.7548,
+      "step": 896
+    },
+    {
+      "epoch": 0.07973333333333334,
+      "grad_norm": 127.17219543457031,
+      "learning_rate": 3e-06,
+      "loss": 30.7848,
+      "step": 897
+    },
+    {
+      "epoch": 0.07982222222222222,
+      "grad_norm": 118.670654296875,
+      "learning_rate": 3e-06,
+      "loss": 30.7772,
+      "step": 898
+    },
+    {
+      "epoch": 0.07991111111111111,
+      "grad_norm": 120.19160461425781,
+      "learning_rate": 3e-06,
+      "loss": 27.401,
+      "step": 899
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 137.2371063232422,
+      "learning_rate": 3e-06,
+      "loss": 30.9334,
+      "step": 900
+    },
+    {
+      "completion_length": 255.89583587646484,
+      "epoch": 0.08008888888888889,
+      "grad_norm": 88.54483032226562,
+      "learning_rate": 3e-06,
+      "loss": 3.6467,
+      "reward": 1.3437500596046448,
+      "reward_std": 0.30770808458328247,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.6666666567325592,
+      "step": 901,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.08017777777777778,
+      "grad_norm": 87.48735046386719,
+      "learning_rate": 3e-06,
+      "loss": 1.8615,
+      "step": 902
+    },
+    {
+      "epoch": 0.08026666666666667,
+      "grad_norm": 86.97764587402344,
+      "learning_rate": 3e-06,
+      "loss": 2.8386,
+      "step": 903
+    },
+    {
+      "epoch": 0.08035555555555556,
+      "grad_norm": 105.64205932617188,
+      "learning_rate": 3e-06,
+      "loss": -3.9129,
+      "step": 904
+    },
+    {
+      "epoch": 0.08044444444444444,
+      "grad_norm": 80.34683227539062,
+      "learning_rate": 3e-06,
+      "loss": 1.4293,
+      "step": 905
+    },
+    {
+      "epoch": 0.08053333333333333,
+      "grad_norm": 91.16341400146484,
+      "learning_rate": 3e-06,
+      "loss": 2.803,
+      "step": 906
+    },
+    {
+      "epoch": 0.08062222222222222,
+      "grad_norm": 96.49407196044922,
+      "learning_rate": 3e-06,
+      "loss": 2.4431,
+      "step": 907
+    },
+    {
+      "epoch": 0.08071111111111111,
+      "grad_norm": 84.40055084228516,
+      "learning_rate": 3e-06,
+      "loss": 0.5478,
+      "step": 908
+    },
+    {
+      "epoch": 0.0808,
+      "grad_norm": 78.79622650146484,
+      "learning_rate": 3e-06,
+      "loss": 1.4422,
+      "step": 909
+    },
+    {
+      "epoch": 0.08088888888888889,
+      "grad_norm": 128.47531127929688,
+      "learning_rate": 3e-06,
+      "loss": -5.1302,
+      "step": 910
+    },
+    {
+      "epoch": 0.08097777777777777,
+      "grad_norm": 79.19956970214844,
+      "learning_rate": 3e-06,
+      "loss": -0.2354,
+      "step": 911
+    },
+    {
+      "epoch": 0.08106666666666666,
+      "grad_norm": 107.92975616455078,
+      "learning_rate": 3e-06,
+      "loss": 1.3494,
+      "step": 912
+    },
+    {
+      "completion_length": 236.7916717529297,
+      "epoch": 0.08115555555555555,
+      "grad_norm": 72.56483459472656,
+      "learning_rate": 3e-06,
+      "loss": -4.7678,
+      "reward": 1.0,
+      "reward_std": 0.23116151988506317,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 913,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.08124444444444444,
+      "grad_norm": 84.60346984863281,
+      "learning_rate": 3e-06,
+      "loss": -1.6215,
+      "step": 914
+    },
+    {
+      "epoch": 0.08133333333333333,
+      "grad_norm": 80.78106689453125,
+      "learning_rate": 3e-06,
+      "loss": 3.2587,
+      "step": 915
+    },
+    {
+      "epoch": 0.08142222222222223,
+      "grad_norm": 71.9332275390625,
+      "learning_rate": 3e-06,
+      "loss": -4.7685,
+      "step": 916
+    },
+    {
+      "epoch": 0.08151111111111112,
+      "grad_norm": 98.66748046875,
+      "learning_rate": 3e-06,
+      "loss": -6.589,
+      "step": 917
+    },
+    {
+      "epoch": 0.0816,
+      "grad_norm": 135.69175720214844,
+      "learning_rate": 3e-06,
+      "loss": -7.5017,
+      "step": 918
+    },
+    {
+      "epoch": 0.0816888888888889,
+      "grad_norm": 73.60437774658203,
+      "learning_rate": 3e-06,
+      "loss": -5.1495,
+      "step": 919
+    },
+    {
+      "epoch": 0.08177777777777778,
+      "grad_norm": 90.75928497314453,
+      "learning_rate": 3e-06,
+      "loss": -2.384,
+      "step": 920
+    },
+    {
+      "epoch": 0.08186666666666667,
+      "grad_norm": 82.05548095703125,
+      "learning_rate": 3e-06,
+      "loss": 2.8112,
+      "step": 921
+    },
+    {
+      "epoch": 0.08195555555555556,
+      "grad_norm": 77.72819519042969,
+      "learning_rate": 3e-06,
+      "loss": -5.4085,
+      "step": 922
+    },
+    {
+      "epoch": 0.08204444444444445,
+      "grad_norm": 100.81270599365234,
+      "learning_rate": 3e-06,
+      "loss": -7.311,
+      "step": 923
+    },
+    {
+      "epoch": 0.08213333333333334,
+      "grad_norm": 86.93998718261719,
+      "learning_rate": 3e-06,
+      "loss": -7.627,
+      "step": 924
+    },
+    {
+      "completion_length": 244.2916717529297,
+      "epoch": 0.08222222222222222,
+      "grad_norm": 124.40040588378906,
+      "learning_rate": 3e-06,
+      "loss": 2.5607,
+      "reward": 1.7708333730697632,
+      "reward_std": 0.5440726578235626,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7291666567325592,
+      "rewards/correctness_reward_func_math": 1.0416666567325592,
+      "step": 925,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.08231111111111111,
+      "grad_norm": 100.88472747802734,
+      "learning_rate": 3e-06,
+      "loss": -11.1251,
+      "step": 926
+    },
+    {
+      "epoch": 0.0824,
+      "grad_norm": 139.0868377685547,
+      "learning_rate": 3e-06,
+      "loss": -4.6789,
+      "step": 927
+    },
+    {
+      "epoch": 0.08248888888888889,
+      "grad_norm": 105.37358093261719,
+      "learning_rate": 3e-06,
+      "loss": -1.7812,
+      "step": 928
+    },
+    {
+      "epoch": 0.08257777777777778,
+      "grad_norm": 122.72453308105469,
+      "learning_rate": 3e-06,
+      "loss": 5.9917,
+      "step": 929
+    },
+    {
+      "epoch": 0.08266666666666667,
+      "grad_norm": 125.92015075683594,
+      "learning_rate": 3e-06,
+      "loss": -2.2776,
+      "step": 930
+    },
+    {
+      "epoch": 0.08275555555555555,
+      "grad_norm": 122.41661834716797,
+      "learning_rate": 3e-06,
+      "loss": 0.8417,
+      "step": 931
+    },
+    {
+      "epoch": 0.08284444444444444,
+      "grad_norm": 117.53387451171875,
+      "learning_rate": 3e-06,
+      "loss": -13.1484,
+      "step": 932
+    },
+    {
+      "epoch": 0.08293333333333333,
+      "grad_norm": 126.8305892944336,
+      "learning_rate": 3e-06,
+      "loss": -7.6824,
+      "step": 933
+    },
+    {
+      "epoch": 0.08302222222222222,
+      "grad_norm": 111.0191421508789,
+      "learning_rate": 3e-06,
+      "loss": -3.9524,
+      "step": 934
+    },
+    {
+      "epoch": 0.08311111111111111,
+      "grad_norm": 131.84397888183594,
+      "learning_rate": 3e-06,
+      "loss": 2.9848,
+      "step": 935
+    },
+    {
+      "epoch": 0.0832,
+      "grad_norm": 124.30160522460938,
+      "learning_rate": 3e-06,
+      "loss": -5.5558,
+      "step": 936
+    },
+    {
+      "completion_length": 240.9791717529297,
+      "epoch": 0.08328888888888888,
+      "grad_norm": 137.87579345703125,
+      "learning_rate": 3e-06,
+      "loss": -18.5164,
+      "reward": 1.1354166865348816,
+      "reward_std": 0.4608011841773987,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.4583333432674408,
+      "step": 937,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.08337777777777777,
+      "grad_norm": 132.598388671875,
+      "learning_rate": 3e-06,
+      "loss": -26.5032,
+      "step": 938
+    },
+    {
+      "epoch": 0.08346666666666666,
+      "grad_norm": 147.24671936035156,
+      "learning_rate": 3e-06,
+      "loss": -20.2196,
+      "step": 939
+    },
+    {
+      "epoch": 0.08355555555555555,
+      "grad_norm": 140.57591247558594,
+      "learning_rate": 3e-06,
+      "loss": -19.0462,
+      "step": 940
+    },
+    {
+      "epoch": 0.08364444444444444,
+      "grad_norm": 124.26339721679688,
+      "learning_rate": 3e-06,
+      "loss": -22.6895,
+      "step": 941
+    },
+    {
+      "epoch": 0.08373333333333334,
+      "grad_norm": 126.45221710205078,
+      "learning_rate": 3e-06,
+      "loss": -24.6262,
+      "step": 942
+    },
+    {
+      "epoch": 0.08382222222222223,
+      "grad_norm": 135.6764373779297,
+      "learning_rate": 3e-06,
+      "loss": -20.8866,
+      "step": 943
+    },
+    {
+      "epoch": 0.08391111111111112,
+      "grad_norm": 139.4601287841797,
+      "learning_rate": 3e-06,
+      "loss": -28.928,
+      "step": 944
+    },
+    {
+      "epoch": 0.084,
+      "grad_norm": 173.5882568359375,
+      "learning_rate": 3e-06,
+      "loss": -23.599,
+      "step": 945
+    },
+    {
+      "epoch": 0.0840888888888889,
+      "grad_norm": 131.79933166503906,
+      "learning_rate": 3e-06,
+      "loss": -22.4616,
+      "step": 946
+    },
+    {
+      "epoch": 0.08417777777777778,
+      "grad_norm": 128.2574920654297,
+      "learning_rate": 3e-06,
+      "loss": -26.6084,
+      "step": 947
+    },
+    {
+      "epoch": 0.08426666666666667,
+      "grad_norm": 133.56024169921875,
+      "learning_rate": 3e-06,
+      "loss": -29.0544,
+      "step": 948
+    },
+    {
+      "completion_length": 254.3541717529297,
+      "epoch": 0.08435555555555556,
+      "grad_norm": 113.98381805419922,
+      "learning_rate": 3e-06,
+      "loss": 2.0708,
+      "reward": 1.291666716337204,
+      "reward_std": 0.4701542258262634,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.541666679084301,
+      "step": 949,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.08444444444444445,
+      "grad_norm": 158.377197265625,
+      "learning_rate": 3e-06,
+      "loss": 0.0861,
+      "step": 950
+    },
+    {
+      "epoch": 0.08453333333333334,
+      "grad_norm": 152.6723175048828,
+      "learning_rate": 3e-06,
+      "loss": 8.1982,
+      "step": 951
+    },
+    {
+      "epoch": 0.08462222222222222,
+      "grad_norm": 122.41393280029297,
+      "learning_rate": 3e-06,
+      "loss": -2.2863,
+      "step": 952
+    },
+    {
+      "epoch": 0.08471111111111111,
+      "grad_norm": 137.0810089111328,
+      "learning_rate": 3e-06,
+      "loss": -1.879,
+      "step": 953
+    },
+    {
+      "epoch": 0.0848,
+      "grad_norm": 122.0219497680664,
+      "learning_rate": 3e-06,
+      "loss": -4.6058,
+      "step": 954
+    },
+    {
+      "epoch": 0.08488888888888889,
+      "grad_norm": 122.55841064453125,
+      "learning_rate": 3e-06,
+      "loss": 0.6988,
+      "step": 955
+    },
+    {
+      "epoch": 0.08497777777777778,
+      "grad_norm": 144.296875,
+      "learning_rate": 3e-06,
+      "loss": -2.2706,
+      "step": 956
+    },
+    {
+      "epoch": 0.08506666666666667,
+      "grad_norm": 130.82684326171875,
+      "learning_rate": 3e-06,
+      "loss": 6.122,
+      "step": 957
+    },
+    {
+      "epoch": 0.08515555555555555,
+      "grad_norm": 121.61994934082031,
+      "learning_rate": 3e-06,
+      "loss": -4.2647,
+      "step": 958
+    },
+    {
+      "epoch": 0.08524444444444444,
+      "grad_norm": 124.73589324951172,
+      "learning_rate": 3e-06,
+      "loss": -4.8135,
+      "step": 959
+    },
+    {
+      "epoch": 0.08533333333333333,
+      "grad_norm": 109.87874603271484,
+      "learning_rate": 3e-06,
+      "loss": -7.1554,
+      "step": 960
+    },
+    {
+      "completion_length": 244.50000762939453,
+      "epoch": 0.08542222222222222,
+      "grad_norm": 74.7403793334961,
+      "learning_rate": 3e-06,
+      "loss": 9.6961,
+      "reward": 2.1666667461395264,
+      "reward_std": 0.20412414520978928,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.4166666269302368,
+      "step": 961,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.08551111111111111,
+      "grad_norm": 68.04969024658203,
+      "learning_rate": 3e-06,
+      "loss": 12.6362,
+      "step": 962
+    },
+    {
+      "epoch": 0.0856,
+      "grad_norm": 69.76187133789062,
+      "learning_rate": 3e-06,
+      "loss": 13.3129,
+      "step": 963
+    },
+    {
+      "epoch": 0.08568888888888888,
+      "grad_norm": 81.5944595336914,
+      "learning_rate": 3e-06,
+      "loss": 9.7628,
+      "step": 964
+    },
+    {
+      "epoch": 0.08577777777777777,
+      "grad_norm": 83.20171356201172,
+      "learning_rate": 3e-06,
+      "loss": 8.2983,
+      "step": 965
+    },
+    {
+      "epoch": 0.08586666666666666,
+      "grad_norm": 78.74623107910156,
+      "learning_rate": 3e-06,
+      "loss": 3.5912,
+      "step": 966
+    },
+    {
+      "epoch": 0.08595555555555555,
+      "grad_norm": 65.22360229492188,
+      "learning_rate": 3e-06,
+      "loss": 8.6179,
+      "step": 967
+    },
+    {
+      "epoch": 0.08604444444444445,
+      "grad_norm": 67.84490966796875,
+      "learning_rate": 3e-06,
+      "loss": 10.9563,
+      "step": 968
+    },
+    {
+      "epoch": 0.08613333333333334,
+      "grad_norm": 66.93883514404297,
+      "learning_rate": 3e-06,
+      "loss": 11.5826,
+      "step": 969
+    },
+    {
+      "epoch": 0.08622222222222223,
+      "grad_norm": 75.27574157714844,
+      "learning_rate": 3e-06,
+      "loss": 8.5011,
+      "step": 970
+    },
+    {
+      "epoch": 0.08631111111111112,
+      "grad_norm": 68.24022674560547,
+      "learning_rate": 3e-06,
+      "loss": 6.8511,
+      "step": 971
+    },
+    {
+      "epoch": 0.0864,
+      "grad_norm": 70.8918685913086,
+      "learning_rate": 3e-06,
+      "loss": 2.4563,
+      "step": 972
+    },
+    {
+      "completion_length": 251.0416717529297,
+      "epoch": 0.08648888888888889,
+      "grad_norm": 130.6646270751953,
+      "learning_rate": 3e-06,
+      "loss": -5.8085,
+      "reward": 1.4479166865348816,
+      "reward_std": 0.5305383503437042,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.708333358168602,
+      "step": 973,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.08657777777777778,
+      "grad_norm": 134.2749786376953,
+      "learning_rate": 3e-06,
+      "loss": 1.4252,
+      "step": 974
+    },
+    {
+      "epoch": 0.08666666666666667,
+      "grad_norm": 137.93409729003906,
+      "learning_rate": 3e-06,
+      "loss": -1.3707,
+      "step": 975
+    },
+    {
+      "epoch": 0.08675555555555556,
+      "grad_norm": 121.03262329101562,
+      "learning_rate": 3e-06,
+      "loss": 1.4106,
+      "step": 976
+    },
+    {
+      "epoch": 0.08684444444444445,
+      "grad_norm": 132.82774353027344,
+      "learning_rate": 3e-06,
+      "loss": -4.6112,
+      "step": 977
+    },
+    {
+      "epoch": 0.08693333333333333,
+      "grad_norm": 163.79421997070312,
+      "learning_rate": 3e-06,
+      "loss": -9.213,
+      "step": 978
+    },
+    {
+      "epoch": 0.08702222222222222,
+      "grad_norm": 135.431396484375,
+      "learning_rate": 3e-06,
+      "loss": -9.618,
+      "step": 979
+    },
+    {
+      "epoch": 0.08711111111111111,
+      "grad_norm": 130.899658203125,
+      "learning_rate": 3e-06,
+      "loss": -2.0931,
+      "step": 980
+    },
+    {
+      "epoch": 0.0872,
+      "grad_norm": 133.3518524169922,
+      "learning_rate": 3e-06,
+      "loss": -5.8637,
+      "step": 981
+    },
+    {
+      "epoch": 0.08728888888888889,
+      "grad_norm": 144.94068908691406,
+      "learning_rate": 3e-06,
+      "loss": -2.6211,
+      "step": 982
+    },
+    {
+      "epoch": 0.08737777777777778,
+      "grad_norm": 141.72738647460938,
+      "learning_rate": 3e-06,
+      "loss": -9.573,
+      "step": 983
+    },
+    {
+      "epoch": 0.08746666666666666,
+      "grad_norm": 148.67123413085938,
+      "learning_rate": 3e-06,
+      "loss": -12.8677,
+      "step": 984
+    },
+    {
+      "completion_length": 255.89583587646484,
+      "epoch": 0.08755555555555555,
+      "grad_norm": 148.83518981933594,
+      "learning_rate": 3e-06,
+      "loss": -2.2086,
+      "reward": 1.3958333432674408,
+      "reward_std": 0.48216672986745834,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7291666865348816,
+      "rewards/correctness_reward_func_math": 0.6666666865348816,
+      "step": 985,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.08764444444444444,
+      "grad_norm": 141.38470458984375,
+      "learning_rate": 3e-06,
+      "loss": -5.2298,
+      "step": 986
+    },
+    {
+      "epoch": 0.08773333333333333,
+      "grad_norm": 114.79426574707031,
+      "learning_rate": 3e-06,
+      "loss": 0.6894,
+      "step": 987
+    },
+    {
+      "epoch": 0.08782222222222222,
+      "grad_norm": 137.67416381835938,
+      "learning_rate": 3e-06,
+      "loss": 0.8801,
+      "step": 988
+    },
+    {
+      "epoch": 0.0879111111111111,
+      "grad_norm": 138.06517028808594,
+      "learning_rate": 3e-06,
+      "loss": -1.6148,
+      "step": 989
+    },
+    {
+      "epoch": 0.088,
+      "grad_norm": 153.39608764648438,
+      "learning_rate": 3e-06,
+      "loss": 0.5819,
+      "step": 990
+    },
+    {
+      "epoch": 0.08808888888888888,
+      "grad_norm": 159.18431091308594,
+      "learning_rate": 3e-06,
+      "loss": -3.8668,
+      "step": 991
+    },
+    {
+      "epoch": 0.08817777777777777,
+      "grad_norm": 141.29696655273438,
+      "learning_rate": 3e-06,
+      "loss": -7.4773,
+      "step": 992
+    },
+    {
+      "epoch": 0.08826666666666666,
+      "grad_norm": 129.83058166503906,
+      "learning_rate": 3e-06,
+      "loss": -1.6063,
+      "step": 993
+    },
+    {
+      "epoch": 0.08835555555555556,
+      "grad_norm": 133.49261474609375,
+      "learning_rate": 3e-06,
+      "loss": -0.5547,
+      "step": 994
+    },
+    {
+      "epoch": 0.08844444444444445,
+      "grad_norm": 125.11674499511719,
+      "learning_rate": 3e-06,
+      "loss": -3.5547,
+      "step": 995
+    },
+    {
+      "epoch": 0.08853333333333334,
+      "grad_norm": 113.9607925415039,
+      "learning_rate": 3e-06,
+      "loss": -1.1078,
+      "step": 996
+    },
+    {
+      "completion_length": 254.1041717529297,
+      "epoch": 0.08862222222222223,
+      "grad_norm": 112.0015869140625,
+      "learning_rate": 3e-06,
+      "loss": 0.6874,
+      "reward": 1.6666666865348816,
+      "reward_std": 0.3557328134775162,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7083333432674408,
+      "rewards/correctness_reward_func_math": 0.9583333432674408,
+      "step": 997,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.08871111111111112,
+      "grad_norm": 117.1604232788086,
+      "learning_rate": 3e-06,
+      "loss": 10.6509,
+      "step": 998
+    },
+    {
+      "epoch": 0.0888,
+      "grad_norm": 97.7497787475586,
+      "learning_rate": 3e-06,
+      "loss": 1.431,
+      "step": 999
+    },
+    {
+      "epoch": 0.08888888888888889,
+      "grad_norm": 99.68553161621094,
+      "learning_rate": 3e-06,
+      "loss": 9.1808,
+      "step": 1000
+    },
+    {
+      "epoch": 0.08897777777777778,
+      "grad_norm": 127.76787567138672,
+      "learning_rate": 3e-06,
+      "loss": 7.5762,
+      "step": 1001
+    },
+    {
+      "epoch": 0.08906666666666667,
+      "grad_norm": 93.00830841064453,
+      "learning_rate": 3e-06,
+      "loss": 5.1825,
+      "step": 1002
+    },
+    {
+      "epoch": 0.08915555555555556,
+      "grad_norm": 100.87223815917969,
+      "learning_rate": 3e-06,
+      "loss": -0.2862,
+      "step": 1003
+    },
+    {
+      "epoch": 0.08924444444444445,
+      "grad_norm": 125.37133026123047,
+      "learning_rate": 3e-06,
+      "loss": 9.2038,
+      "step": 1004
+    },
+    {
+      "epoch": 0.08933333333333333,
+      "grad_norm": 107.84559631347656,
+      "learning_rate": 3e-06,
+      "loss": -0.0948,
+      "step": 1005
+    },
+    {
+      "epoch": 0.08942222222222222,
+      "grad_norm": 126.59029388427734,
+      "learning_rate": 3e-06,
+      "loss": 7.2332,
+      "step": 1006
+    },
+    {
+      "epoch": 0.08951111111111111,
+      "grad_norm": 120.74652099609375,
+      "learning_rate": 3e-06,
+      "loss": 6.3094,
+      "step": 1007
+    },
+    {
+      "epoch": 0.0896,
+      "grad_norm": 94.37996673583984,
+      "learning_rate": 3e-06,
+      "loss": 3.469,
+      "step": 1008
+    },
+    {
+      "completion_length": 242.25000762939453,
+      "epoch": 0.08968888888888889,
+      "grad_norm": 61.877044677734375,
+      "learning_rate": 3e-06,
+      "loss": 0.3003,
+      "reward": 1.0104166865348816,
+      "reward_std": 0.11467799544334412,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.3333333358168602,
+      "step": 1009,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.08977777777777778,
+      "grad_norm": 48.69044494628906,
+      "learning_rate": 3e-06,
+      "loss": -1.93,
+      "step": 1010
+    },
+    {
+      "epoch": 0.08986666666666666,
+      "grad_norm": 62.56666946411133,
+      "learning_rate": 3e-06,
+      "loss": 0.7216,
+      "step": 1011
+    },
+    {
+      "epoch": 0.08995555555555555,
+      "grad_norm": 54.1674690246582,
+      "learning_rate": 3e-06,
+      "loss": -1.7886,
+      "step": 1012
+    },
+    {
+      "epoch": 0.09004444444444444,
+      "grad_norm": 52.60224533081055,
+      "learning_rate": 3e-06,
+      "loss": 0.5879,
+      "step": 1013
+    },
+    {
+      "epoch": 0.09013333333333333,
+      "grad_norm": 45.58321762084961,
+      "learning_rate": 3e-06,
+      "loss": 0.5204,
+      "step": 1014
+    },
+    {
+      "epoch": 0.09022222222222222,
+      "grad_norm": 57.8793830871582,
+      "learning_rate": 3e-06,
+      "loss": -0.4941,
+      "step": 1015
+    },
+    {
+      "epoch": 0.0903111111111111,
+      "grad_norm": 51.80791091918945,
+      "learning_rate": 3e-06,
+      "loss": -2.7426,
+      "step": 1016
+    },
+    {
+      "epoch": 0.0904,
+      "grad_norm": 56.86159896850586,
+      "learning_rate": 3e-06,
+      "loss": -0.3923,
+      "step": 1017
+    },
+    {
+      "epoch": 0.09048888888888888,
+      "grad_norm": 48.4435920715332,
+      "learning_rate": 3e-06,
+      "loss": -3.0717,
+      "step": 1018
+    },
+    {
+      "epoch": 0.09057777777777777,
+      "grad_norm": 52.369598388671875,
+      "learning_rate": 3e-06,
+      "loss": -0.6683,
+      "step": 1019
+    },
+    {
+      "epoch": 0.09066666666666667,
+      "grad_norm": 45.13036346435547,
+      "learning_rate": 3e-06,
+      "loss": -0.8826,
+      "step": 1020
+    },
+    {
+      "completion_length": 251.9166717529297,
+      "epoch": 0.09075555555555556,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "reward": 1.5,
+      "reward_std": 0.0,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 1021,
+      "zero_std_ratio": 1.0
+    },
+    {
+      "epoch": 0.09084444444444445,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1022
+    },
+    {
+      "epoch": 0.09093333333333334,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1023
+    },
+    {
+      "epoch": 0.09102222222222223,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1024
+    },
+    {
+      "epoch": 0.09111111111111111,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1025
+    },
+    {
+      "epoch": 0.0912,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1026
+    },
+    {
+      "epoch": 0.09128888888888889,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1027
+    },
+    {
+      "epoch": 0.09137777777777778,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1028
+    },
+    {
+      "epoch": 0.09146666666666667,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1029
+    },
+    {
+      "epoch": 0.09155555555555556,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1030
+    },
+    {
+      "epoch": 0.09164444444444445,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1031
+    },
+    {
+      "epoch": 0.09173333333333333,
+      "grad_norm": 0.0,
+      "learning_rate": 3e-06,
+      "loss": 0.0,
+      "step": 1032
+    },
+    {
+      "completion_length": 252.3541717529297,
+      "epoch": 0.09182222222222222,
+      "grad_norm": 126.83654022216797,
+      "learning_rate": 3e-06,
+      "loss": -8.3335,
+      "reward": 1.625,
+      "reward_std": 0.3410547971725464,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.875,
+      "step": 1033,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.09191111111111111,
+      "grad_norm": 102.25955963134766,
+      "learning_rate": 3e-06,
+      "loss": -6.4906,
+      "step": 1034
+    },
+    {
+      "epoch": 0.092,
+      "grad_norm": 128.81582641601562,
+      "learning_rate": 3e-06,
+      "loss": -6.558,
+      "step": 1035
+    },
+    {
+      "epoch": 0.09208888888888889,
+      "grad_norm": 112.67058563232422,
+      "learning_rate": 3e-06,
+      "loss": -9.3061,
+      "step": 1036
+    },
+    {
+      "epoch": 0.09217777777777778,
+      "grad_norm": 108.50650024414062,
+      "learning_rate": 3e-06,
+      "loss": -6.0381,
+      "step": 1037
+    },
+    {
+      "epoch": 0.09226666666666666,
+      "grad_norm": 108.7009506225586,
+      "learning_rate": 3e-06,
+      "loss": 0.0853,
+      "step": 1038
+    },
+    {
+      "epoch": 0.09235555555555555,
+      "grad_norm": 114.24646759033203,
+      "learning_rate": 3e-06,
+      "loss": -9.0476,
+      "step": 1039
+    },
+    {
+      "epoch": 0.09244444444444444,
+      "grad_norm": 99.69547271728516,
+      "learning_rate": 3e-06,
+      "loss": -8.1766,
+      "step": 1040
+    },
+    {
+      "epoch": 0.09253333333333333,
+      "grad_norm": 142.9137725830078,
+      "learning_rate": 3e-06,
+      "loss": -7.6658,
+      "step": 1041
+    },
+    {
+      "epoch": 0.09262222222222222,
+      "grad_norm": 113.05297088623047,
+      "learning_rate": 3e-06,
+      "loss": -10.693,
+      "step": 1042
+    },
+    {
+      "epoch": 0.0927111111111111,
+      "grad_norm": 114.48544311523438,
+      "learning_rate": 3e-06,
+      "loss": -8.4665,
+      "step": 1043
+    },
+    {
+      "epoch": 0.0928,
+      "grad_norm": 131.56544494628906,
+      "learning_rate": 3e-06,
+      "loss": -1.0749,
+      "step": 1044
+    },
+    {
+      "completion_length": 251.3125,
+      "epoch": 0.09288888888888888,
+      "grad_norm": 162.5680389404297,
+      "learning_rate": 3e-06,
+      "loss": 2.2852,
+      "reward": 1.4479166865348816,
+      "reward_std": 0.5932036638259888,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.708333358168602,
+      "step": 1045,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.09297777777777778,
+      "grad_norm": 183.73635864257812,
+      "learning_rate": 3e-06,
+      "loss": 4.3862,
+      "step": 1046
+    },
+    {
+      "epoch": 0.09306666666666667,
+      "grad_norm": 149.2581329345703,
+      "learning_rate": 3e-06,
+      "loss": 3.2506,
+      "step": 1047
+    },
+    {
+      "epoch": 0.09315555555555556,
+      "grad_norm": 146.53892517089844,
+      "learning_rate": 3e-06,
+      "loss": 6.7977,
+      "step": 1048
+    },
+    {
+      "epoch": 0.09324444444444445,
+      "grad_norm": 187.605224609375,
+      "learning_rate": 3e-06,
+      "loss": -4.9329,
+      "step": 1049
+    },
+    {
+      "epoch": 0.09333333333333334,
+      "grad_norm": 135.69638061523438,
+      "learning_rate": 3e-06,
+      "loss": 1.364,
+      "step": 1050
+    },
+    {
+      "epoch": 0.09342222222222223,
+      "grad_norm": 155.4630126953125,
+      "learning_rate": 3e-06,
+      "loss": -0.1606,
+      "step": 1051
+    },
+    {
+      "epoch": 0.09351111111111111,
+      "grad_norm": 137.24606323242188,
+      "learning_rate": 3e-06,
+      "loss": 2.4667,
+      "step": 1052
+    },
+    {
+      "epoch": 0.0936,
+      "grad_norm": 145.3031768798828,
+      "learning_rate": 3e-06,
+      "loss": 0.3328,
+      "step": 1053
+    },
+    {
+      "epoch": 0.09368888888888889,
+      "grad_norm": 138.85008239746094,
+      "learning_rate": 3e-06,
+      "loss": 3.6961,
+      "step": 1054
+    },
+    {
+      "epoch": 0.09377777777777778,
+      "grad_norm": 151.01002502441406,
+      "learning_rate": 3e-06,
+      "loss": -8.223,
+      "step": 1055
+    },
+    {
+      "epoch": 0.09386666666666667,
+      "grad_norm": 130.21697998046875,
+      "learning_rate": 3e-06,
+      "loss": -1.2589,
+      "step": 1056
+    },
+    {
+      "completion_length": 255.4791717529297,
+      "epoch": 0.09395555555555556,
+      "grad_norm": 129.92930603027344,
+      "learning_rate": 3e-06,
+      "loss": -4.7404,
+      "reward": 1.0,
+      "reward_std": 0.43528565764427185,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 1057,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.09404444444444444,
+      "grad_norm": 134.7594451904297,
+      "learning_rate": 3e-06,
+      "loss": -7.4851,
+      "step": 1058
+    },
+    {
+      "epoch": 0.09413333333333333,
+      "grad_norm": 145.62059020996094,
+      "learning_rate": 3e-06,
+      "loss": -11.0884,
+      "step": 1059
+    },
+    {
+      "epoch": 0.09422222222222222,
+      "grad_norm": 116.93612670898438,
+      "learning_rate": 3e-06,
+      "loss": 6.683,
+      "step": 1060
+    },
+    {
+      "epoch": 0.09431111111111111,
+      "grad_norm": 126.82006072998047,
+      "learning_rate": 3e-06,
+      "loss": -7.7905,
+      "step": 1061
+    },
+    {
+      "epoch": 0.0944,
+      "grad_norm": 252.49966430664062,
+      "learning_rate": 3e-06,
+      "loss": 4.6391,
+      "step": 1062
+    },
+    {
+      "epoch": 0.09448888888888889,
+      "grad_norm": 129.0404052734375,
+      "learning_rate": 3e-06,
+      "loss": -6.2658,
+      "step": 1063
+    },
+    {
+      "epoch": 0.09457777777777777,
+      "grad_norm": 140.08370971679688,
+      "learning_rate": 3e-06,
+      "loss": -8.4286,
+      "step": 1064
+    },
+    {
+      "epoch": 0.09466666666666666,
+      "grad_norm": 114.8161392211914,
+      "learning_rate": 3e-06,
+      "loss": -12.9169,
+      "step": 1065
+    },
+    {
+      "epoch": 0.09475555555555555,
+      "grad_norm": 112.29281616210938,
+      "learning_rate": 3e-06,
+      "loss": 4.7452,
+      "step": 1066
+    },
+    {
+      "epoch": 0.09484444444444444,
+      "grad_norm": 140.55029296875,
+      "learning_rate": 3e-06,
+      "loss": -9.5303,
+      "step": 1067
+    },
+    {
+      "epoch": 0.09493333333333333,
+      "grad_norm": 147.0571746826172,
+      "learning_rate": 3e-06,
+      "loss": 3.0677,
+      "step": 1068
+    },
+    {
+      "completion_length": 245.0416717529297,
+      "epoch": 0.09502222222222222,
+      "grad_norm": 101.13983154296875,
+      "learning_rate": 3e-06,
+      "loss": 7.3268,
+      "reward": 0.7916666865348816,
+      "reward_std": 0.27350127696990967,
+      "rewards/boxed_and_answer_tags_format_reward": 0.4999999850988388,
+      "rewards/correctness_reward_func_math": 0.2916666679084301,
+      "step": 1069,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.0951111111111111,
+      "grad_norm": 121.06708526611328,
+      "learning_rate": 3e-06,
+      "loss": 4.489,
+      "step": 1070
+    },
+    {
+      "epoch": 0.0952,
+      "grad_norm": 127.9291763305664,
+      "learning_rate": 3e-06,
+      "loss": 1.7732,
+      "step": 1071
+    },
+    {
+      "epoch": 0.0952888888888889,
+      "grad_norm": 120.67790222167969,
+      "learning_rate": 3e-06,
+      "loss": 7.2385,
+      "step": 1072
+    },
+    {
+      "epoch": 0.09537777777777778,
+      "grad_norm": 98.45962524414062,
+      "learning_rate": 3e-06,
+      "loss": -9.8857,
+      "step": 1073
+    },
+    {
+      "epoch": 0.09546666666666667,
+      "grad_norm": 110.4314193725586,
+      "learning_rate": 3e-06,
+      "loss": -1.6899,
+      "step": 1074
+    },
+    {
+      "epoch": 0.09555555555555556,
+      "grad_norm": 101.835693359375,
+      "learning_rate": 3e-06,
+      "loss": 5.3228,
+      "step": 1075
+    },
+    {
+      "epoch": 0.09564444444444445,
+      "grad_norm": 119.22904205322266,
+      "learning_rate": 3e-06,
+      "loss": 3.0508,
+      "step": 1076
+    },
+    {
+      "epoch": 0.09573333333333334,
+      "grad_norm": 126.35284423828125,
+      "learning_rate": 3e-06,
+      "loss": -1.1565,
+      "step": 1077
+    },
+    {
+      "epoch": 0.09582222222222223,
+      "grad_norm": 129.94705200195312,
+      "learning_rate": 3e-06,
+      "loss": 4.3463,
+      "step": 1078
+    },
+    {
+      "epoch": 0.09591111111111111,
+      "grad_norm": 95.31863403320312,
+      "learning_rate": 3e-06,
+      "loss": -12.0709,
+      "step": 1079
+    },
+    {
+      "epoch": 0.096,
+      "grad_norm": 111.38770294189453,
+      "learning_rate": 3e-06,
+      "loss": -4.804,
+      "step": 1080
+    },
+    {
+      "completion_length": 227.33333587646484,
+      "epoch": 0.09608888888888889,
+      "grad_norm": 93.04568481445312,
+      "learning_rate": 3e-06,
+      "loss": -1.8293,
+      "reward": 1.2604166865348816,
+      "reward_std": 0.4915197938680649,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.5833333134651184,
+      "step": 1081,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.09617777777777778,
+      "grad_norm": 186.89431762695312,
+      "learning_rate": 3e-06,
+      "loss": -1.6852,
+      "step": 1082
+    },
+    {
+      "epoch": 0.09626666666666667,
+      "grad_norm": 132.6072998046875,
+      "learning_rate": 3e-06,
+      "loss": -19.4138,
+      "step": 1083
+    },
+    {
+      "epoch": 0.09635555555555556,
+      "grad_norm": 128.269287109375,
+      "learning_rate": 3e-06,
+      "loss": -7.5699,
+      "step": 1084
+    },
+    {
+      "epoch": 0.09644444444444444,
+      "grad_norm": 104.99845123291016,
+      "learning_rate": 3e-06,
+      "loss": -14.2839,
+      "step": 1085
+    },
+    {
+      "epoch": 0.09653333333333333,
+      "grad_norm": 127.01644897460938,
+      "learning_rate": 3e-06,
+      "loss": -6.2089,
+      "step": 1086
+    },
+    {
+      "epoch": 0.09662222222222222,
+      "grad_norm": 99.6088638305664,
+      "learning_rate": 3e-06,
+      "loss": -3.0049,
+      "step": 1087
+    },
+    {
+      "epoch": 0.09671111111111111,
+      "grad_norm": 120.30072784423828,
+      "learning_rate": 3e-06,
+      "loss": -3.4027,
+      "step": 1088
+    },
+    {
+      "epoch": 0.0968,
+      "grad_norm": 121.11226654052734,
+      "learning_rate": 3e-06,
+      "loss": -20.6903,
+      "step": 1089
+    },
+    {
+      "epoch": 0.09688888888888889,
+      "grad_norm": 101.49308013916016,
+      "learning_rate": 3e-06,
+      "loss": -8.6379,
+      "step": 1090
+    },
+    {
+      "epoch": 0.09697777777777777,
+      "grad_norm": 117.81327819824219,
+      "learning_rate": 3e-06,
+      "loss": -15.1799,
+      "step": 1091
+    },
+    {
+      "epoch": 0.09706666666666666,
+      "grad_norm": 114.92808532714844,
+      "learning_rate": 3e-06,
+      "loss": -8.7164,
+      "step": 1092
+    },
+    {
+      "completion_length": 245.68750762939453,
+      "epoch": 0.09715555555555555,
+      "grad_norm": 137.37049865722656,
+      "learning_rate": 3e-06,
+      "loss": -17.5632,
+      "reward": 1.2604166865348816,
+      "reward_std": 0.3936076760292053,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.5833333134651184,
+      "step": 1093,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.09724444444444444,
+      "grad_norm": 104.14005279541016,
+      "learning_rate": 3e-06,
+      "loss": -13.8548,
+      "step": 1094
+    },
+    {
+      "epoch": 0.09733333333333333,
+      "grad_norm": 111.48066711425781,
+      "learning_rate": 3e-06,
+      "loss": -14.4865,
+      "step": 1095
+    },
+    {
+      "epoch": 0.09742222222222222,
+      "grad_norm": 115.86460876464844,
+      "learning_rate": 3e-06,
+      "loss": -12.7074,
+      "step": 1096
+    },
+    {
+      "epoch": 0.09751111111111112,
+      "grad_norm": 108.50313568115234,
+      "learning_rate": 3e-06,
+      "loss": -17.9489,
+      "step": 1097
+    },
+    {
+      "epoch": 0.0976,
+      "grad_norm": 113.4880599975586,
+      "learning_rate": 3e-06,
+      "loss": -16.1087,
+      "step": 1098
+    },
+    {
+      "epoch": 0.0976888888888889,
+      "grad_norm": 109.29180145263672,
+      "learning_rate": 3e-06,
+      "loss": -19.2428,
+      "step": 1099
+    },
+    {
+      "epoch": 0.09777777777777778,
+      "grad_norm": 107.1020278930664,
+      "learning_rate": 3e-06,
+      "loss": -15.2564,
+      "step": 1100
+    },
+    {
+      "epoch": 0.09786666666666667,
+      "grad_norm": 131.5577392578125,
+      "learning_rate": 3e-06,
+      "loss": -15.6807,
+      "step": 1101
+    },
+    {
+      "epoch": 0.09795555555555556,
+      "grad_norm": 121.74998474121094,
+      "learning_rate": 3e-06,
+      "loss": -15.0192,
+      "step": 1102
+    },
+    {
+      "epoch": 0.09804444444444445,
+      "grad_norm": 107.87700653076172,
+      "learning_rate": 3e-06,
+      "loss": -19.6246,
+      "step": 1103
+    },
+    {
+      "epoch": 0.09813333333333334,
+      "grad_norm": 132.6728515625,
+      "learning_rate": 3e-06,
+      "loss": -17.4119,
+      "step": 1104
+    },
+    {
+      "completion_length": 247.375,
+      "epoch": 0.09822222222222222,
+      "grad_norm": 194.66749572753906,
+      "learning_rate": 3e-06,
+      "loss": -59.2499,
+      "reward": 1.7083333730697632,
+      "reward_std": 0.6184598803520203,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.9583333432674408,
+      "step": 1105,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.09831111111111111,
+      "grad_norm": 179.941162109375,
+      "learning_rate": 3e-06,
+      "loss": -46.7423,
+      "step": 1106
+    },
+    {
+      "epoch": 0.0984,
+      "grad_norm": 232.29762268066406,
+      "learning_rate": 3e-06,
+      "loss": -54.0588,
+      "step": 1107
+    },
+    {
+      "epoch": 0.09848888888888889,
+      "grad_norm": 208.61793518066406,
+      "learning_rate": 3e-06,
+      "loss": -72.7937,
+      "step": 1108
+    },
+    {
+      "epoch": 0.09857777777777778,
+      "grad_norm": 180.18431091308594,
+      "learning_rate": 3e-06,
+      "loss": -37.6563,
+      "step": 1109
+    },
+    {
+      "epoch": 0.09866666666666667,
+      "grad_norm": 191.83653259277344,
+      "learning_rate": 3e-06,
+      "loss": -56.5117,
+      "step": 1110
+    },
+    {
+      "epoch": 0.09875555555555555,
+      "grad_norm": 192.1591796875,
+      "learning_rate": 3e-06,
+      "loss": -63.0639,
+      "step": 1111
+    },
+    {
+      "epoch": 0.09884444444444444,
+      "grad_norm": 183.53610229492188,
+      "learning_rate": 3e-06,
+      "loss": -50.2258,
+      "step": 1112
+    },
+    {
+      "epoch": 0.09893333333333333,
+      "grad_norm": 233.22872924804688,
+      "learning_rate": 3e-06,
+      "loss": -58.1688,
+      "step": 1113
+    },
+    {
+      "epoch": 0.09902222222222222,
+      "grad_norm": 219.78233337402344,
+      "learning_rate": 3e-06,
+      "loss": -78.4412,
+      "step": 1114
+    },
+    {
+      "epoch": 0.09911111111111111,
+      "grad_norm": 189.9258270263672,
+      "learning_rate": 3e-06,
+      "loss": -42.5197,
+      "step": 1115
+    },
+    {
+      "epoch": 0.0992,
+      "grad_norm": 204.02183532714844,
+      "learning_rate": 3e-06,
+      "loss": -61.9828,
+      "step": 1116
+    },
+    {
+      "completion_length": 248.9166717529297,
+      "epoch": 0.09928888888888888,
+      "grad_norm": 114.85681915283203,
+      "learning_rate": 3e-06,
+      "loss": 1.5984,
+      "reward": 1.2395833730697632,
+      "reward_std": 0.2296396717429161,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.4999999850988388,
+      "step": 1117,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.09937777777777777,
+      "grad_norm": 113.66478729248047,
+      "learning_rate": 3e-06,
+      "loss": -2.2296,
+      "step": 1118
+    },
+    {
+      "epoch": 0.09946666666666666,
+      "grad_norm": 124.4161148071289,
+      "learning_rate": 3e-06,
+      "loss": 4.382,
+      "step": 1119
+    },
+    {
+      "epoch": 0.09955555555555555,
+      "grad_norm": 131.39085388183594,
+      "learning_rate": 3e-06,
+      "loss": 0.5734,
+      "step": 1120
+    },
+    {
+      "epoch": 0.09964444444444444,
+      "grad_norm": 119.46894073486328,
+      "learning_rate": 3e-06,
+      "loss": 5.0378,
+      "step": 1121
+    },
+    {
+      "epoch": 0.09973333333333333,
+      "grad_norm": 113.26507568359375,
+      "learning_rate": 3e-06,
+      "loss": 1.4215,
+      "step": 1122
+    },
+    {
+      "epoch": 0.09982222222222223,
+      "grad_norm": 120.69562530517578,
+      "learning_rate": 3e-06,
+      "loss": 0.0203,
+      "step": 1123
+    },
+    {
+      "epoch": 0.09991111111111112,
+      "grad_norm": 128.0107421875,
+      "learning_rate": 3e-06,
+      "loss": -3.7284,
+      "step": 1124
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 139.31997680664062,
+      "learning_rate": 3e-06,
+      "loss": 2.9527,
+      "step": 1125
+    },
+    {
+      "epoch": 0.1000888888888889,
+      "grad_norm": 111.49156188964844,
+      "learning_rate": 3e-06,
+      "loss": -1.217,
+      "step": 1126
+    },
+    {
+      "epoch": 0.10017777777777778,
+      "grad_norm": 107.92985534667969,
+      "learning_rate": 3e-06,
+      "loss": 3.7084,
+      "step": 1127
+    },
+    {
+      "epoch": 0.10026666666666667,
+      "grad_norm": 112.88748168945312,
+      "learning_rate": 3e-06,
+      "loss": -0.5577,
+      "step": 1128
+    },
+    {
+      "completion_length": 245.18750762939453,
+      "epoch": 0.10035555555555556,
+      "grad_norm": 124.6043930053711,
+      "learning_rate": 3e-06,
+      "loss": -20.6745,
+      "reward": 1.4583333432674408,
+      "reward_std": 0.5722163170576096,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.8333333283662796,
+      "step": 1129,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.10044444444444445,
+      "grad_norm": 118.25672149658203,
+      "learning_rate": 3e-06,
+      "loss": -32.0798,
+      "step": 1130
+    },
+    {
+      "epoch": 0.10053333333333334,
+      "grad_norm": 189.2974853515625,
+      "learning_rate": 3e-06,
+      "loss": -23.7609,
+      "step": 1131
+    },
+    {
+      "epoch": 0.10062222222222222,
+      "grad_norm": 139.94485473632812,
+      "learning_rate": 3e-06,
+      "loss": -30.0811,
+      "step": 1132
+    },
+    {
+      "epoch": 0.10071111111111111,
+      "grad_norm": 132.92324829101562,
+      "learning_rate": 3e-06,
+      "loss": -29.1304,
+      "step": 1133
+    },
+    {
+      "epoch": 0.1008,
+      "grad_norm": 129.6322021484375,
+      "learning_rate": 3e-06,
+      "loss": -25.8999,
+      "step": 1134
+    },
+    {
+      "epoch": 0.10088888888888889,
+      "grad_norm": 122.53899383544922,
+      "learning_rate": 3e-06,
+      "loss": -22.266,
+      "step": 1135
+    },
+    {
+      "epoch": 0.10097777777777778,
+      "grad_norm": 111.6375961303711,
+      "learning_rate": 3e-06,
+      "loss": -32.9394,
+      "step": 1136
+    },
+    {
+      "epoch": 0.10106666666666667,
+      "grad_norm": 162.63771057128906,
+      "learning_rate": 3e-06,
+      "loss": -25.9897,
+      "step": 1137
+    },
+    {
+      "epoch": 0.10115555555555555,
+      "grad_norm": 154.28424072265625,
+      "learning_rate": 3e-06,
+      "loss": -32.7633,
+      "step": 1138
+    },
+    {
+      "epoch": 0.10124444444444444,
+      "grad_norm": 132.94351196289062,
+      "learning_rate": 3e-06,
+      "loss": -31.2033,
+      "step": 1139
+    },
+    {
+      "epoch": 0.10133333333333333,
+      "grad_norm": 131.9947052001953,
+      "learning_rate": 3e-06,
+      "loss": -27.4069,
+      "step": 1140
+    },
+    {
+      "completion_length": 248.625,
+      "epoch": 0.10142222222222222,
+      "grad_norm": 113.37858581542969,
+      "learning_rate": 3e-06,
+      "loss": -14.7283,
+      "reward": 0.9479166865348816,
+      "reward_std": 0.2296396642923355,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6145833432674408,
+      "rewards/correctness_reward_func_math": 0.3333333246409893,
+      "step": 1141,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.10151111111111111,
+      "grad_norm": 112.18434143066406,
+      "learning_rate": 3e-06,
+      "loss": -11.645,
+      "step": 1142
+    },
+    {
+      "epoch": 0.1016,
+      "grad_norm": 105.41796112060547,
+      "learning_rate": 3e-06,
+      "loss": -15.9935,
+      "step": 1143
+    },
+    {
+      "epoch": 0.10168888888888888,
+      "grad_norm": 91.3786849975586,
+      "learning_rate": 3e-06,
+      "loss": -16.6043,
+      "step": 1144
+    },
+    {
+      "epoch": 0.10177777777777777,
+      "grad_norm": 104.3455810546875,
+      "learning_rate": 3e-06,
+      "loss": -3.4885,
+      "step": 1145
+    },
+    {
+      "epoch": 0.10186666666666666,
+      "grad_norm": 87.41930389404297,
+      "learning_rate": 3e-06,
+      "loss": -20.634,
+      "step": 1146
+    },
+    {
+      "epoch": 0.10195555555555555,
+      "grad_norm": 90.50940704345703,
+      "learning_rate": 3e-06,
+      "loss": -16.7866,
+      "step": 1147
+    },
+    {
+      "epoch": 0.10204444444444444,
+      "grad_norm": 117.73979949951172,
+      "learning_rate": 3e-06,
+      "loss": -13.9565,
+      "step": 1148
+    },
+    {
+      "epoch": 0.10213333333333334,
+      "grad_norm": 117.4783706665039,
+      "learning_rate": 3e-06,
+      "loss": -18.3371,
+      "step": 1149
+    },
+    {
+      "epoch": 0.10222222222222223,
+      "grad_norm": 90.1675033569336,
+      "learning_rate": 3e-06,
+      "loss": -19.5285,
+      "step": 1150
+    },
+    {
+      "epoch": 0.10231111111111112,
+      "grad_norm": 114.65315246582031,
+      "learning_rate": 3e-06,
+      "loss": -6.9923,
+      "step": 1151
+    },
+    {
+      "epoch": 0.1024,
+      "grad_norm": 236.66262817382812,
+      "learning_rate": 3e-06,
+      "loss": -23.6847,
+      "step": 1152
+    },
+    {
+      "completion_length": 254.0625,
+      "epoch": 0.1024888888888889,
+      "grad_norm": 112.95475769042969,
+      "learning_rate": 3e-06,
+      "loss": -27.7434,
+      "reward": 0.96875,
+      "reward_std": 0.3734789788722992,
+      "rewards/boxed_and_answer_tags_format_reward": 0.71875,
+      "rewards/correctness_reward_func_math": 0.25,
+      "step": 1153,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.10257777777777778,
+      "grad_norm": 144.55364990234375,
+      "learning_rate": 3e-06,
+      "loss": -22.4297,
+      "step": 1154
+    },
+    {
+      "epoch": 0.10266666666666667,
+      "grad_norm": 128.1786346435547,
+      "learning_rate": 3e-06,
+      "loss": -34.1587,
+      "step": 1155
+    },
+    {
+      "epoch": 0.10275555555555556,
+      "grad_norm": 125.85458374023438,
+      "learning_rate": 3e-06,
+      "loss": -30.6015,
+      "step": 1156
+    },
+    {
+      "epoch": 0.10284444444444445,
+      "grad_norm": 142.524658203125,
+      "learning_rate": 3e-06,
+      "loss": -31.4122,
+      "step": 1157
+    },
+    {
+      "epoch": 0.10293333333333334,
+      "grad_norm": 111.42877197265625,
+      "learning_rate": 3e-06,
+      "loss": -26.1967,
+      "step": 1158
+    },
+    {
+      "epoch": 0.10302222222222222,
+      "grad_norm": 123.39324951171875,
+      "learning_rate": 3e-06,
+      "loss": -29.3019,
+      "step": 1159
+    },
+    {
+      "epoch": 0.10311111111111111,
+      "grad_norm": 151.65802001953125,
+      "learning_rate": 3e-06,
+      "loss": -24.3001,
+      "step": 1160
+    },
+    {
+      "epoch": 0.1032,
+      "grad_norm": 127.43438720703125,
+      "learning_rate": 3e-06,
+      "loss": -36.1734,
+      "step": 1161
+    },
+    {
+      "epoch": 0.10328888888888889,
+      "grad_norm": 123.67347717285156,
+      "learning_rate": 3e-06,
+      "loss": -33.6541,
+      "step": 1162
+    },
+    {
+      "epoch": 0.10337777777777778,
+      "grad_norm": 147.1012420654297,
+      "learning_rate": 3e-06,
+      "loss": -33.954,
+      "step": 1163
+    },
+    {
+      "epoch": 0.10346666666666667,
+      "grad_norm": 125.47201538085938,
+      "learning_rate": 3e-06,
+      "loss": -28.7879,
+      "step": 1164
+    },
+    {
+      "completion_length": 236.64583587646484,
+      "epoch": 0.10355555555555555,
+      "grad_norm": 348.24346923828125,
+      "learning_rate": 3e-06,
+      "loss": -9.8131,
+      "reward": 1.4375000596046448,
+      "reward_std": 0.4971916079521179,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 1165,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.10364444444444444,
+      "grad_norm": 189.84584045410156,
+      "learning_rate": 3e-06,
+      "loss": -5.3377,
+      "step": 1166
+    },
+    {
+      "epoch": 0.10373333333333333,
+      "grad_norm": 192.2544403076172,
+      "learning_rate": 3e-06,
+      "loss": -1.7722,
+      "step": 1167
+    },
+    {
+      "epoch": 0.10382222222222222,
+      "grad_norm": 180.6472625732422,
+      "learning_rate": 3e-06,
+      "loss": -6.7487,
+      "step": 1168
+    },
+    {
+      "epoch": 0.1039111111111111,
+      "grad_norm": 197.0133819580078,
+      "learning_rate": 3e-06,
+      "loss": -2.4648,
+      "step": 1169
+    },
+    {
+      "epoch": 0.104,
+      "grad_norm": 256.5722351074219,
+      "learning_rate": 3e-06,
+      "loss": -17.2997,
+      "step": 1170
+    },
+    {
+      "epoch": 0.10408888888888888,
+      "grad_norm": 179.86228942871094,
+      "learning_rate": 3e-06,
+      "loss": -12.5209,
+      "step": 1171
+    },
+    {
+      "epoch": 0.10417777777777777,
+      "grad_norm": 195.3128204345703,
+      "learning_rate": 3e-06,
+      "loss": -9.6999,
+      "step": 1172
+    },
+    {
+      "epoch": 0.10426666666666666,
+      "grad_norm": 204.9373321533203,
+      "learning_rate": 3e-06,
+      "loss": -6.9052,
+      "step": 1173
+    },
+    {
+      "epoch": 0.10435555555555555,
+      "grad_norm": 192.8905792236328,
+      "learning_rate": 3e-06,
+      "loss": -10.4987,
+      "step": 1174
+    },
+    {
+      "epoch": 0.10444444444444445,
+      "grad_norm": 181.99449157714844,
+      "learning_rate": 3e-06,
+      "loss": -7.0688,
+      "step": 1175
+    },
+    {
+      "epoch": 0.10453333333333334,
+      "grad_norm": 189.1867218017578,
+      "learning_rate": 3e-06,
+      "loss": -22.4777,
+      "step": 1176
+    },
+    {
+      "completion_length": 250.33333587646484,
+      "epoch": 0.10462222222222223,
+      "grad_norm": 158.67459106445312,
+      "learning_rate": 3e-06,
+      "loss": -8.4502,
+      "reward": 0.9583333432674408,
+      "reward_std": 0.5451789498329163,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5,
+      "rewards/correctness_reward_func_math": 0.4583333358168602,
+      "step": 1177,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.10471111111111112,
+      "grad_norm": 158.85533142089844,
+      "learning_rate": 3e-06,
+      "loss": -13.1446,
+      "step": 1178
+    },
+    {
+      "epoch": 0.1048,
+      "grad_norm": 155.9822235107422,
+      "learning_rate": 3e-06,
+      "loss": -6.7899,
+      "step": 1179
+    },
+    {
+      "epoch": 0.10488888888888889,
+      "grad_norm": 150.00985717773438,
+      "learning_rate": 3e-06,
+      "loss": 5.268,
+      "step": 1180
+    },
+    {
+      "epoch": 0.10497777777777778,
+      "grad_norm": 140.22618103027344,
+      "learning_rate": 3e-06,
+      "loss": -3.0269,
+      "step": 1181
+    },
+    {
+      "epoch": 0.10506666666666667,
+      "grad_norm": 130.9547119140625,
+      "learning_rate": 3e-06,
+      "loss": -6.5307,
+      "step": 1182
+    },
+    {
+      "epoch": 0.10515555555555556,
+      "grad_norm": 157.96466064453125,
+      "learning_rate": 3e-06,
+      "loss": -10.1506,
+      "step": 1183
+    },
+    {
+      "epoch": 0.10524444444444445,
+      "grad_norm": 162.57582092285156,
+      "learning_rate": 3e-06,
+      "loss": -14.3085,
+      "step": 1184
+    },
+    {
+      "epoch": 0.10533333333333333,
+      "grad_norm": 163.5466766357422,
+      "learning_rate": 3e-06,
+      "loss": -9.5339,
+      "step": 1185
+    },
+    {
+      "epoch": 0.10542222222222222,
+      "grad_norm": 126.0348129272461,
+      "learning_rate": 3e-06,
+      "loss": 2.7285,
+      "step": 1186
+    },
+    {
+      "epoch": 0.10551111111111111,
+      "grad_norm": 152.0486297607422,
+      "learning_rate": 3e-06,
+      "loss": -5.0055,
+      "step": 1187
+    },
+    {
+      "epoch": 0.1056,
+      "grad_norm": 135.672607421875,
+      "learning_rate": 3e-06,
+      "loss": -7.9305,
+      "step": 1188
+    },
+    {
+      "completion_length": 242.43750762939453,
+      "epoch": 0.10568888888888889,
+      "grad_norm": 102.83517456054688,
+      "learning_rate": 3e-06,
+      "loss": -9.6079,
+      "reward": 1.5833333730697632,
+      "reward_std": 0.3332235962152481,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.8333333134651184,
+      "step": 1189,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.10577777777777778,
+      "grad_norm": 104.57038116455078,
+      "learning_rate": 3e-06,
+      "loss": 3.4859,
+      "step": 1190
+    },
+    {
+      "epoch": 0.10586666666666666,
+      "grad_norm": 130.5141143798828,
+      "learning_rate": 3e-06,
+      "loss": 9.6096,
+      "step": 1191
+    },
+    {
+      "epoch": 0.10595555555555555,
+      "grad_norm": 121.0637435913086,
+      "learning_rate": 3e-06,
+      "loss": -7.0112,
+      "step": 1192
+    },
+    {
+      "epoch": 0.10604444444444444,
+      "grad_norm": 116.66060638427734,
+      "learning_rate": 3e-06,
+      "loss": -0.6318,
+      "step": 1193
+    },
+    {
+      "epoch": 0.10613333333333333,
+      "grad_norm": 93.47602844238281,
+      "learning_rate": 3e-06,
+      "loss": -7.121,
+      "step": 1194
+    },
+    {
+      "epoch": 0.10622222222222222,
+      "grad_norm": 97.91071319580078,
+      "learning_rate": 3e-06,
+      "loss": -11.2048,
+      "step": 1195
+    },
+    {
+      "epoch": 0.1063111111111111,
+      "grad_norm": 99.79684448242188,
+      "learning_rate": 3e-06,
+      "loss": 2.4369,
+      "step": 1196
+    },
+    {
+      "epoch": 0.1064,
+      "grad_norm": 103.4743423461914,
+      "learning_rate": 3e-06,
+      "loss": 7.3113,
+      "step": 1197
+    },
+    {
+      "epoch": 0.10648888888888888,
+      "grad_norm": 115.71762084960938,
+      "learning_rate": 3e-06,
+      "loss": -8.8378,
+      "step": 1198
+    },
+    {
+      "epoch": 0.10657777777777777,
+      "grad_norm": 116.43769073486328,
+      "learning_rate": 3e-06,
+      "loss": -2.6688,
+      "step": 1199
+    },
+    {
+      "epoch": 0.10666666666666667,
+      "grad_norm": 93.20524597167969,
+      "learning_rate": 3e-06,
+      "loss": -9.0718,
+      "step": 1200
+    },
+    {
+      "completion_length": 242.77083587646484,
+      "epoch": 0.10675555555555556,
+      "grad_norm": 156.0432586669922,
+      "learning_rate": 3e-06,
+      "loss": 6.0861,
+      "reward": 1.6770833730697632,
+      "reward_std": 0.5419133305549622,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 1.0,
+      "step": 1201,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.10684444444444445,
+      "grad_norm": 173.78958129882812,
+      "learning_rate": 3e-06,
+      "loss": 9.0247,
+      "step": 1202
+    },
+    {
+      "epoch": 0.10693333333333334,
+      "grad_norm": 152.82534790039062,
+      "learning_rate": 3e-06,
+      "loss": 5.6834,
+      "step": 1203
+    },
+    {
+      "epoch": 0.10702222222222223,
+      "grad_norm": 151.876953125,
+      "learning_rate": 3e-06,
+      "loss": 6.9018,
+      "step": 1204
+    },
+    {
+      "epoch": 0.10711111111111112,
+      "grad_norm": 165.88400268554688,
+      "learning_rate": 3e-06,
+      "loss": 9.1234,
+      "step": 1205
+    },
+    {
+      "epoch": 0.1072,
+      "grad_norm": 167.48348999023438,
+      "learning_rate": 3e-06,
+      "loss": 9.4483,
+      "step": 1206
+    },
+    {
+      "epoch": 0.10728888888888889,
+      "grad_norm": 149.90536499023438,
+      "learning_rate": 3e-06,
+      "loss": 2.7725,
+      "step": 1207
+    },
+    {
+      "epoch": 0.10737777777777778,
+      "grad_norm": 169.23411560058594,
+      "learning_rate": 3e-06,
+      "loss": 6.9296,
+      "step": 1208
+    },
+    {
+      "epoch": 0.10746666666666667,
+      "grad_norm": 233.46914672851562,
+      "learning_rate": 3e-06,
+      "loss": 3.5024,
+      "step": 1209
+    },
+    {
+      "epoch": 0.10755555555555556,
+      "grad_norm": 147.3250732421875,
+      "learning_rate": 3e-06,
+      "loss": 3.7723,
+      "step": 1210
+    },
+    {
+      "epoch": 0.10764444444444445,
+      "grad_norm": 155.88914489746094,
+      "learning_rate": 3e-06,
+      "loss": 7.0329,
+      "step": 1211
+    },
+    {
+      "epoch": 0.10773333333333333,
+      "grad_norm": 153.82785034179688,
+      "learning_rate": 3e-06,
+      "loss": 7.0486,
+      "step": 1212
+    },
+    {
+      "completion_length": 247.6666717529297,
+      "epoch": 0.10782222222222222,
+      "grad_norm": 158.41856384277344,
+      "learning_rate": 3e-06,
+      "loss": 18.2887,
+      "reward": 1.614583432674408,
+      "reward_std": 0.3665703386068344,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.8750000149011612,
+      "step": 1213,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.10791111111111111,
+      "grad_norm": 163.51866149902344,
+      "learning_rate": 3e-06,
+      "loss": 5.6461,
+      "step": 1214
+    },
+    {
+      "epoch": 0.108,
+      "grad_norm": 131.4403839111328,
+      "learning_rate": 3e-06,
+      "loss": 3.3667,
+      "step": 1215
+    },
+    {
+      "epoch": 0.10808888888888889,
+      "grad_norm": 136.77757263183594,
+      "learning_rate": 3e-06,
+      "loss": 4.406,
+      "step": 1216
+    },
+    {
+      "epoch": 0.10817777777777778,
+      "grad_norm": 113.4407958984375,
+      "learning_rate": 3e-06,
+      "loss": 12.3995,
+      "step": 1217
+    },
+    {
+      "epoch": 0.10826666666666666,
+      "grad_norm": 136.1168212890625,
+      "learning_rate": 3e-06,
+      "loss": 4.6776,
+      "step": 1218
+    },
+    {
+      "epoch": 0.10835555555555555,
+      "grad_norm": 144.9372100830078,
+      "learning_rate": 3e-06,
+      "loss": 17.5724,
+      "step": 1219
+    },
+    {
+      "epoch": 0.10844444444444444,
+      "grad_norm": 125.29820251464844,
+      "learning_rate": 3e-06,
+      "loss": 4.1424,
+      "step": 1220
+    },
+    {
+      "epoch": 0.10853333333333333,
+      "grad_norm": 138.5077667236328,
+      "learning_rate": 3e-06,
+      "loss": 1.3947,
+      "step": 1221
+    },
+    {
+      "epoch": 0.10862222222222222,
+      "grad_norm": 128.62693786621094,
+      "learning_rate": 3e-06,
+      "loss": 2.4864,
+      "step": 1222
+    },
+    {
+      "epoch": 0.1087111111111111,
+      "grad_norm": 107.87577056884766,
+      "learning_rate": 3e-06,
+      "loss": 10.4239,
+      "step": 1223
+    },
+    {
+      "epoch": 0.1088,
+      "grad_norm": 127.07205963134766,
+      "learning_rate": 3e-06,
+      "loss": 1.3109,
+      "step": 1224
+    },
+    {
+      "completion_length": 238.89584350585938,
+      "epoch": 0.10888888888888888,
+      "grad_norm": 154.35325622558594,
+      "learning_rate": 3e-06,
+      "loss": 8.6596,
+      "reward": 1.5208333730697632,
+      "reward_std": 0.3680921420454979,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.8333333432674408,
+      "step": 1225,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.10897777777777778,
+      "grad_norm": 165.58445739746094,
+      "learning_rate": 3e-06,
+      "loss": 1.8278,
+      "step": 1226
+    },
+    {
+      "epoch": 0.10906666666666667,
+      "grad_norm": 360.30029296875,
+      "learning_rate": 3e-06,
+      "loss": -9.1779,
+      "step": 1227
+    },
+    {
+      "epoch": 0.10915555555555556,
+      "grad_norm": 152.518310546875,
+      "learning_rate": 3e-06,
+      "loss": -9.97,
+      "step": 1228
+    },
+    {
+      "epoch": 0.10924444444444445,
+      "grad_norm": 146.67100524902344,
+      "learning_rate": 3e-06,
+      "loss": -6.3997,
+      "step": 1229
+    },
+    {
+      "epoch": 0.10933333333333334,
+      "grad_norm": 138.9771728515625,
+      "learning_rate": 3e-06,
+      "loss": -6.0985,
+      "step": 1230
+    },
+    {
+      "epoch": 0.10942222222222223,
+      "grad_norm": 159.66302490234375,
+      "learning_rate": 3e-06,
+      "loss": 5.5207,
+      "step": 1231
+    },
+    {
+      "epoch": 0.10951111111111111,
+      "grad_norm": 138.22695922851562,
+      "learning_rate": 3e-06,
+      "loss": -0.4886,
+      "step": 1232
+    },
+    {
+      "epoch": 0.1096,
+      "grad_norm": 160.3970489501953,
+      "learning_rate": 3e-06,
+      "loss": -12.1071,
+      "step": 1233
+    },
+    {
+      "epoch": 0.10968888888888889,
+      "grad_norm": 141.63226318359375,
+      "learning_rate": 3e-06,
+      "loss": -13.6489,
+      "step": 1234
+    },
+    {
+      "epoch": 0.10977777777777778,
+      "grad_norm": 146.45748901367188,
+      "learning_rate": 3e-06,
+      "loss": -10.0014,
+      "step": 1235
+    },
+    {
+      "epoch": 0.10986666666666667,
+      "grad_norm": 140.94992065429688,
+      "learning_rate": 3e-06,
+      "loss": -9.6508,
+      "step": 1236
+    },
+    {
+      "completion_length": 255.4166717529297,
+      "epoch": 0.10995555555555556,
+      "grad_norm": 160.60601806640625,
+      "learning_rate": 3e-06,
+      "loss": 1.8826,
+      "reward": 0.8020833432674408,
+      "reward_std": 0.3381742835044861,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6354166865348816,
+      "rewards/correctness_reward_func_math": 0.1666666679084301,
+      "step": 1237,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.11004444444444444,
+      "grad_norm": 186.42630004882812,
+      "learning_rate": 3e-06,
+      "loss": -4.6822,
+      "step": 1238
+    },
+    {
+      "epoch": 0.11013333333333333,
+      "grad_norm": 141.29820251464844,
+      "learning_rate": 3e-06,
+      "loss": 11.1707,
+      "step": 1239
+    },
+    {
+      "epoch": 0.11022222222222222,
+      "grad_norm": 144.1931610107422,
+      "learning_rate": 3e-06,
+      "loss": 3.3338,
+      "step": 1240
+    },
+    {
+      "epoch": 0.11031111111111111,
+      "grad_norm": 135.27630615234375,
+      "learning_rate": 3e-06,
+      "loss": 4.9367,
+      "step": 1241
+    },
+    {
+      "epoch": 0.1104,
+      "grad_norm": 175.80433654785156,
+      "learning_rate": 3e-06,
+      "loss": -13.1628,
+      "step": 1242
+    },
+    {
+      "epoch": 0.11048888888888889,
+      "grad_norm": 165.0531463623047,
+      "learning_rate": 3e-06,
+      "loss": -0.3721,
+      "step": 1243
+    },
+    {
+      "epoch": 0.11057777777777777,
+      "grad_norm": 264.06695556640625,
+      "learning_rate": 3e-06,
+      "loss": -7.2717,
+      "step": 1244
+    },
+    {
+      "epoch": 0.11066666666666666,
+      "grad_norm": 145.43423461914062,
+      "learning_rate": 3e-06,
+      "loss": 8.5606,
+      "step": 1245
+    },
+    {
+      "epoch": 0.11075555555555555,
+      "grad_norm": 145.75946044921875,
+      "learning_rate": 3e-06,
+      "loss": 0.9458,
+      "step": 1246
+    },
+    {
+      "epoch": 0.11084444444444444,
+      "grad_norm": 125.21862030029297,
+      "learning_rate": 3e-06,
+      "loss": 3.1516,
+      "step": 1247
+    },
+    {
+      "epoch": 0.11093333333333333,
+      "grad_norm": 177.2153778076172,
+      "learning_rate": 3e-06,
+      "loss": -17.2454,
+      "step": 1248
+    },
+    {
+      "completion_length": 251.6041717529297,
+      "epoch": 0.11102222222222222,
+      "grad_norm": 371.7901306152344,
+      "learning_rate": 3e-06,
+      "loss": -30.6046,
+      "reward": 1.9895833730697632,
+      "reward_std": 0.8478911817073822,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 1.25,
+      "step": 1249,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 241.84666442871094,
+      "learning_rate": 3e-06,
+      "loss": -47.9038,
+      "step": 1250
+    },
+    {
+      "epoch": 0.1112,
+      "grad_norm": 187.25973510742188,
+      "learning_rate": 3e-06,
+      "loss": -31.5847,
+      "step": 1251
+    },
+    {
+      "epoch": 0.1112888888888889,
+      "grad_norm": 220.72537231445312,
+      "learning_rate": 3e-06,
+      "loss": -37.9464,
+      "step": 1252
+    },
+    {
+      "epoch": 0.11137777777777778,
+      "grad_norm": 215.938720703125,
+      "learning_rate": 3e-06,
+      "loss": -36.917,
+      "step": 1253
+    },
+    {
+      "epoch": 0.11146666666666667,
+      "grad_norm": 255.27755737304688,
+      "learning_rate": 3e-06,
+      "loss": -36.4102,
+      "step": 1254
+    },
+    {
+      "epoch": 0.11155555555555556,
+      "grad_norm": 252.8043212890625,
+      "learning_rate": 3e-06,
+      "loss": -32.6218,
+      "step": 1255
+    },
+    {
+      "epoch": 0.11164444444444445,
+      "grad_norm": 225.40321350097656,
+      "learning_rate": 3e-06,
+      "loss": -51.6723,
+      "step": 1256
+    },
+    {
+      "epoch": 0.11173333333333334,
+      "grad_norm": 208.3738250732422,
+      "learning_rate": 3e-06,
+      "loss": -35.6313,
+      "step": 1257
+    },
+    {
+      "epoch": 0.11182222222222223,
+      "grad_norm": 215.90704345703125,
+      "learning_rate": 3e-06,
+      "loss": -41.5126,
+      "step": 1258
+    },
+    {
+      "epoch": 0.11191111111111111,
+      "grad_norm": 242.8232879638672,
+      "learning_rate": 3e-06,
+      "loss": -42.0952,
+      "step": 1259
+    },
+    {
+      "epoch": 0.112,
+      "grad_norm": 220.46678161621094,
+      "learning_rate": 3e-06,
+      "loss": -40.7929,
+      "step": 1260
+    },
+    {
+      "completion_length": 252.2291717529297,
+      "epoch": 0.11208888888888889,
+      "grad_norm": 112.288330078125,
+      "learning_rate": 3e-06,
+      "loss": -28.9692,
+      "reward": 0.8333333730697632,
+      "reward_std": 0.23116151988506317,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.2083333283662796,
+      "step": 1261,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.11217777777777778,
+      "grad_norm": 111.31488800048828,
+      "learning_rate": 3e-06,
+      "loss": -22.9427,
+      "step": 1262
+    },
+    {
+      "epoch": 0.11226666666666667,
+      "grad_norm": 130.2353057861328,
+      "learning_rate": 3e-06,
+      "loss": -30.2762,
+      "step": 1263
+    },
+    {
+      "epoch": 0.11235555555555556,
+      "grad_norm": 154.38973999023438,
+      "learning_rate": 3e-06,
+      "loss": -39.3815,
+      "step": 1264
+    },
+    {
+      "epoch": 0.11244444444444444,
+      "grad_norm": 109.89620971679688,
+      "learning_rate": 3e-06,
+      "loss": -28.5723,
+      "step": 1265
+    },
+    {
+      "epoch": 0.11253333333333333,
+      "grad_norm": 104.00005340576172,
+      "learning_rate": 3e-06,
+      "loss": -25.6052,
+      "step": 1266
+    },
+    {
+      "epoch": 0.11262222222222222,
+      "grad_norm": 121.78623962402344,
+      "learning_rate": 3e-06,
+      "loss": -30.0983,
+      "step": 1267
+    },
+    {
+      "epoch": 0.11271111111111111,
+      "grad_norm": 119.55603790283203,
+      "learning_rate": 3e-06,
+      "loss": -23.8716,
+      "step": 1268
+    },
+    {
+      "epoch": 0.1128,
+      "grad_norm": 124.7007827758789,
+      "learning_rate": 3e-06,
+      "loss": -31.7822,
+      "step": 1269
+    },
+    {
+      "epoch": 0.11288888888888889,
+      "grad_norm": 133.42088317871094,
+      "learning_rate": 3e-06,
+      "loss": -42.2768,
+      "step": 1270
+    },
+    {
+      "epoch": 0.11297777777777777,
+      "grad_norm": 128.7488555908203,
+      "learning_rate": 3e-06,
+      "loss": -30.9053,
+      "step": 1271
+    },
+    {
+      "epoch": 0.11306666666666666,
+      "grad_norm": 128.32147216796875,
+      "learning_rate": 3e-06,
+      "loss": -27.7986,
+      "step": 1272
+    },
+    {
+      "completion_length": 251.02083587646484,
+      "epoch": 0.11315555555555555,
+      "grad_norm": 276.54681396484375,
+      "learning_rate": 3e-06,
+      "loss": -8.3431,
+      "reward": 1.7916666865348816,
+      "reward_std": 0.4701542258262634,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.0416666567325592,
+      "step": 1273,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.11324444444444444,
+      "grad_norm": 234.2393341064453,
+      "learning_rate": 3e-06,
+      "loss": -13.7623,
+      "step": 1274
+    },
+    {
+      "epoch": 0.11333333333333333,
+      "grad_norm": 236.026611328125,
+      "learning_rate": 3e-06,
+      "loss": -4.7751,
+      "step": 1275
+    },
+    {
+      "epoch": 0.11342222222222222,
+      "grad_norm": 245.18170166015625,
+      "learning_rate": 3e-06,
+      "loss": -16.0847,
+      "step": 1276
+    },
+    {
+      "epoch": 0.1135111111111111,
+      "grad_norm": 314.53057861328125,
+      "learning_rate": 3e-06,
+      "loss": -8.6135,
+      "step": 1277
+    },
+    {
+      "epoch": 0.1136,
+      "grad_norm": 207.09188842773438,
+      "learning_rate": 3e-06,
+      "loss": -13.115,
+      "step": 1278
+    },
+    {
+      "epoch": 0.1136888888888889,
+      "grad_norm": 252.8507080078125,
+      "learning_rate": 3e-06,
+      "loss": -12.3574,
+      "step": 1279
+    },
+    {
+      "epoch": 0.11377777777777778,
+      "grad_norm": 241.2433319091797,
+      "learning_rate": 3e-06,
+      "loss": -18.4114,
+      "step": 1280
+    },
+    {
+      "epoch": 0.11386666666666667,
+      "grad_norm": 217.98683166503906,
+      "learning_rate": 3e-06,
+      "loss": -7.0416,
+      "step": 1281
+    },
+    {
+      "epoch": 0.11395555555555556,
+      "grad_norm": 252.15773010253906,
+      "learning_rate": 3e-06,
+      "loss": -19.5222,
+      "step": 1282
+    },
+    {
+      "epoch": 0.11404444444444445,
+      "grad_norm": 227.17205810546875,
+      "learning_rate": 3e-06,
+      "loss": -13.2521,
+      "step": 1283
+    },
+    {
+      "epoch": 0.11413333333333334,
+      "grad_norm": 235.24327087402344,
+      "learning_rate": 3e-06,
+      "loss": -17.9427,
+      "step": 1284
+    },
+    {
+      "completion_length": 253.25,
+      "epoch": 0.11422222222222222,
+      "grad_norm": 206.74746704101562,
+      "learning_rate": 3e-06,
+      "loss": 0.218,
+      "reward": 1.7187500596046448,
+      "reward_std": 0.5723656415939331,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 1.0416666567325592,
+      "step": 1285,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.11431111111111111,
+      "grad_norm": 207.45079040527344,
+      "learning_rate": 3e-06,
+      "loss": 23.7089,
+      "step": 1286
+    },
+    {
+      "epoch": 0.1144,
+      "grad_norm": 201.48416137695312,
+      "learning_rate": 3e-06,
+      "loss": 16.6222,
+      "step": 1287
+    },
+    {
+      "epoch": 0.11448888888888889,
+      "grad_norm": 240.6068115234375,
+      "learning_rate": 3e-06,
+      "loss": 2.8229,
+      "step": 1288
+    },
+    {
+      "epoch": 0.11457777777777778,
+      "grad_norm": 267.2186279296875,
+      "learning_rate": 3e-06,
+      "loss": -6.3361,
+      "step": 1289
+    },
+    {
+      "epoch": 0.11466666666666667,
+      "grad_norm": 254.57681274414062,
+      "learning_rate": 3e-06,
+      "loss": 2.6077,
+      "step": 1290
+    },
+    {
+      "epoch": 0.11475555555555556,
+      "grad_norm": 215.68649291992188,
+      "learning_rate": 3e-06,
+      "loss": -2.7734,
+      "step": 1291
+    },
+    {
+      "epoch": 0.11484444444444444,
+      "grad_norm": 227.70590209960938,
+      "learning_rate": 3e-06,
+      "loss": 18.5952,
+      "step": 1292
+    },
+    {
+      "epoch": 0.11493333333333333,
+      "grad_norm": 215.9313201904297,
+      "learning_rate": 3e-06,
+      "loss": 13.7655,
+      "step": 1293
+    },
+    {
+      "epoch": 0.11502222222222222,
+      "grad_norm": 251.1554718017578,
+      "learning_rate": 3e-06,
+      "loss": -1.3305,
+      "step": 1294
+    },
+    {
+      "epoch": 0.11511111111111111,
+      "grad_norm": 274.58538818359375,
+      "learning_rate": 3e-06,
+      "loss": -8.8694,
+      "step": 1295
+    },
+    {
+      "epoch": 0.1152,
+      "grad_norm": 238.6010284423828,
+      "learning_rate": 3e-06,
+      "loss": -1.8695,
+      "step": 1296
+    },
+    {
+      "completion_length": 252.1666717529297,
+      "epoch": 0.11528888888888889,
+      "grad_norm": 220.25823974609375,
+      "learning_rate": 3e-06,
+      "loss": -7.0592,
+      "reward": 1.0625,
+      "reward_std": 0.39512956142425537,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 1297,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.11537777777777777,
+      "grad_norm": 190.90309143066406,
+      "learning_rate": 3e-06,
+      "loss": -21.8086,
+      "step": 1298
+    },
+    {
+      "epoch": 0.11546666666666666,
+      "grad_norm": 167.63180541992188,
+      "learning_rate": 3e-06,
+      "loss": -10.4386,
+      "step": 1299
+    },
+    {
+      "epoch": 0.11555555555555555,
+      "grad_norm": 196.72048950195312,
+      "learning_rate": 3e-06,
+      "loss": -10.7261,
+      "step": 1300
+    },
+    {
+      "epoch": 0.11564444444444444,
+      "grad_norm": 243.5116424560547,
+      "learning_rate": 3e-06,
+      "loss": -21.7371,
+      "step": 1301
+    },
+    {
+      "epoch": 0.11573333333333333,
+      "grad_norm": 163.6327667236328,
+      "learning_rate": 3e-06,
+      "loss": -21.7888,
+      "step": 1302
+    },
+    {
+      "epoch": 0.11582222222222222,
+      "grad_norm": 217.04978942871094,
+      "learning_rate": 3e-06,
+      "loss": -7.9678,
+      "step": 1303
+    },
+    {
+      "epoch": 0.11591111111111112,
+      "grad_norm": 182.5911865234375,
+      "learning_rate": 3e-06,
+      "loss": -22.9758,
+      "step": 1304
+    },
+    {
+      "epoch": 0.116,
+      "grad_norm": 167.71888732910156,
+      "learning_rate": 3e-06,
+      "loss": -12.1331,
+      "step": 1305
+    },
+    {
+      "epoch": 0.1160888888888889,
+      "grad_norm": 168.7008819580078,
+      "learning_rate": 3e-06,
+      "loss": -13.2942,
+      "step": 1306
+    },
+    {
+      "epoch": 0.11617777777777778,
+      "grad_norm": 210.5468292236328,
+      "learning_rate": 3e-06,
+      "loss": -22.5768,
+      "step": 1307
+    },
+    {
+      "epoch": 0.11626666666666667,
+      "grad_norm": 179.2998046875,
+      "learning_rate": 3e-06,
+      "loss": -23.8159,
+      "step": 1308
+    },
+    {
+      "completion_length": 254.5416717529297,
+      "epoch": 0.11635555555555556,
+      "grad_norm": 219.9465789794922,
+      "learning_rate": 3e-06,
+      "loss": 7.1814,
+      "reward": 1.2916667461395264,
+      "reward_std": 0.48936043679714203,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.5416666716337204,
+      "step": 1309,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.11644444444444445,
+      "grad_norm": 224.42604064941406,
+      "learning_rate": 3e-06,
+      "loss": 8.354,
+      "step": 1310
+    },
+    {
+      "epoch": 0.11653333333333334,
+      "grad_norm": 262.8431091308594,
+      "learning_rate": 3e-06,
+      "loss": 5.5188,
+      "step": 1311
+    },
+    {
+      "epoch": 0.11662222222222222,
+      "grad_norm": 446.6980285644531,
+      "learning_rate": 3e-06,
+      "loss": 7.2822,
+      "step": 1312
+    },
+    {
+      "epoch": 0.11671111111111111,
+      "grad_norm": 200.77650451660156,
+      "learning_rate": 3e-06,
+      "loss": -0.5339,
+      "step": 1313
+    },
+    {
+      "epoch": 0.1168,
+      "grad_norm": 230.9583740234375,
+      "learning_rate": 3e-06,
+      "loss": -3.6372,
+      "step": 1314
+    },
+    {
+      "epoch": 0.11688888888888889,
+      "grad_norm": 257.7409362792969,
+      "learning_rate": 3e-06,
+      "loss": 5.1944,
+      "step": 1315
+    },
+    {
+      "epoch": 0.11697777777777778,
+      "grad_norm": 215.5637664794922,
+      "learning_rate": 3e-06,
+      "loss": 6.6842,
+      "step": 1316
+    },
+    {
+      "epoch": 0.11706666666666667,
+      "grad_norm": 228.1314697265625,
+      "learning_rate": 3e-06,
+      "loss": 3.1597,
+      "step": 1317
+    },
+    {
+      "epoch": 0.11715555555555555,
+      "grad_norm": 227.79237365722656,
+      "learning_rate": 3e-06,
+      "loss": 6.0615,
+      "step": 1318
+    },
+    {
+      "epoch": 0.11724444444444444,
+      "grad_norm": 364.67889404296875,
+      "learning_rate": 3e-06,
+      "loss": -4.7295,
+      "step": 1319
+    },
+    {
+      "epoch": 0.11733333333333333,
+      "grad_norm": 435.36578369140625,
+      "learning_rate": 3e-06,
+      "loss": -7.3331,
+      "step": 1320
+    },
+    {
+      "completion_length": 254.12500762939453,
+      "epoch": 0.11742222222222222,
+      "grad_norm": 181.09559631347656,
+      "learning_rate": 3e-06,
+      "loss": 9.7639,
+      "reward": 1.3437500596046448,
+      "reward_std": 0.3936076909303665,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.6666666865348816,
+      "step": 1321,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.11751111111111111,
+      "grad_norm": 184.9253387451172,
+      "learning_rate": 3e-06,
+      "loss": -7.8109,
+      "step": 1322
+    },
+    {
+      "epoch": 0.1176,
+      "grad_norm": 178.65553283691406,
+      "learning_rate": 3e-06,
+      "loss": -0.5036,
+      "step": 1323
+    },
+    {
+      "epoch": 0.11768888888888888,
+      "grad_norm": 172.2812042236328,
+      "learning_rate": 3e-06,
+      "loss": -5.0317,
+      "step": 1324
+    },
+    {
+      "epoch": 0.11777777777777777,
+      "grad_norm": 189.1818084716797,
+      "learning_rate": 3e-06,
+      "loss": -5.0896,
+      "step": 1325
+    },
+    {
+      "epoch": 0.11786666666666666,
+      "grad_norm": 195.15562438964844,
+      "learning_rate": 3e-06,
+      "loss": -6.7484,
+      "step": 1326
+    },
+    {
+      "epoch": 0.11795555555555555,
+      "grad_norm": 180.81617736816406,
+      "learning_rate": 3e-06,
+      "loss": 6.0087,
+      "step": 1327
+    },
+    {
+      "epoch": 0.11804444444444444,
+      "grad_norm": 181.2151336669922,
+      "learning_rate": 3e-06,
+      "loss": -12.1462,
+      "step": 1328
+    },
+    {
+      "epoch": 0.11813333333333334,
+      "grad_norm": 232.097900390625,
+      "learning_rate": 3e-06,
+      "loss": -6.1373,
+      "step": 1329
+    },
+    {
+      "epoch": 0.11822222222222223,
+      "grad_norm": 180.55462646484375,
+      "learning_rate": 3e-06,
+      "loss": -10.2139,
+      "step": 1330
+    },
+    {
+      "epoch": 0.11831111111111112,
+      "grad_norm": 192.82818603515625,
+      "learning_rate": 3e-06,
+      "loss": -11.1656,
+      "step": 1331
+    },
+    {
+      "epoch": 0.1184,
+      "grad_norm": 190.95399475097656,
+      "learning_rate": 3e-06,
+      "loss": -13.1171,
+      "step": 1332
+    },
+    {
+      "completion_length": 254.81250762939453,
+      "epoch": 0.1184888888888889,
+      "grad_norm": 65.9253921508789,
+      "learning_rate": 3e-06,
+      "loss": 2.6383,
+      "reward": 0.9791666865348816,
+      "reward_std": 0.10206206887960434,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.2916666679084301,
+      "step": 1333,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.11857777777777778,
+      "grad_norm": 66.89152526855469,
+      "learning_rate": 3e-06,
+      "loss": 0.0904,
+      "step": 1334
+    },
+    {
+      "epoch": 0.11866666666666667,
+      "grad_norm": 52.20970916748047,
+      "learning_rate": 3e-06,
+      "loss": -2.1923,
+      "step": 1335
+    },
+    {
+      "epoch": 0.11875555555555556,
+      "grad_norm": 62.94733810424805,
+      "learning_rate": 3e-06,
+      "loss": -0.1938,
+      "step": 1336
+    },
+    {
+      "epoch": 0.11884444444444445,
+      "grad_norm": 74.16914367675781,
+      "learning_rate": 3e-06,
+      "loss": -0.0211,
+      "step": 1337
+    },
+    {
+      "epoch": 0.11893333333333334,
+      "grad_norm": 60.86524963378906,
+      "learning_rate": 3e-06,
+      "loss": 1.9824,
+      "step": 1338
+    },
+    {
+      "epoch": 0.11902222222222222,
+      "grad_norm": 68.02703857421875,
+      "learning_rate": 3e-06,
+      "loss": 1.8138,
+      "step": 1339
+    },
+    {
+      "epoch": 0.11911111111111111,
+      "grad_norm": 65.81590270996094,
+      "learning_rate": 3e-06,
+      "loss": -0.8801,
+      "step": 1340
+    },
+    {
+      "epoch": 0.1192,
+      "grad_norm": 49.24789810180664,
+      "learning_rate": 3e-06,
+      "loss": -2.9454,
+      "step": 1341
+    },
+    {
+      "epoch": 0.11928888888888889,
+      "grad_norm": 66.03591918945312,
+      "learning_rate": 3e-06,
+      "loss": -2.1792,
+      "step": 1342
+    },
+    {
+      "epoch": 0.11937777777777778,
+      "grad_norm": 68.27337646484375,
+      "learning_rate": 3e-06,
+      "loss": -1.5692,
+      "step": 1343
+    },
+    {
+      "epoch": 0.11946666666666667,
+      "grad_norm": 76.02365112304688,
+      "learning_rate": 3e-06,
+      "loss": 1.3993,
+      "step": 1344
+    },
+    {
+      "completion_length": 242.02084350585938,
+      "epoch": 0.11955555555555555,
+      "grad_norm": 159.37770080566406,
+      "learning_rate": 3e-06,
+      "loss": -16.5166,
+      "reward": 1.3125,
+      "reward_std": 0.43528567254543304,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.625,
+      "step": 1345,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.11964444444444444,
+      "grad_norm": 156.10609436035156,
+      "learning_rate": 3e-06,
+      "loss": -19.2254,
+      "step": 1346
+    },
+    {
+      "epoch": 0.11973333333333333,
+      "grad_norm": 159.9058074951172,
+      "learning_rate": 3e-06,
+      "loss": -21.4448,
+      "step": 1347
+    },
+    {
+      "epoch": 0.11982222222222222,
+      "grad_norm": 217.29722595214844,
+      "learning_rate": 3e-06,
+      "loss": -22.2701,
+      "step": 1348
+    },
+    {
+      "epoch": 0.11991111111111111,
+      "grad_norm": 220.8108673095703,
+      "learning_rate": 3e-06,
+      "loss": -9.9609,
+      "step": 1349
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 255.4961395263672,
+      "learning_rate": 3e-06,
+      "loss": -10.7417,
+      "step": 1350
+    },
+    {
+      "epoch": 0.12008888888888888,
+      "grad_norm": 159.0548858642578,
+      "learning_rate": 3e-06,
+      "loss": -18.3206,
+      "step": 1351
+    },
+    {
+      "epoch": 0.12017777777777777,
+      "grad_norm": 151.7304229736328,
+      "learning_rate": 3e-06,
+      "loss": -21.7915,
+      "step": 1352
+    },
+    {
+      "epoch": 0.12026666666666666,
+      "grad_norm": 162.3264923095703,
+      "learning_rate": 3e-06,
+      "loss": -24.2052,
+      "step": 1353
+    },
+    {
+      "epoch": 0.12035555555555555,
+      "grad_norm": 188.41310119628906,
+      "learning_rate": 3e-06,
+      "loss": -26.0542,
+      "step": 1354
+    },
+    {
+      "epoch": 0.12044444444444445,
+      "grad_norm": 182.1976776123047,
+      "learning_rate": 3e-06,
+      "loss": -14.6291,
+      "step": 1355
+    },
+    {
+      "epoch": 0.12053333333333334,
+      "grad_norm": 227.4834747314453,
+      "learning_rate": 3e-06,
+      "loss": -16.78,
+      "step": 1356
+    },
+    {
+      "completion_length": 236.93750762939453,
+      "epoch": 0.12062222222222223,
+      "grad_norm": 244.82566833496094,
+      "learning_rate": 3e-06,
+      "loss": -29.5738,
+      "reward": 2.125,
+      "reward_std": 0.39512956142425537,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.375,
+      "step": 1357,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.12071111111111112,
+      "grad_norm": 560.8680419921875,
+      "learning_rate": 3e-06,
+      "loss": -35.2668,
+      "step": 1358
+    },
+    {
+      "epoch": 0.1208,
+      "grad_norm": 303.6029968261719,
+      "learning_rate": 3e-06,
+      "loss": -21.3551,
+      "step": 1359
+    },
+    {
+      "epoch": 0.12088888888888889,
+      "grad_norm": 247.47055053710938,
+      "learning_rate": 3e-06,
+      "loss": -30.3557,
+      "step": 1360
+    },
+    {
+      "epoch": 0.12097777777777778,
+      "grad_norm": 311.3307189941406,
+      "learning_rate": 3e-06,
+      "loss": -31.387,
+      "step": 1361
+    },
+    {
+      "epoch": 0.12106666666666667,
+      "grad_norm": 296.8590087890625,
+      "learning_rate": 3e-06,
+      "loss": -33.6326,
+      "step": 1362
+    },
+    {
+      "epoch": 0.12115555555555556,
+      "grad_norm": 253.0756378173828,
+      "learning_rate": 3e-06,
+      "loss": -31.4888,
+      "step": 1363
+    },
+    {
+      "epoch": 0.12124444444444445,
+      "grad_norm": 293.0926513671875,
+      "learning_rate": 3e-06,
+      "loss": -38.9989,
+      "step": 1364
+    },
+    {
+      "epoch": 0.12133333333333333,
+      "grad_norm": 318.564208984375,
+      "learning_rate": 3e-06,
+      "loss": -25.8663,
+      "step": 1365
+    },
+    {
+      "epoch": 0.12142222222222222,
+      "grad_norm": 262.5808410644531,
+      "learning_rate": 3e-06,
+      "loss": -33.3059,
+      "step": 1366
+    },
+    {
+      "epoch": 0.12151111111111111,
+      "grad_norm": 297.9688720703125,
+      "learning_rate": 3e-06,
+      "loss": -36.1247,
+      "step": 1367
+    },
+    {
+      "epoch": 0.1216,
+      "grad_norm": 319.11968994140625,
+      "learning_rate": 3e-06,
+      "loss": -37.8042,
+      "step": 1368
+    },
+    {
+      "completion_length": 241.52083587646484,
+      "epoch": 0.12168888888888889,
+      "grad_norm": 149.28683471679688,
+      "learning_rate": 3e-06,
+      "loss": -5.2185,
+      "reward": 0.979166716337204,
+      "reward_std": 0.306186206638813,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.2916666641831398,
+      "step": 1369,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.12177777777777778,
+      "grad_norm": 234.47120666503906,
+      "learning_rate": 3e-06,
+      "loss": -0.5601,
+      "step": 1370
+    },
+    {
+      "epoch": 0.12186666666666666,
+      "grad_norm": 168.30697631835938,
+      "learning_rate": 3e-06,
+      "loss": -6.9666,
+      "step": 1371
+    },
+    {
+      "epoch": 0.12195555555555555,
+      "grad_norm": 304.0892333984375,
+      "learning_rate": 3e-06,
+      "loss": -11.5806,
+      "step": 1372
+    },
+    {
+      "epoch": 0.12204444444444444,
+      "grad_norm": 210.6204071044922,
+      "learning_rate": 3e-06,
+      "loss": 3.7489,
+      "step": 1373
+    },
+    {
+      "epoch": 0.12213333333333333,
+      "grad_norm": 227.05795288085938,
+      "learning_rate": 3e-06,
+      "loss": -19.6276,
+      "step": 1374
+    },
+    {
+      "epoch": 0.12222222222222222,
+      "grad_norm": 148.8232879638672,
+      "learning_rate": 3e-06,
+      "loss": -8.3126,
+      "step": 1375
+    },
+    {
+      "epoch": 0.1223111111111111,
+      "grad_norm": 216.28646850585938,
+      "learning_rate": 3e-06,
+      "loss": -4.662,
+      "step": 1376
+    },
+    {
+      "epoch": 0.1224,
+      "grad_norm": 196.22518920898438,
+      "learning_rate": 3e-06,
+      "loss": -10.8248,
+      "step": 1377
+    },
+    {
+      "epoch": 0.12248888888888888,
+      "grad_norm": 238.0521697998047,
+      "learning_rate": 3e-06,
+      "loss": -16.9448,
+      "step": 1378
+    },
+    {
+      "epoch": 0.12257777777777777,
+      "grad_norm": 195.62860107421875,
+      "learning_rate": 3e-06,
+      "loss": -0.2301,
+      "step": 1379
+    },
+    {
+      "epoch": 0.12266666666666666,
+      "grad_norm": 201.47740173339844,
+      "learning_rate": 3e-06,
+      "loss": -26.1931,
+      "step": 1380
+    },
+    {
+      "completion_length": 223.375,
+      "epoch": 0.12275555555555556,
+      "grad_norm": 340.7679443359375,
+      "learning_rate": 3e-06,
+      "loss": -35.7739,
+      "reward": 1.5000000596046448,
+      "reward_std": 0.4779854714870453,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 1381,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.12284444444444445,
+      "grad_norm": 357.4112548828125,
+      "learning_rate": 3e-06,
+      "loss": -25.5447,
+      "step": 1382
+    },
+    {
+      "epoch": 0.12293333333333334,
+      "grad_norm": 355.23419189453125,
+      "learning_rate": 3e-06,
+      "loss": -29.5073,
+      "step": 1383
+    },
+    {
+      "epoch": 0.12302222222222223,
+      "grad_norm": 362.6241455078125,
+      "learning_rate": 3e-06,
+      "loss": -15.9052,
+      "step": 1384
+    },
+    {
+      "epoch": 0.12311111111111112,
+      "grad_norm": 388.32904052734375,
+      "learning_rate": 3e-06,
+      "loss": -16.7547,
+      "step": 1385
+    },
+    {
+      "epoch": 0.1232,
+      "grad_norm": 398.5157775878906,
+      "learning_rate": 3e-06,
+      "loss": -32.0836,
+      "step": 1386
+    },
+    {
+      "epoch": 0.12328888888888889,
+      "grad_norm": 256.07763671875,
+      "learning_rate": 3e-06,
+      "loss": -36.567,
+      "step": 1387
+    },
+    {
+      "epoch": 0.12337777777777778,
+      "grad_norm": 273.55108642578125,
+      "learning_rate": 3e-06,
+      "loss": -32.0325,
+      "step": 1388
+    },
+    {
+      "epoch": 0.12346666666666667,
+      "grad_norm": 298.35675048828125,
+      "learning_rate": 3e-06,
+      "loss": -33.9138,
+      "step": 1389
+    },
+    {
+      "epoch": 0.12355555555555556,
+      "grad_norm": 319.2604064941406,
+      "learning_rate": 3e-06,
+      "loss": -22.5196,
+      "step": 1390
+    },
+    {
+      "epoch": 0.12364444444444445,
+      "grad_norm": 321.7362976074219,
+      "learning_rate": 3e-06,
+      "loss": -22.5907,
+      "step": 1391
+    },
+    {
+      "epoch": 0.12373333333333333,
+      "grad_norm": 289.5376281738281,
+      "learning_rate": 3e-06,
+      "loss": -39.4107,
+      "step": 1392
+    },
+    {
+      "completion_length": 239.6041717529297,
+      "epoch": 0.12382222222222222,
+      "grad_norm": 189.32923889160156,
+      "learning_rate": 3e-06,
+      "loss": 33.0773,
+      "reward": 1.3541666865348816,
+      "reward_std": 0.23899273574352264,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.6666666567325592,
+      "step": 1393,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.12391111111111111,
+      "grad_norm": 280.063720703125,
+      "learning_rate": 3e-06,
+      "loss": 28.1537,
+      "step": 1394
+    },
+    {
+      "epoch": 0.124,
+      "grad_norm": 239.7598114013672,
+      "learning_rate": 3e-06,
+      "loss": 30.4262,
+      "step": 1395
+    },
+    {
+      "epoch": 0.12408888888888889,
+      "grad_norm": 230.4679718017578,
+      "learning_rate": 3e-06,
+      "loss": 26.9258,
+      "step": 1396
+    },
+    {
+      "epoch": 0.12417777777777778,
+      "grad_norm": 241.6261444091797,
+      "learning_rate": 3e-06,
+      "loss": 39.8305,
+      "step": 1397
+    },
+    {
+      "epoch": 0.12426666666666666,
+      "grad_norm": 305.56097412109375,
+      "learning_rate": 3e-06,
+      "loss": 29.9909,
+      "step": 1398
+    },
+    {
+      "epoch": 0.12435555555555555,
+      "grad_norm": 190.80581665039062,
+      "learning_rate": 3e-06,
+      "loss": 28.9829,
+      "step": 1399
+    },
+    {
+      "epoch": 0.12444444444444444,
+      "grad_norm": 272.18365478515625,
+      "learning_rate": 3e-06,
+      "loss": 21.7391,
+      "step": 1400
+    },
+    {
+      "epoch": 0.12453333333333333,
+      "grad_norm": 236.96322631835938,
+      "learning_rate": 3e-06,
+      "loss": 24.1388,
+      "step": 1401
+    },
+    {
+      "epoch": 0.12462222222222222,
+      "grad_norm": 257.829833984375,
+      "learning_rate": 3e-06,
+      "loss": 19.7315,
+      "step": 1402
+    },
+    {
+      "epoch": 0.1247111111111111,
+      "grad_norm": 237.62989807128906,
+      "learning_rate": 3e-06,
+      "loss": 32.9331,
+      "step": 1403
+    },
+    {
+      "epoch": 0.1248,
+      "grad_norm": 299.9111633300781,
+      "learning_rate": 3e-06,
+      "loss": 20.7263,
+      "step": 1404
+    },
+    {
+      "completion_length": 253.62500762939453,
+      "epoch": 0.12488888888888888,
+      "grad_norm": 351.87091064453125,
+      "learning_rate": 3e-06,
+      "loss": 1.3314,
+      "reward": 1.5416666865348816,
+      "reward_std": 0.48936042189598083,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.7916666716337204,
+      "step": 1405,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.12497777777777777,
+      "grad_norm": 318.410888671875,
+      "learning_rate": 3e-06,
+      "loss": -6.872,
+      "step": 1406
+    },
+    {
+      "epoch": 0.12506666666666666,
+      "grad_norm": 296.7760925292969,
+      "learning_rate": 3e-06,
+      "loss": 7.1205,
+      "step": 1407
+    },
+    {
+      "epoch": 0.12515555555555555,
+      "grad_norm": 264.91400146484375,
+      "learning_rate": 3e-06,
+      "loss": 18.3195,
+      "step": 1408
+    },
+    {
+      "epoch": 0.12524444444444444,
+      "grad_norm": 309.8560485839844,
+      "learning_rate": 3e-06,
+      "loss": -9.7835,
+      "step": 1409
+    },
+    {
+      "epoch": 0.12533333333333332,
+      "grad_norm": 343.082763671875,
+      "learning_rate": 3e-06,
+      "loss": 8.5194,
+      "step": 1410
+    },
+    {
+      "epoch": 0.1254222222222222,
+      "grad_norm": 349.1229248046875,
+      "learning_rate": 3e-06,
+      "loss": -2.322,
+      "step": 1411
+    },
+    {
+      "epoch": 0.1255111111111111,
+      "grad_norm": 343.0054626464844,
+      "learning_rate": 3e-06,
+      "loss": -10.4852,
+      "step": 1412
+    },
+    {
+      "epoch": 0.1256,
+      "grad_norm": 293.6500549316406,
+      "learning_rate": 3e-06,
+      "loss": 3.7367,
+      "step": 1413
+    },
+    {
+      "epoch": 0.12568888888888888,
+      "grad_norm": 244.24459838867188,
+      "learning_rate": 3e-06,
+      "loss": 12.8548,
+      "step": 1414
+    },
+    {
+      "epoch": 0.12577777777777777,
+      "grad_norm": 308.9840393066406,
+      "learning_rate": 3e-06,
+      "loss": -14.9646,
+      "step": 1415
+    },
+    {
+      "epoch": 0.12586666666666665,
+      "grad_norm": 302.1766052246094,
+      "learning_rate": 3e-06,
+      "loss": 4.8047,
+      "step": 1416
+    },
+    {
+      "completion_length": 249.6666717529297,
+      "epoch": 0.12595555555555554,
+      "grad_norm": 189.9821319580078,
+      "learning_rate": 3e-06,
+      "loss": -21.868,
+      "reward": 1.4583333730697632,
+      "reward_std": 0.4701542258262634,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.7083333134651184,
+      "step": 1417,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.12604444444444443,
+      "grad_norm": 230.71632385253906,
+      "learning_rate": 3e-06,
+      "loss": -6.3819,
+      "step": 1418
+    },
+    {
+      "epoch": 0.12613333333333332,
+      "grad_norm": 245.25421142578125,
+      "learning_rate": 3e-06,
+      "loss": 10.5483,
+      "step": 1419
+    },
+    {
+      "epoch": 0.12622222222222224,
+      "grad_norm": 153.20816040039062,
+      "learning_rate": 3e-06,
+      "loss": -8.6118,
+      "step": 1420
+    },
+    {
+      "epoch": 0.12631111111111112,
+      "grad_norm": 167.58921813964844,
+      "learning_rate": 3e-06,
+      "loss": -8.8097,
+      "step": 1421
+    },
+    {
+      "epoch": 0.1264,
+      "grad_norm": 190.3168182373047,
+      "learning_rate": 3e-06,
+      "loss": -14.9707,
+      "step": 1422
+    },
+    {
+      "epoch": 0.1264888888888889,
+      "grad_norm": 191.2883758544922,
+      "learning_rate": 3e-06,
+      "loss": -23.6776,
+      "step": 1423
+    },
+    {
+      "epoch": 0.1265777777777778,
+      "grad_norm": 223.39498901367188,
+      "learning_rate": 3e-06,
+      "loss": -8.999,
+      "step": 1424
+    },
+    {
+      "epoch": 0.12666666666666668,
+      "grad_norm": 228.12818908691406,
+      "learning_rate": 3e-06,
+      "loss": 7.6836,
+      "step": 1425
+    },
+    {
+      "epoch": 0.12675555555555557,
+      "grad_norm": 165.7634735107422,
+      "learning_rate": 3e-06,
+      "loss": -12.3289,
+      "step": 1426
+    },
+    {
+      "epoch": 0.12684444444444445,
+      "grad_norm": 169.15396118164062,
+      "learning_rate": 3e-06,
+      "loss": -13.59,
+      "step": 1427
+    },
+    {
+      "epoch": 0.12693333333333334,
+      "grad_norm": 204.25228881835938,
+      "learning_rate": 3e-06,
+      "loss": -19.9151,
+      "step": 1428
+    },
+    {
+      "completion_length": 228.02083587646484,
+      "epoch": 0.12702222222222223,
+      "grad_norm": 167.8898468017578,
+      "learning_rate": 3e-06,
+      "loss": 7.6334,
+      "reward": 1.2083333432674408,
+      "reward_std": 0.3602609783411026,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.4583333358168602,
+      "step": 1429,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.12711111111111112,
+      "grad_norm": 224.77040100097656,
+      "learning_rate": 3e-06,
+      "loss": -1.7483,
+      "step": 1430
+    },
+    {
+      "epoch": 0.1272,
+      "grad_norm": 220.69076538085938,
+      "learning_rate": 3e-06,
+      "loss": 14.763,
+      "step": 1431
+    },
+    {
+      "epoch": 0.1272888888888889,
+      "grad_norm": 150.462646484375,
+      "learning_rate": 3e-06,
+      "loss": 17.6324,
+      "step": 1432
+    },
+    {
+      "epoch": 0.12737777777777778,
+      "grad_norm": 203.99217224121094,
+      "learning_rate": 3e-06,
+      "loss": 17.7821,
+      "step": 1433
+    },
+    {
+      "epoch": 0.12746666666666667,
+      "grad_norm": 188.85665893554688,
+      "learning_rate": 3e-06,
+      "loss": 13.1598,
+      "step": 1434
+    },
+    {
+      "epoch": 0.12755555555555556,
+      "grad_norm": 155.11329650878906,
+      "learning_rate": 3e-06,
+      "loss": 6.2584,
+      "step": 1435
+    },
+    {
+      "epoch": 0.12764444444444445,
+      "grad_norm": 205.92800903320312,
+      "learning_rate": 3e-06,
+      "loss": -3.0899,
+      "step": 1436
+    },
+    {
+      "epoch": 0.12773333333333334,
+      "grad_norm": 201.03298950195312,
+      "learning_rate": 3e-06,
+      "loss": 12.9818,
+      "step": 1437
+    },
+    {
+      "epoch": 0.12782222222222223,
+      "grad_norm": 167.153076171875,
+      "learning_rate": 3e-06,
+      "loss": 15.1051,
+      "step": 1438
+    },
+    {
+      "epoch": 0.12791111111111111,
+      "grad_norm": 223.22909545898438,
+      "learning_rate": 3e-06,
+      "loss": 14.2815,
+      "step": 1439
+    },
+    {
+      "epoch": 0.128,
+      "grad_norm": 173.97694396972656,
+      "learning_rate": 3e-06,
+      "loss": 9.4034,
+      "step": 1440
+    },
+    {
+      "completion_length": 239.5416717529297,
+      "epoch": 0.1280888888888889,
+      "grad_norm": 299.9499816894531,
+      "learning_rate": 3e-06,
+      "loss": -27.7606,
+      "reward": 1.5208333730697632,
+      "reward_std": 0.3680921643972397,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.8333333432674408,
+      "step": 1441,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.12817777777777778,
+      "grad_norm": 236.49612426757812,
+      "learning_rate": 3e-06,
+      "loss": -27.7132,
+      "step": 1442
+    },
+    {
+      "epoch": 0.12826666666666667,
+      "grad_norm": 186.70510864257812,
+      "learning_rate": 3e-06,
+      "loss": -28.2186,
+      "step": 1443
+    },
+    {
+      "epoch": 0.12835555555555556,
+      "grad_norm": 294.79656982421875,
+      "learning_rate": 3e-06,
+      "loss": -18.5453,
+      "step": 1444
+    },
+    {
+      "epoch": 0.12844444444444444,
+      "grad_norm": 257.8788146972656,
+      "learning_rate": 3e-06,
+      "loss": -26.3342,
+      "step": 1445
+    },
+    {
+      "epoch": 0.12853333333333333,
+      "grad_norm": 280.1625061035156,
+      "learning_rate": 3e-06,
+      "loss": -37.5225,
+      "step": 1446
+    },
+    {
+      "epoch": 0.12862222222222222,
+      "grad_norm": 257.7731018066406,
+      "learning_rate": 3e-06,
+      "loss": -31.1507,
+      "step": 1447
+    },
+    {
+      "epoch": 0.1287111111111111,
+      "grad_norm": 221.82879638671875,
+      "learning_rate": 3e-06,
+      "loss": -30.7869,
+      "step": 1448
+    },
+    {
+      "epoch": 0.1288,
+      "grad_norm": 227.20188903808594,
+      "learning_rate": 3e-06,
+      "loss": -31.8637,
+      "step": 1449
+    },
+    {
+      "epoch": 0.1288888888888889,
+      "grad_norm": 319.30633544921875,
+      "learning_rate": 3e-06,
+      "loss": -24.0296,
+      "step": 1450
+    },
+    {
+      "epoch": 0.12897777777777777,
+      "grad_norm": 206.82269287109375,
+      "learning_rate": 3e-06,
+      "loss": -28.7864,
+      "step": 1451
+    },
+    {
+      "epoch": 0.12906666666666666,
+      "grad_norm": 196.76171875,
+      "learning_rate": 3e-06,
+      "loss": -42.0569,
+      "step": 1452
+    },
+    {
+      "completion_length": 248.1666717529297,
+      "epoch": 0.12915555555555555,
+      "grad_norm": 210.2032928466797,
+      "learning_rate": 3e-06,
+      "loss": -15.8435,
+      "reward": 1.5000000596046448,
+      "reward_std": 0.3680921941995621,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 1453,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.12924444444444444,
+      "grad_norm": 209.77688598632812,
+      "learning_rate": 3e-06,
+      "loss": -14.2805,
+      "step": 1454
+    },
+    {
+      "epoch": 0.12933333333333333,
+      "grad_norm": 662.65966796875,
+      "learning_rate": 3e-06,
+      "loss": -15.5908,
+      "step": 1455
+    },
+    {
+      "epoch": 0.12942222222222222,
+      "grad_norm": 176.72958374023438,
+      "learning_rate": 3e-06,
+      "loss": -15.1466,
+      "step": 1456
+    },
+    {
+      "epoch": 0.1295111111111111,
+      "grad_norm": 224.30841064453125,
+      "learning_rate": 3e-06,
+      "loss": -23.5471,
+      "step": 1457
+    },
+    {
+      "epoch": 0.1296,
+      "grad_norm": 187.04263305664062,
+      "learning_rate": 3e-06,
+      "loss": -15.5373,
+      "step": 1458
+    },
+    {
+      "epoch": 0.12968888888888888,
+      "grad_norm": 216.18629455566406,
+      "learning_rate": 3e-06,
+      "loss": -19.5843,
+      "step": 1459
+    },
+    {
+      "epoch": 0.12977777777777777,
+      "grad_norm": 204.7811279296875,
+      "learning_rate": 3e-06,
+      "loss": -17.6737,
+      "step": 1460
+    },
+    {
+      "epoch": 0.12986666666666666,
+      "grad_norm": 357.2877197265625,
+      "learning_rate": 3e-06,
+      "loss": -19.0236,
+      "step": 1461
+    },
+    {
+      "epoch": 0.12995555555555555,
+      "grad_norm": 173.1217803955078,
+      "learning_rate": 3e-06,
+      "loss": -18.2209,
+      "step": 1462
+    },
+    {
+      "epoch": 0.13004444444444443,
+      "grad_norm": 202.11126708984375,
+      "learning_rate": 3e-06,
+      "loss": -26.1672,
+      "step": 1463
+    },
+    {
+      "epoch": 0.13013333333333332,
+      "grad_norm": 215.072265625,
+      "learning_rate": 3e-06,
+      "loss": -18.211,
+      "step": 1464
+    },
+    {
+      "completion_length": 248.45833587646484,
+      "epoch": 0.1302222222222222,
+      "grad_norm": 189.24551391601562,
+      "learning_rate": 3e-06,
+      "loss": -26.8632,
+      "reward": 1.1041666865348816,
+      "reward_std": 0.23899272084236145,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.4166666716337204,
+      "step": 1465,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.1303111111111111,
+      "grad_norm": 150.01397705078125,
+      "learning_rate": 3e-06,
+      "loss": -36.2487,
+      "step": 1466
+    },
+    {
+      "epoch": 0.1304,
+      "grad_norm": 239.17184448242188,
+      "learning_rate": 3e-06,
+      "loss": -32.3168,
+      "step": 1467
+    },
+    {
+      "epoch": 0.13048888888888888,
+      "grad_norm": 214.16525268554688,
+      "learning_rate": 3e-06,
+      "loss": -36.6605,
+      "step": 1468
+    },
+    {
+      "epoch": 0.13057777777777776,
+      "grad_norm": 200.94650268554688,
+      "learning_rate": 3e-06,
+      "loss": -23.324,
+      "step": 1469
+    },
+    {
+      "epoch": 0.13066666666666665,
+      "grad_norm": 202.95838928222656,
+      "learning_rate": 3e-06,
+      "loss": -34.8067,
+      "step": 1470
+    },
+    {
+      "epoch": 0.13075555555555557,
+      "grad_norm": 147.8043670654297,
+      "learning_rate": 3e-06,
+      "loss": -29.223,
+      "step": 1471
+    },
+    {
+      "epoch": 0.13084444444444446,
+      "grad_norm": 137.1348114013672,
+      "learning_rate": 3e-06,
+      "loss": -38.384,
+      "step": 1472
+    },
+    {
+      "epoch": 0.13093333333333335,
+      "grad_norm": 198.6035614013672,
+      "learning_rate": 3e-06,
+      "loss": -35.8388,
+      "step": 1473
+    },
+    {
+      "epoch": 0.13102222222222223,
+      "grad_norm": 225.8579864501953,
+      "learning_rate": 3e-06,
+      "loss": -39.2579,
+      "step": 1474
+    },
+    {
+      "epoch": 0.13111111111111112,
+      "grad_norm": 193.4779052734375,
+      "learning_rate": 3e-06,
+      "loss": -25.8384,
+      "step": 1475
+    },
+    {
+      "epoch": 0.1312,
+      "grad_norm": 188.15464782714844,
+      "learning_rate": 3e-06,
+      "loss": -37.6788,
+      "step": 1476
+    },
+    {
+      "completion_length": 236.9375,
+      "epoch": 0.1312888888888889,
+      "grad_norm": 433.47833251953125,
+      "learning_rate": 3e-06,
+      "loss": -44.9082,
+      "reward": 1.9166667461395264,
+      "reward_std": 0.6611596345901489,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.1666666865348816,
+      "step": 1477,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.1313777777777778,
+      "grad_norm": 327.1938171386719,
+      "learning_rate": 3e-06,
+      "loss": -49.2004,
+      "step": 1478
+    },
+    {
+      "epoch": 0.13146666666666668,
+      "grad_norm": 295.2423095703125,
+      "learning_rate": 3e-06,
+      "loss": -47.3065,
+      "step": 1479
+    },
+    {
+      "epoch": 0.13155555555555556,
+      "grad_norm": 288.9835510253906,
+      "learning_rate": 3e-06,
+      "loss": -35.3959,
+      "step": 1480
+    },
+    {
+      "epoch": 0.13164444444444445,
+      "grad_norm": 310.02056884765625,
+      "learning_rate": 3e-06,
+      "loss": -36.4699,
+      "step": 1481
+    },
+    {
+      "epoch": 0.13173333333333334,
+      "grad_norm": 288.0977478027344,
+      "learning_rate": 3e-06,
+      "loss": -36.6703,
+      "step": 1482
+    },
+    {
+      "epoch": 0.13182222222222223,
+      "grad_norm": 266.6463317871094,
+      "learning_rate": 3e-06,
+      "loss": -47.7283,
+      "step": 1483
+    },
+    {
+      "epoch": 0.13191111111111112,
+      "grad_norm": 405.67071533203125,
+      "learning_rate": 3e-06,
+      "loss": -51.2972,
+      "step": 1484
+    },
+    {
+      "epoch": 0.132,
+      "grad_norm": 401.3482360839844,
+      "learning_rate": 3e-06,
+      "loss": -52.0379,
+      "step": 1485
+    },
+    {
+      "epoch": 0.1320888888888889,
+      "grad_norm": 280.8495788574219,
+      "learning_rate": 3e-06,
+      "loss": -39.6706,
+      "step": 1486
+    },
+    {
+      "epoch": 0.13217777777777778,
+      "grad_norm": 329.5269775390625,
+      "learning_rate": 3e-06,
+      "loss": -41.6526,
+      "step": 1487
+    },
+    {
+      "epoch": 0.13226666666666667,
+      "grad_norm": 335.015380859375,
+      "learning_rate": 3e-06,
+      "loss": -42.414,
+      "step": 1488
+    },
+    {
+      "completion_length": 244.70833587646484,
+      "epoch": 0.13235555555555556,
+      "grad_norm": 365.8779602050781,
+      "learning_rate": 3e-06,
+      "loss": 0.0487,
+      "reward": 1.1875,
+      "reward_std": 0.23116151988506317,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5625,
+      "rewards/correctness_reward_func_math": 0.625,
+      "step": 1489,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.13244444444444445,
+      "grad_norm": 270.31494140625,
+      "learning_rate": 3e-06,
+      "loss": -3.1951,
+      "step": 1490
+    },
+    {
+      "epoch": 0.13253333333333334,
+      "grad_norm": 350.0372619628906,
+      "learning_rate": 3e-06,
+      "loss": 2.4081,
+      "step": 1491
+    },
+    {
+      "epoch": 0.13262222222222222,
+      "grad_norm": 297.7118835449219,
+      "learning_rate": 3e-06,
+      "loss": 1.5965,
+      "step": 1492
+    },
+    {
+      "epoch": 0.1327111111111111,
+      "grad_norm": 251.00436401367188,
+      "learning_rate": 3e-06,
+      "loss": -8.6016,
+      "step": 1493
+    },
+    {
+      "epoch": 0.1328,
+      "grad_norm": 364.3514709472656,
+      "learning_rate": 3e-06,
+      "loss": 5.7272,
+      "step": 1494
+    },
+    {
+      "epoch": 0.1328888888888889,
+      "grad_norm": 327.8075256347656,
+      "learning_rate": 3e-06,
+      "loss": -6.1829,
+      "step": 1495
+    },
+    {
+      "epoch": 0.13297777777777778,
+      "grad_norm": 816.3570556640625,
+      "learning_rate": 3e-06,
+      "loss": -10.451,
+      "step": 1496
+    },
+    {
+      "epoch": 0.13306666666666667,
+      "grad_norm": 319.318115234375,
+      "learning_rate": 3e-06,
+      "loss": -7.2984,
+      "step": 1497
+    },
+    {
+      "epoch": 0.13315555555555555,
+      "grad_norm": 306.3028259277344,
+      "learning_rate": 3e-06,
+      "loss": -7.2124,
+      "step": 1498
+    },
+    {
+      "epoch": 0.13324444444444444,
+      "grad_norm": 194.4175567626953,
+      "learning_rate": 3e-06,
+      "loss": -14.7213,
+      "step": 1499
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 289.50408935546875,
+      "learning_rate": 3e-06,
+      "loss": -8.9605,
+      "step": 1500
+    },
+    {
+      "completion_length": 245.8541717529297,
+      "epoch": 0.13342222222222222,
+      "grad_norm": 177.177734375,
+      "learning_rate": 3e-06,
+      "loss": -19.9228,
+      "reward": 1.0000000596046448,
+      "reward_std": 0.23116152733564377,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 1501,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.1335111111111111,
+      "grad_norm": 256.49334716796875,
+      "learning_rate": 3e-06,
+      "loss": -12.3773,
+      "step": 1502
+    },
+    {
+      "epoch": 0.1336,
+      "grad_norm": 200.69879150390625,
+      "learning_rate": 3e-06,
+      "loss": -12.0433,
+      "step": 1503
+    },
+    {
+      "epoch": 0.13368888888888888,
+      "grad_norm": 195.25538635253906,
+      "learning_rate": 3e-06,
+      "loss": -5.3936,
+      "step": 1504
+    },
+    {
+      "epoch": 0.13377777777777777,
+      "grad_norm": 179.9781036376953,
+      "learning_rate": 3e-06,
+      "loss": -11.4454,
+      "step": 1505
+    },
+    {
+      "epoch": 0.13386666666666666,
+      "grad_norm": 255.14865112304688,
+      "learning_rate": 3e-06,
+      "loss": -14.0628,
+      "step": 1506
+    },
+    {
+      "epoch": 0.13395555555555555,
+      "grad_norm": 185.94732666015625,
+      "learning_rate": 3e-06,
+      "loss": -21.484,
+      "step": 1507
+    },
+    {
+      "epoch": 0.13404444444444444,
+      "grad_norm": 233.72573852539062,
+      "learning_rate": 3e-06,
+      "loss": -15.3775,
+      "step": 1508
+    },
+    {
+      "epoch": 0.13413333333333333,
+      "grad_norm": 329.163818359375,
+      "learning_rate": 3e-06,
+      "loss": -14.9938,
+      "step": 1509
+    },
+    {
+      "epoch": 0.13422222222222221,
+      "grad_norm": 199.61465454101562,
+      "learning_rate": 3e-06,
+      "loss": -7.8012,
+      "step": 1510
+    },
+    {
+      "epoch": 0.1343111111111111,
+      "grad_norm": 168.71255493164062,
+      "learning_rate": 3e-06,
+      "loss": -14.2907,
+      "step": 1511
+    },
+    {
+      "epoch": 0.1344,
+      "grad_norm": 246.87896728515625,
+      "learning_rate": 3e-06,
+      "loss": -15.4667,
+      "step": 1512
+    },
+    {
+      "completion_length": 240.18750762939453,
+      "epoch": 0.13448888888888888,
+      "grad_norm": 273.19427490234375,
+      "learning_rate": 3e-06,
+      "loss": -30.8428,
+      "reward": 1.5625,
+      "reward_std": 0.3410547822713852,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.875,
+      "step": 1513,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.13457777777777777,
+      "grad_norm": 198.8998565673828,
+      "learning_rate": 3e-06,
+      "loss": -36.7035,
+      "step": 1514
+    },
+    {
+      "epoch": 0.13466666666666666,
+      "grad_norm": 266.1399230957031,
+      "learning_rate": 3e-06,
+      "loss": -29.4233,
+      "step": 1515
+    },
+    {
+      "epoch": 0.13475555555555555,
+      "grad_norm": 270.2858581542969,
+      "learning_rate": 3e-06,
+      "loss": -36.6868,
+      "step": 1516
+    },
+    {
+      "epoch": 0.13484444444444443,
+      "grad_norm": 190.39768981933594,
+      "learning_rate": 3e-06,
+      "loss": -35.0192,
+      "step": 1517
+    },
+    {
+      "epoch": 0.13493333333333332,
+      "grad_norm": 319.56494140625,
+      "learning_rate": 3e-06,
+      "loss": -35.0648,
+      "step": 1518
+    },
+    {
+      "epoch": 0.1350222222222222,
+      "grad_norm": 224.65713500976562,
+      "learning_rate": 3e-06,
+      "loss": -35.3229,
+      "step": 1519
+    },
+    {
+      "epoch": 0.1351111111111111,
+      "grad_norm": 188.5618438720703,
+      "learning_rate": 3e-06,
+      "loss": -41.7633,
+      "step": 1520
+    },
+    {
+      "epoch": 0.1352,
+      "grad_norm": 334.20281982421875,
+      "learning_rate": 3e-06,
+      "loss": -34.6749,
+      "step": 1521
+    },
+    {
+      "epoch": 0.13528888888888888,
+      "grad_norm": 232.7653350830078,
+      "learning_rate": 3e-06,
+      "loss": -42.387,
+      "step": 1522
+    },
+    {
+      "epoch": 0.1353777777777778,
+      "grad_norm": 179.99102783203125,
+      "learning_rate": 3e-06,
+      "loss": -40.5993,
+      "step": 1523
+    },
+    {
+      "epoch": 0.13546666666666668,
+      "grad_norm": 266.2838134765625,
+      "learning_rate": 3e-06,
+      "loss": -43.1418,
+      "step": 1524
+    },
+    {
+      "completion_length": 251.06250762939453,
+      "epoch": 0.13555555555555557,
+      "grad_norm": 379.1654052734375,
+      "learning_rate": 3e-06,
+      "loss": 5.5695,
+      "reward": 1.0520833730697632,
+      "reward_std": 0.5148759335279465,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 1525,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.13564444444444446,
+      "grad_norm": 382.55987548828125,
+      "learning_rate": 3e-06,
+      "loss": -2.5248,
+      "step": 1526
+    },
+    {
+      "epoch": 0.13573333333333334,
+      "grad_norm": 535.1154174804688,
+      "learning_rate": 3e-06,
+      "loss": 2.5235,
+      "step": 1527
+    },
+    {
+      "epoch": 0.13582222222222223,
+      "grad_norm": 382.97515869140625,
+      "learning_rate": 3e-06,
+      "loss": 25.0744,
+      "step": 1528
+    },
+    {
+      "epoch": 0.13591111111111112,
+      "grad_norm": 417.135009765625,
+      "learning_rate": 3e-06,
+      "loss": -7.6026,
+      "step": 1529
+    },
+    {
+      "epoch": 0.136,
+      "grad_norm": 452.19580078125,
+      "learning_rate": 3e-06,
+      "loss": -7.3517,
+      "step": 1530
+    },
+    {
+      "epoch": 0.1360888888888889,
+      "grad_norm": 404.5818176269531,
+      "learning_rate": 3e-06,
+      "loss": 4.4363,
+      "step": 1531
+    },
+    {
+      "epoch": 0.1361777777777778,
+      "grad_norm": 378.475341796875,
+      "learning_rate": 3e-06,
+      "loss": -7.4401,
+      "step": 1532
+    },
+    {
+      "epoch": 0.13626666666666667,
+      "grad_norm": 446.7852478027344,
+      "learning_rate": 3e-06,
+      "loss": -2.1065,
+      "step": 1533
+    },
+    {
+      "epoch": 0.13635555555555556,
+      "grad_norm": 313.30340576171875,
+      "learning_rate": 3e-06,
+      "loss": 20.5916,
+      "step": 1534
+    },
+    {
+      "epoch": 0.13644444444444445,
+      "grad_norm": 385.8117370605469,
+      "learning_rate": 3e-06,
+      "loss": -12.1391,
+      "step": 1535
+    },
+    {
+      "epoch": 0.13653333333333334,
+      "grad_norm": 381.8157653808594,
+      "learning_rate": 3e-06,
+      "loss": -10.1047,
+      "step": 1536
+    },
+    {
+      "completion_length": 251.9166717529297,
+      "epoch": 0.13662222222222223,
+      "grad_norm": 151.30177307128906,
+      "learning_rate": 3e-06,
+      "loss": 7.8283,
+      "reward": 1.5,
+      "reward_std": 0.1369306445121765,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.875,
+      "step": 1537,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.13671111111111112,
+      "grad_norm": 152.4805908203125,
+      "learning_rate": 3e-06,
+      "loss": 2.9257,
+      "step": 1538
+    },
+    {
+      "epoch": 0.1368,
+      "grad_norm": 143.77023315429688,
+      "learning_rate": 3e-06,
+      "loss": 1.5204,
+      "step": 1539
+    },
+    {
+      "epoch": 0.1368888888888889,
+      "grad_norm": 134.1708526611328,
+      "learning_rate": 3e-06,
+      "loss": 2.3719,
+      "step": 1540
+    },
+    {
+      "epoch": 0.13697777777777778,
+      "grad_norm": 152.87826538085938,
+      "learning_rate": 3e-06,
+      "loss": 4.1701,
+      "step": 1541
+    },
+    {
+      "epoch": 0.13706666666666667,
+      "grad_norm": 140.14395141601562,
+      "learning_rate": 3e-06,
+      "loss": 5.3651,
+      "step": 1542
+    },
+    {
+      "epoch": 0.13715555555555556,
+      "grad_norm": 126.84275817871094,
+      "learning_rate": 3e-06,
+      "loss": 7.1642,
+      "step": 1543
+    },
+    {
+      "epoch": 0.13724444444444445,
+      "grad_norm": 168.67564392089844,
+      "learning_rate": 3e-06,
+      "loss": -0.2961,
+      "step": 1544
+    },
+    {
+      "epoch": 0.13733333333333334,
+      "grad_norm": 163.78794860839844,
+      "learning_rate": 3e-06,
+      "loss": -0.8864,
+      "step": 1545
+    },
+    {
+      "epoch": 0.13742222222222222,
+      "grad_norm": 125.58057403564453,
+      "learning_rate": 3e-06,
+      "loss": -0.0556,
+      "step": 1546
+    },
+    {
+      "epoch": 0.1375111111111111,
+      "grad_norm": 173.51966857910156,
+      "learning_rate": 3e-06,
+      "loss": 3.8643,
+      "step": 1547
+    },
+    {
+      "epoch": 0.1376,
+      "grad_norm": 120.23555755615234,
+      "learning_rate": 3e-06,
+      "loss": 1.9312,
+      "step": 1548
+    },
+    {
+      "completion_length": 254.1041717529297,
+      "epoch": 0.1376888888888889,
+      "grad_norm": 355.8426208496094,
+      "learning_rate": 3e-06,
+      "loss": -6.5178,
+      "reward": 1.1250000596046448,
+      "reward_std": 0.23116152733564377,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 1549,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.13777777777777778,
+      "grad_norm": 321.2608947753906,
+      "learning_rate": 3e-06,
+      "loss": -6.46,
+      "step": 1550
+    },
+    {
+      "epoch": 0.13786666666666667,
+      "grad_norm": 220.72544860839844,
+      "learning_rate": 3e-06,
+      "loss": -3.6261,
+      "step": 1551
+    },
+    {
+      "epoch": 0.13795555555555555,
+      "grad_norm": 475.98944091796875,
+      "learning_rate": 3e-06,
+      "loss": -20.6756,
+      "step": 1552
+    },
+    {
+      "epoch": 0.13804444444444444,
+      "grad_norm": 345.5748291015625,
+      "learning_rate": 3e-06,
+      "loss": -3.3905,
+      "step": 1553
+    },
+    {
+      "epoch": 0.13813333333333333,
+      "grad_norm": 285.5007629394531,
+      "learning_rate": 3e-06,
+      "loss": -2.0237,
+      "step": 1554
+    },
+    {
+      "epoch": 0.13822222222222222,
+      "grad_norm": 395.1439514160156,
+      "learning_rate": 3e-06,
+      "loss": -8.0344,
+      "step": 1555
+    },
+    {
+      "epoch": 0.1383111111111111,
+      "grad_norm": 300.19091796875,
+      "learning_rate": 3e-06,
+      "loss": -9.601,
+      "step": 1556
+    },
+    {
+      "epoch": 0.1384,
+      "grad_norm": 285.38763427734375,
+      "learning_rate": 3e-06,
+      "loss": -7.3983,
+      "step": 1557
+    },
+    {
+      "epoch": 0.13848888888888888,
+      "grad_norm": 463.34136962890625,
+      "learning_rate": 3e-06,
+      "loss": -25.5177,
+      "step": 1558
+    },
+    {
+      "epoch": 0.13857777777777777,
+      "grad_norm": 298.12005615234375,
+      "learning_rate": 3e-06,
+      "loss": -4.867,
+      "step": 1559
+    },
+    {
+      "epoch": 0.13866666666666666,
+      "grad_norm": 274.00836181640625,
+      "learning_rate": 3e-06,
+      "loss": -3.8962,
+      "step": 1560
+    },
+    {
+      "completion_length": 242.33334350585938,
+      "epoch": 0.13875555555555555,
+      "grad_norm": 310.84564208984375,
+      "learning_rate": 3e-06,
+      "loss": -16.4582,
+      "reward": 1.541666716337204,
+      "reward_std": 0.3061862289905548,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.7916666492819786,
+      "step": 1561,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.13884444444444444,
+      "grad_norm": 398.2083435058594,
+      "learning_rate": 3e-06,
+      "loss": -20.4413,
+      "step": 1562
+    },
+    {
+      "epoch": 0.13893333333333333,
+      "grad_norm": 378.9441833496094,
+      "learning_rate": 3e-06,
+      "loss": -16.1798,
+      "step": 1563
+    },
+    {
+      "epoch": 0.1390222222222222,
+      "grad_norm": 350.91192626953125,
+      "learning_rate": 3e-06,
+      "loss": -20.2111,
+      "step": 1564
+    },
+    {
+      "epoch": 0.1391111111111111,
+      "grad_norm": 365.54754638671875,
+      "learning_rate": 3e-06,
+      "loss": -17.1017,
+      "step": 1565
+    },
+    {
+      "epoch": 0.1392,
+      "grad_norm": 551.0444946289062,
+      "learning_rate": 3e-06,
+      "loss": -29.4716,
+      "step": 1566
+    },
+    {
+      "epoch": 0.13928888888888888,
+      "grad_norm": 320.9019775390625,
+      "learning_rate": 3e-06,
+      "loss": -20.3973,
+      "step": 1567
+    },
+    {
+      "epoch": 0.13937777777777777,
+      "grad_norm": 456.51190185546875,
+      "learning_rate": 3e-06,
+      "loss": -24.394,
+      "step": 1568
+    },
+    {
+      "epoch": 0.13946666666666666,
+      "grad_norm": 406.370361328125,
+      "learning_rate": 3e-06,
+      "loss": -20.4216,
+      "step": 1569
+    },
+    {
+      "epoch": 0.13955555555555554,
+      "grad_norm": 356.31982421875,
+      "learning_rate": 3e-06,
+      "loss": -24.4687,
+      "step": 1570
+    },
+    {
+      "epoch": 0.13964444444444443,
+      "grad_norm": 342.86468505859375,
+      "learning_rate": 3e-06,
+      "loss": -23.7611,
+      "step": 1571
+    },
+    {
+      "epoch": 0.13973333333333332,
+      "grad_norm": 536.0755615234375,
+      "learning_rate": 3e-06,
+      "loss": -33.8363,
+      "step": 1572
+    },
+    {
+      "completion_length": 254.18750762939453,
+      "epoch": 0.1398222222222222,
+      "grad_norm": 186.11480712890625,
+      "learning_rate": 3e-06,
+      "loss": 4.595,
+      "reward": 0.9791666865348816,
+      "reward_std": 0.23899271339178085,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5625,
+      "rewards/correctness_reward_func_math": 0.4166666567325592,
+      "step": 1573,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.13991111111111112,
+      "grad_norm": 166.36386108398438,
+      "learning_rate": 3e-06,
+      "loss": 6.7414,
+      "step": 1574
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 213.84388732910156,
+      "learning_rate": 3e-06,
+      "loss": 4.949,
+      "step": 1575
+    },
+    {
+      "epoch": 0.1400888888888889,
+      "grad_norm": 196.5020294189453,
+      "learning_rate": 3e-06,
+      "loss": 8.9449,
+      "step": 1576
+    },
+    {
+      "epoch": 0.1401777777777778,
+      "grad_norm": 216.48086547851562,
+      "learning_rate": 3e-06,
+      "loss": 7.0276,
+      "step": 1577
+    },
+    {
+      "epoch": 0.14026666666666668,
+      "grad_norm": 188.523681640625,
+      "learning_rate": 3e-06,
+      "loss": 5.1123,
+      "step": 1578
+    },
+    {
+      "epoch": 0.14035555555555557,
+      "grad_norm": 223.4918975830078,
+      "learning_rate": 3e-06,
+      "loss": 2.6615,
+      "step": 1579
+    },
+    {
+      "epoch": 0.14044444444444446,
+      "grad_norm": 218.71112060546875,
+      "learning_rate": 3e-06,
+      "loss": 3.1185,
+      "step": 1580
+    },
+    {
+      "epoch": 0.14053333333333334,
+      "grad_norm": 223.2500762939453,
+      "learning_rate": 3e-06,
+      "loss": 1.1539,
+      "step": 1581
+    },
+    {
+      "epoch": 0.14062222222222223,
+      "grad_norm": 176.69094848632812,
+      "learning_rate": 3e-06,
+      "loss": 5.1857,
+      "step": 1582
+    },
+    {
+      "epoch": 0.14071111111111112,
+      "grad_norm": 243.36929321289062,
+      "learning_rate": 3e-06,
+      "loss": 4.2964,
+      "step": 1583
+    },
+    {
+      "epoch": 0.1408,
+      "grad_norm": 269.2211608886719,
+      "learning_rate": 3e-06,
+      "loss": 1.8612,
+      "step": 1584
+    },
+    {
+      "completion_length": 245.9166717529297,
+      "epoch": 0.1408888888888889,
+      "grad_norm": 542.5291137695312,
+      "learning_rate": 3e-06,
+      "loss": 14.2794,
+      "reward": 0.9791666865348816,
+      "reward_std": 0.5643851011991501,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.2916666716337204,
+      "step": 1585,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.14097777777777779,
+      "grad_norm": 575.147705078125,
+      "learning_rate": 3e-06,
+      "loss": 17.5223,
+      "step": 1586
+    },
+    {
+      "epoch": 0.14106666666666667,
+      "grad_norm": 405.6719970703125,
+      "learning_rate": 3e-06,
+      "loss": 9.7247,
+      "step": 1587
+    },
+    {
+      "epoch": 0.14115555555555556,
+      "grad_norm": 528.659912109375,
+      "learning_rate": 3e-06,
+      "loss": 35.2638,
+      "step": 1588
+    },
+    {
+      "epoch": 0.14124444444444445,
+      "grad_norm": 514.4658203125,
+      "learning_rate": 3e-06,
+      "loss": 14.2266,
+      "step": 1589
+    },
+    {
+      "epoch": 0.14133333333333334,
+      "grad_norm": 535.8659057617188,
+      "learning_rate": 3e-06,
+      "loss": 13.3064,
+      "step": 1590
+    },
+    {
+      "epoch": 0.14142222222222223,
+      "grad_norm": 538.135009765625,
+      "learning_rate": 3e-06,
+      "loss": 7.7669,
+      "step": 1591
+    },
+    {
+      "epoch": 0.14151111111111112,
+      "grad_norm": 547.5822143554688,
+      "learning_rate": 3e-06,
+      "loss": 13.0588,
+      "step": 1592
+    },
+    {
+      "epoch": 0.1416,
+      "grad_norm": 415.9985046386719,
+      "learning_rate": 3e-06,
+      "loss": 4.3967,
+      "step": 1593
+    },
+    {
+      "epoch": 0.1416888888888889,
+      "grad_norm": 553.2412719726562,
+      "learning_rate": 3e-06,
+      "loss": 26.3526,
+      "step": 1594
+    },
+    {
+      "epoch": 0.14177777777777778,
+      "grad_norm": 512.5444946289062,
+      "learning_rate": 3e-06,
+      "loss": 5.7301,
+      "step": 1595
+    },
+    {
+      "epoch": 0.14186666666666667,
+      "grad_norm": 580.6036987304688,
+      "learning_rate": 3e-06,
+      "loss": 4.7651,
+      "step": 1596
+    },
+    {
+      "completion_length": 235.7916717529297,
+      "epoch": 0.14195555555555556,
+      "grad_norm": 347.6015319824219,
+      "learning_rate": 3e-06,
+      "loss": -18.3809,
+      "reward": 1.5416666865348816,
+      "reward_std": 0.3332235887646675,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.9166666567325592,
+      "step": 1597,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.14204444444444445,
+      "grad_norm": 317.97137451171875,
+      "learning_rate": 3e-06,
+      "loss": 2.7823,
+      "step": 1598
+    },
+    {
+      "epoch": 0.14213333333333333,
+      "grad_norm": 202.9899444580078,
+      "learning_rate": 3e-06,
+      "loss": -5.0647,
+      "step": 1599
+    },
+    {
+      "epoch": 0.14222222222222222,
+      "grad_norm": 786.5781860351562,
+      "learning_rate": 3e-06,
+      "loss": -4.8509,
+      "step": 1600
+    },
+    {
+      "epoch": 0.1423111111111111,
+      "grad_norm": 290.1077575683594,
+      "learning_rate": 3e-06,
+      "loss": -8.7908,
+      "step": 1601
+    },
+    {
+      "epoch": 0.1424,
+      "grad_norm": 198.83493041992188,
+      "learning_rate": 3e-06,
+      "loss": 8.0596,
+      "step": 1602
+    },
+    {
+      "epoch": 0.1424888888888889,
+      "grad_norm": 270.3849792480469,
+      "learning_rate": 3e-06,
+      "loss": -21.899,
+      "step": 1603
+    },
+    {
+      "epoch": 0.14257777777777778,
+      "grad_norm": 1309.112060546875,
+      "learning_rate": 3e-06,
+      "loss": -1.111,
+      "step": 1604
+    },
+    {
+      "epoch": 0.14266666666666666,
+      "grad_norm": 212.12266540527344,
+      "learning_rate": 3e-06,
+      "loss": -7.2685,
+      "step": 1605
+    },
+    {
+      "epoch": 0.14275555555555555,
+      "grad_norm": 242.00680541992188,
+      "learning_rate": 3e-06,
+      "loss": -7.1693,
+      "step": 1606
+    },
+    {
+      "epoch": 0.14284444444444444,
+      "grad_norm": 271.86090087890625,
+      "learning_rate": 3e-06,
+      "loss": -10.3562,
+      "step": 1607
+    },
+    {
+      "epoch": 0.14293333333333333,
+      "grad_norm": 233.82144165039062,
+      "learning_rate": 3e-06,
+      "loss": 4.7331,
+      "step": 1608
+    },
+    {
+      "completion_length": 254.1666717529297,
+      "epoch": 0.14302222222222222,
+      "grad_norm": 183.18118286132812,
+      "learning_rate": 3e-06,
+      "loss": -27.8021,
+      "reward": 1.8958333730697632,
+      "reward_std": 0.10206206887960434,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 1.2083333134651184,
+      "step": 1609,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.1431111111111111,
+      "grad_norm": 393.4090270996094,
+      "learning_rate": 3e-06,
+      "loss": -29.9682,
+      "step": 1610
+    },
+    {
+      "epoch": 0.1432,
+      "grad_norm": 232.8114776611328,
+      "learning_rate": 3e-06,
+      "loss": -21.6474,
+      "step": 1611
+    },
+    {
+      "epoch": 0.14328888888888888,
+      "grad_norm": 304.4367370605469,
+      "learning_rate": 3e-06,
+      "loss": -21.4802,
+      "step": 1612
+    },
+    {
+      "epoch": 0.14337777777777777,
+      "grad_norm": 162.92181396484375,
+      "learning_rate": 3e-06,
+      "loss": -28.1092,
+      "step": 1613
+    },
+    {
+      "epoch": 0.14346666666666666,
+      "grad_norm": 209.14356994628906,
+      "learning_rate": 3e-06,
+      "loss": -19.8933,
+      "step": 1614
+    },
+    {
+      "epoch": 0.14355555555555555,
+      "grad_norm": 697.1129760742188,
+      "learning_rate": 3e-06,
+      "loss": -27.563,
+      "step": 1615
+    },
+    {
+      "epoch": 0.14364444444444444,
+      "grad_norm": 311.41851806640625,
+      "learning_rate": 3e-06,
+      "loss": -33.0986,
+      "step": 1616
+    },
+    {
+      "epoch": 0.14373333333333332,
+      "grad_norm": 257.729248046875,
+      "learning_rate": 3e-06,
+      "loss": -26.3909,
+      "step": 1617
+    },
+    {
+      "epoch": 0.1438222222222222,
+      "grad_norm": 258.4046936035156,
+      "learning_rate": 3e-06,
+      "loss": -26.2488,
+      "step": 1618
+    },
+    {
+      "epoch": 0.1439111111111111,
+      "grad_norm": 130.1263885498047,
+      "learning_rate": 3e-06,
+      "loss": -30.1948,
+      "step": 1619
+    },
+    {
+      "epoch": 0.144,
+      "grad_norm": 182.67807006835938,
+      "learning_rate": 3e-06,
+      "loss": -23.4364,
+      "step": 1620
+    },
+    {
+      "completion_length": 249.4375,
+      "epoch": 0.14408888888888888,
+      "grad_norm": 354.3665771484375,
+      "learning_rate": 3e-06,
+      "loss": 8.1684,
+      "reward": 1.3229166865348816,
+      "reward_std": 0.15461497008800507,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.5833333432674408,
+      "step": 1621,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.14417777777777777,
+      "grad_norm": 424.4937438964844,
+      "learning_rate": 3e-06,
+      "loss": -0.1374,
+      "step": 1622
+    },
+    {
+      "epoch": 0.14426666666666665,
+      "grad_norm": 441.7712097167969,
+      "learning_rate": 3e-06,
+      "loss": -5.8721,
+      "step": 1623
+    },
+    {
+      "epoch": 0.14435555555555554,
+      "grad_norm": 474.1778259277344,
+      "learning_rate": 3e-06,
+      "loss": -4.5006,
+      "step": 1624
+    },
+    {
+      "epoch": 0.14444444444444443,
+      "grad_norm": 464.2291564941406,
+      "learning_rate": 3e-06,
+      "loss": -9.1126,
+      "step": 1625
+    },
+    {
+      "epoch": 0.14453333333333335,
+      "grad_norm": 375.41595458984375,
+      "learning_rate": 3e-06,
+      "loss": -1.7981,
+      "step": 1626
+    },
+    {
+      "epoch": 0.14462222222222224,
+      "grad_norm": 591.2286376953125,
+      "learning_rate": 3e-06,
+      "loss": -2.7549,
+      "step": 1627
+    },
+    {
+      "epoch": 0.14471111111111112,
+      "grad_norm": 269.577880859375,
+      "learning_rate": 3e-06,
+      "loss": -15.3618,
+      "step": 1628
+    },
+    {
+      "epoch": 0.1448,
+      "grad_norm": 366.1959533691406,
+      "learning_rate": 3e-06,
+      "loss": -17.4499,
+      "step": 1629
+    },
+    {
+      "epoch": 0.1448888888888889,
+      "grad_norm": 248.89236450195312,
+      "learning_rate": 3e-06,
+      "loss": -18.4987,
+      "step": 1630
+    },
+    {
+      "epoch": 0.1449777777777778,
+      "grad_norm": 173.4352569580078,
+      "learning_rate": 3e-06,
+      "loss": -16.0213,
+      "step": 1631
+    },
+    {
+      "epoch": 0.14506666666666668,
+      "grad_norm": 558.5076904296875,
+      "learning_rate": 3e-06,
+      "loss": -18.2749,
+      "step": 1632
+    },
+    {
+      "completion_length": 251.20833587646484,
+      "epoch": 0.14515555555555557,
+      "grad_norm": 273.303955078125,
+      "learning_rate": 3e-06,
+      "loss": -13.4676,
+      "reward": 1.6145833730697632,
+      "reward_std": 0.33129163831472397,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6979166567325592,
+      "rewards/correctness_reward_func_math": 0.9166666865348816,
+      "step": 1633,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.14524444444444445,
+      "grad_norm": 275.0503845214844,
+      "learning_rate": 3e-06,
+      "loss": -3.2551,
+      "step": 1634
+    },
+    {
+      "epoch": 0.14533333333333334,
+      "grad_norm": 399.2369384765625,
+      "learning_rate": 3e-06,
+      "loss": -5.5698,
+      "step": 1635
+    },
+    {
+      "epoch": 0.14542222222222223,
+      "grad_norm": 328.1588439941406,
+      "learning_rate": 3e-06,
+      "loss": -18.8269,
+      "step": 1636
+    },
+    {
+      "epoch": 0.14551111111111112,
+      "grad_norm": 239.95028686523438,
+      "learning_rate": 3e-06,
+      "loss": -6.0394,
+      "step": 1637
+    },
+    {
+      "epoch": 0.1456,
+      "grad_norm": 290.9996643066406,
+      "learning_rate": 3e-06,
+      "loss": -3.1009,
+      "step": 1638
+    },
+    {
+      "epoch": 0.1456888888888889,
+      "grad_norm": 334.3653869628906,
+      "learning_rate": 3e-06,
+      "loss": -17.4325,
+      "step": 1639
+    },
+    {
+      "epoch": 0.14577777777777778,
+      "grad_norm": 301.37139892578125,
+      "learning_rate": 3e-06,
+      "loss": -7.3761,
+      "step": 1640
+    },
+    {
+      "epoch": 0.14586666666666667,
+      "grad_norm": 593.1119995117188,
+      "learning_rate": 3e-06,
+      "loss": -12.5855,
+      "step": 1641
+    },
+    {
+      "epoch": 0.14595555555555556,
+      "grad_norm": 394.4963073730469,
+      "learning_rate": 3e-06,
+      "loss": -21.8609,
+      "step": 1642
+    },
+    {
+      "epoch": 0.14604444444444445,
+      "grad_norm": 239.55348205566406,
+      "learning_rate": 3e-06,
+      "loss": -9.8962,
+      "step": 1643
+    },
+    {
+      "epoch": 0.14613333333333334,
+      "grad_norm": 279.3672180175781,
+      "learning_rate": 3e-06,
+      "loss": -8.5061,
+      "step": 1644
+    },
+    {
+      "completion_length": 241.9166717529297,
+      "epoch": 0.14622222222222223,
+      "grad_norm": 477.223388671875,
+      "learning_rate": 3e-06,
+      "loss": -12.8053,
+      "reward": 1.1875000596046448,
+      "reward_std": 0.3332235887646675,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.4999999850988388,
+      "step": 1645,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.14631111111111111,
+      "grad_norm": 255.73138427734375,
+      "learning_rate": 3e-06,
+      "loss": 6.7089,
+      "step": 1646
+    },
+    {
+      "epoch": 0.1464,
+      "grad_norm": 414.9920959472656,
+      "learning_rate": 3e-06,
+      "loss": -17.6565,
+      "step": 1647
+    },
+    {
+      "epoch": 0.1464888888888889,
+      "grad_norm": 362.6864318847656,
+      "learning_rate": 3e-06,
+      "loss": -5.9471,
+      "step": 1648
+    },
+    {
+      "epoch": 0.14657777777777778,
+      "grad_norm": 257.5548400878906,
+      "learning_rate": 3e-06,
+      "loss": 7.0843,
+      "step": 1649
+    },
+    {
+      "epoch": 0.14666666666666667,
+      "grad_norm": 407.9867248535156,
+      "learning_rate": 3e-06,
+      "loss": -8.5254,
+      "step": 1650
+    },
+    {
+      "epoch": 0.14675555555555556,
+      "grad_norm": 326.8368225097656,
+      "learning_rate": 3e-06,
+      "loss": -14.7068,
+      "step": 1651
+    },
+    {
+      "epoch": 0.14684444444444444,
+      "grad_norm": 222.26805114746094,
+      "learning_rate": 3e-06,
+      "loss": 4.1678,
+      "step": 1652
+    },
+    {
+      "epoch": 0.14693333333333333,
+      "grad_norm": 361.6373596191406,
+      "learning_rate": 3e-06,
+      "loss": -22.3731,
+      "step": 1653
+    },
+    {
+      "epoch": 0.14702222222222222,
+      "grad_norm": 331.17803955078125,
+      "learning_rate": 3e-06,
+      "loss": -9.9171,
+      "step": 1654
+    },
+    {
+      "epoch": 0.1471111111111111,
+      "grad_norm": 365.830078125,
+      "learning_rate": 3e-06,
+      "loss": 1.295,
+      "step": 1655
+    },
+    {
+      "epoch": 0.1472,
+      "grad_norm": 400.97723388671875,
+      "learning_rate": 3e-06,
+      "loss": -15.896,
+      "step": 1656
+    },
+    {
+      "completion_length": 246.3541717529297,
+      "epoch": 0.14728888888888889,
+      "grad_norm": 407.52398681640625,
+      "learning_rate": 3e-06,
+      "loss": -95.2234,
+      "reward": 1.4583333730697632,
+      "reward_std": 0.6184598803520203,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5,
+      "rewards/correctness_reward_func_math": 0.9583333432674408,
+      "step": 1657,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.14737777777777777,
+      "grad_norm": 386.23016357421875,
+      "learning_rate": 3e-06,
+      "loss": -94.5723,
+      "step": 1658
+    },
+    {
+      "epoch": 0.14746666666666666,
+      "grad_norm": 337.7243347167969,
+      "learning_rate": 3e-06,
+      "loss": -78.7432,
+      "step": 1659
+    },
+    {
+      "epoch": 0.14755555555555555,
+      "grad_norm": 391.17547607421875,
+      "learning_rate": 3e-06,
+      "loss": -95.9404,
+      "step": 1660
+    },
+    {
+      "epoch": 0.14764444444444444,
+      "grad_norm": 462.5135498046875,
+      "learning_rate": 3e-06,
+      "loss": -69.7866,
+      "step": 1661
+    },
+    {
+      "epoch": 0.14773333333333333,
+      "grad_norm": 326.1936340332031,
+      "learning_rate": 3e-06,
+      "loss": -70.0077,
+      "step": 1662
+    },
+    {
+      "epoch": 0.14782222222222222,
+      "grad_norm": 417.0813903808594,
+      "learning_rate": 3e-06,
+      "loss": -104.9765,
+      "step": 1663
+    },
+    {
+      "epoch": 0.1479111111111111,
+      "grad_norm": 407.79150390625,
+      "learning_rate": 3e-06,
+      "loss": -99.052,
+      "step": 1664
+    },
+    {
+      "epoch": 0.148,
+      "grad_norm": 459.3847961425781,
+      "learning_rate": 3e-06,
+      "loss": -89.9644,
+      "step": 1665
+    },
+    {
+      "epoch": 0.14808888888888888,
+      "grad_norm": 449.5730285644531,
+      "learning_rate": 3e-06,
+      "loss": -106.9796,
+      "step": 1666
+    },
+    {
+      "epoch": 0.14817777777777777,
+      "grad_norm": 431.5627746582031,
+      "learning_rate": 3e-06,
+      "loss": -78.6433,
+      "step": 1667
+    },
+    {
+      "epoch": 0.14826666666666666,
+      "grad_norm": 356.53759765625,
+      "learning_rate": 3e-06,
+      "loss": -78.1133,
+      "step": 1668
+    },
+    {
+      "completion_length": 250.12500762939453,
+      "epoch": 0.14835555555555555,
+      "grad_norm": 798.4279174804688,
+      "learning_rate": 3e-06,
+      "loss": 13.9371,
+      "reward": 1.2500000596046448,
+      "reward_std": 0.778884083032608,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.625,
+      "step": 1669,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.14844444444444443,
+      "grad_norm": 674.0550537109375,
+      "learning_rate": 3e-06,
+      "loss": -39.024,
+      "step": 1670
+    },
+    {
+      "epoch": 0.14853333333333332,
+      "grad_norm": 686.0006103515625,
+      "learning_rate": 3e-06,
+      "loss": -26.081,
+      "step": 1671
+    },
+    {
+      "epoch": 0.1486222222222222,
+      "grad_norm": 854.5956420898438,
+      "learning_rate": 3e-06,
+      "loss": -6.9918,
+      "step": 1672
+    },
+    {
+      "epoch": 0.1487111111111111,
+      "grad_norm": 588.1673583984375,
+      "learning_rate": 3e-06,
+      "loss": -41.4908,
+      "step": 1673
+    },
+    {
+      "epoch": 0.1488,
+      "grad_norm": 802.58544921875,
+      "learning_rate": 3e-06,
+      "loss": -9.7048,
+      "step": 1674
+    },
+    {
+      "epoch": 0.14888888888888888,
+      "grad_norm": 842.8211669921875,
+      "learning_rate": 3e-06,
+      "loss": 3.388,
+      "step": 1675
+    },
+    {
+      "epoch": 0.14897777777777776,
+      "grad_norm": 603.185791015625,
+      "learning_rate": 3e-06,
+      "loss": -51.6478,
+      "step": 1676
+    },
+    {
+      "epoch": 0.14906666666666665,
+      "grad_norm": 667.0884399414062,
+      "learning_rate": 3e-06,
+      "loss": -34.1086,
+      "step": 1677
+    },
+    {
+      "epoch": 0.14915555555555557,
+      "grad_norm": 702.9710693359375,
+      "learning_rate": 3e-06,
+      "loss": -14.8077,
+      "step": 1678
+    },
+    {
+      "epoch": 0.14924444444444446,
+      "grad_norm": 887.6446533203125,
+      "learning_rate": 3e-06,
+      "loss": -53.0034,
+      "step": 1679
+    },
+    {
+      "epoch": 0.14933333333333335,
+      "grad_norm": 738.0032348632812,
+      "learning_rate": 3e-06,
+      "loss": -19.4303,
+      "step": 1680
+    },
+    {
+      "completion_length": 252.8541717529297,
+      "epoch": 0.14942222222222223,
+      "grad_norm": 563.926513671875,
+      "learning_rate": 3e-06,
+      "loss": 4.8877,
+      "reward": 1.0625,
+      "reward_std": 0.43528565764427185,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 1681,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.14951111111111112,
+      "grad_norm": 571.0492553710938,
+      "learning_rate": 3e-06,
+      "loss": -12.3935,
+      "step": 1682
+    },
+    {
+      "epoch": 0.1496,
+      "grad_norm": 365.8373107910156,
+      "learning_rate": 3e-06,
+      "loss": -5.9182,
+      "step": 1683
+    },
+    {
+      "epoch": 0.1496888888888889,
+      "grad_norm": 700.759033203125,
+      "learning_rate": 3e-06,
+      "loss": 14.5414,
+      "step": 1684
+    },
+    {
+      "epoch": 0.1497777777777778,
+      "grad_norm": 600.5524291992188,
+      "learning_rate": 3e-06,
+      "loss": 4.631,
+      "step": 1685
+    },
+    {
+      "epoch": 0.14986666666666668,
+      "grad_norm": 505.4430847167969,
+      "learning_rate": 3e-06,
+      "loss": -27.7184,
+      "step": 1686
+    },
+    {
+      "epoch": 0.14995555555555556,
+      "grad_norm": 464.21759033203125,
+      "learning_rate": 3e-06,
+      "loss": -0.9671,
+      "step": 1687
+    },
+    {
+      "epoch": 0.15004444444444445,
+      "grad_norm": 500.7622985839844,
+      "learning_rate": 3e-06,
+      "loss": -18.8001,
+      "step": 1688
+    },
+    {
+      "epoch": 0.15013333333333334,
+      "grad_norm": 369.3395080566406,
+      "learning_rate": 3e-06,
+      "loss": -9.8684,
+      "step": 1689
+    },
+    {
+      "epoch": 0.15022222222222223,
+      "grad_norm": 881.677001953125,
+      "learning_rate": 3e-06,
+      "loss": 6.5321,
+      "step": 1690
+    },
+    {
+      "epoch": 0.15031111111111112,
+      "grad_norm": 586.9358520507812,
+      "learning_rate": 3e-06,
+      "loss": -3.3023,
+      "step": 1691
+    },
+    {
+      "epoch": 0.1504,
+      "grad_norm": 406.6219787597656,
+      "learning_rate": 3e-06,
+      "loss": -33.3837,
+      "step": 1692
+    },
+    {
+      "completion_length": 250.125,
+      "epoch": 0.1504888888888889,
+      "grad_norm": 382.4415283203125,
+      "learning_rate": 3e-06,
+      "loss": 19.5614,
+      "reward": 1.3750000596046448,
+      "reward_std": 0.3602609932422638,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.625,
+      "step": 1693,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.15057777777777778,
+      "grad_norm": 349.4650573730469,
+      "learning_rate": 3e-06,
+      "loss": 22.1681,
+      "step": 1694
+    },
+    {
+      "epoch": 0.15066666666666667,
+      "grad_norm": 306.9416809082031,
+      "learning_rate": 3e-06,
+      "loss": 34.961,
+      "step": 1695
+    },
+    {
+      "epoch": 0.15075555555555556,
+      "grad_norm": 353.93121337890625,
+      "learning_rate": 3e-06,
+      "loss": 41.0782,
+      "step": 1696
+    },
+    {
+      "epoch": 0.15084444444444445,
+      "grad_norm": 213.22996520996094,
+      "learning_rate": 3e-06,
+      "loss": 26.4404,
+      "step": 1697
+    },
+    {
+      "epoch": 0.15093333333333334,
+      "grad_norm": 356.7481994628906,
+      "learning_rate": 3e-06,
+      "loss": 16.8735,
+      "step": 1698
+    },
+    {
+      "epoch": 0.15102222222222222,
+      "grad_norm": 394.1246643066406,
+      "learning_rate": 3e-06,
+      "loss": 15.6199,
+      "step": 1699
+    },
+    {
+      "epoch": 0.1511111111111111,
+      "grad_norm": 358.08831787109375,
+      "learning_rate": 3e-06,
+      "loss": 19.0218,
+      "step": 1700
+    },
+    {
+      "epoch": 0.1512,
+      "grad_norm": 290.66485595703125,
+      "learning_rate": 3e-06,
+      "loss": 30.4685,
+      "step": 1701
+    },
+    {
+      "epoch": 0.1512888888888889,
+      "grad_norm": 427.0489501953125,
+      "learning_rate": 3e-06,
+      "loss": 35.7429,
+      "step": 1702
+    },
+    {
+      "epoch": 0.15137777777777778,
+      "grad_norm": 218.4366912841797,
+      "learning_rate": 3e-06,
+      "loss": 22.0305,
+      "step": 1703
+    },
+    {
+      "epoch": 0.15146666666666667,
+      "grad_norm": 375.6819152832031,
+      "learning_rate": 3e-06,
+      "loss": 14.9698,
+      "step": 1704
+    },
+    {
+      "completion_length": 232.77083587646484,
+      "epoch": 0.15155555555555555,
+      "grad_norm": 213.29344177246094,
+      "learning_rate": 3e-06,
+      "loss": 5.0413,
+      "reward": 1.4791666865348816,
+      "reward_std": 0.10206206887960434,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.7916666567325592,
+      "step": 1705,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.15164444444444444,
+      "grad_norm": 168.3350067138672,
+      "learning_rate": 3e-06,
+      "loss": -3.6939,
+      "step": 1706
+    },
+    {
+      "epoch": 0.15173333333333333,
+      "grad_norm": 159.5379638671875,
+      "learning_rate": 3e-06,
+      "loss": 1.1446,
+      "step": 1707
+    },
+    {
+      "epoch": 0.15182222222222222,
+      "grad_norm": 144.53854370117188,
+      "learning_rate": 3e-06,
+      "loss": 3.3238,
+      "step": 1708
+    },
+    {
+      "epoch": 0.1519111111111111,
+      "grad_norm": 174.06390380859375,
+      "learning_rate": 3e-06,
+      "loss": -0.1332,
+      "step": 1709
+    },
+    {
+      "epoch": 0.152,
+      "grad_norm": 124.43144989013672,
+      "learning_rate": 3e-06,
+      "loss": -4.2987,
+      "step": 1710
+    },
+    {
+      "epoch": 0.15208888888888888,
+      "grad_norm": 242.69232177734375,
+      "learning_rate": 3e-06,
+      "loss": 1.5975,
+      "step": 1711
+    },
+    {
+      "epoch": 0.15217777777777777,
+      "grad_norm": 172.77381896972656,
+      "learning_rate": 3e-06,
+      "loss": -5.9669,
+      "step": 1712
+    },
+    {
+      "epoch": 0.15226666666666666,
+      "grad_norm": 139.244873046875,
+      "learning_rate": 3e-06,
+      "loss": -1.5727,
+      "step": 1713
+    },
+    {
+      "epoch": 0.15235555555555555,
+      "grad_norm": 133.6866455078125,
+      "learning_rate": 3e-06,
+      "loss": -0.2409,
+      "step": 1714
+    },
+    {
+      "epoch": 0.15244444444444444,
+      "grad_norm": 171.1123809814453,
+      "learning_rate": 3e-06,
+      "loss": -3.8625,
+      "step": 1715
+    },
+    {
+      "epoch": 0.15253333333333333,
+      "grad_norm": 110.9591064453125,
+      "learning_rate": 3e-06,
+      "loss": -6.3714,
+      "step": 1716
+    },
+    {
+      "completion_length": 254.12500762939453,
+      "epoch": 0.15262222222222221,
+      "grad_norm": 464.7311096191406,
+      "learning_rate": 3e-06,
+      "loss": -16.6423,
+      "reward": 1.2916666865348816,
+      "reward_std": 0.48936043679714203,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.5416666567325592,
+      "step": 1717,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.1527111111111111,
+      "grad_norm": 429.9415588378906,
+      "learning_rate": 3e-06,
+      "loss": -21.4782,
+      "step": 1718
+    },
+    {
+      "epoch": 0.1528,
+      "grad_norm": 415.4131774902344,
+      "learning_rate": 3e-06,
+      "loss": -38.5279,
+      "step": 1719
+    },
+    {
+      "epoch": 0.15288888888888888,
+      "grad_norm": 589.5310668945312,
+      "learning_rate": 3e-06,
+      "loss": -19.3757,
+      "step": 1720
+    },
+    {
+      "epoch": 0.15297777777777777,
+      "grad_norm": 492.2614440917969,
+      "learning_rate": 3e-06,
+      "loss": -28.3813,
+      "step": 1721
+    },
+    {
+      "epoch": 0.15306666666666666,
+      "grad_norm": 381.6370849609375,
+      "learning_rate": 3e-06,
+      "loss": -23.7667,
+      "step": 1722
+    },
+    {
+      "epoch": 0.15315555555555554,
+      "grad_norm": 448.4298095703125,
+      "learning_rate": 3e-06,
+      "loss": -20.063,
+      "step": 1723
+    },
+    {
+      "epoch": 0.15324444444444443,
+      "grad_norm": 351.5326232910156,
+      "learning_rate": 3e-06,
+      "loss": -25.6929,
+      "step": 1724
+    },
+    {
+      "epoch": 0.15333333333333332,
+      "grad_norm": 951.7434692382812,
+      "learning_rate": 3e-06,
+      "loss": -43.7247,
+      "step": 1725
+    },
+    {
+      "epoch": 0.1534222222222222,
+      "grad_norm": 457.6904602050781,
+      "learning_rate": 3e-06,
+      "loss": -27.8251,
+      "step": 1726
+    },
+    {
+      "epoch": 0.1535111111111111,
+      "grad_norm": 433.86907958984375,
+      "learning_rate": 3e-06,
+      "loss": -37.9928,
+      "step": 1727
+    },
+    {
+      "epoch": 0.1536,
+      "grad_norm": 506.2419128417969,
+      "learning_rate": 3e-06,
+      "loss": -31.6119,
+      "step": 1728
+    },
+    {
+      "completion_length": 250.25000762939453,
+      "epoch": 0.1536888888888889,
+      "grad_norm": 638.6153564453125,
+      "learning_rate": 3e-06,
+      "loss": -60.7029,
+      "reward": 1.4895833730697632,
+      "reward_std": 0.7616997957229614,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 1729,
+      "zero_std_ratio": 0.125
+    },
+    {
+      "epoch": 0.1537777777777778,
+      "grad_norm": 645.9075927734375,
+      "learning_rate": 3e-06,
+      "loss": -100.7913,
+      "step": 1730
+    },
+    {
+      "epoch": 0.15386666666666668,
+      "grad_norm": 769.6129760742188,
+      "learning_rate": 3e-06,
+      "loss": -41.8559,
+      "step": 1731
+    },
+    {
+      "epoch": 0.15395555555555557,
+      "grad_norm": 594.3479614257812,
+      "learning_rate": 3e-06,
+      "loss": -85.2107,
+      "step": 1732
+    },
+    {
+      "epoch": 0.15404444444444446,
+      "grad_norm": 513.4801635742188,
+      "learning_rate": 3e-06,
+      "loss": -78.7403,
+      "step": 1733
+    },
+    {
+      "epoch": 0.15413333333333334,
+      "grad_norm": 596.7926635742188,
+      "learning_rate": 3e-06,
+      "loss": -38.3655,
+      "step": 1734
+    },
+    {
+      "epoch": 0.15422222222222223,
+      "grad_norm": 762.2822875976562,
+      "learning_rate": 3e-06,
+      "loss": -68.4507,
+      "step": 1735
+    },
+    {
+      "epoch": 0.15431111111111112,
+      "grad_norm": 507.63958740234375,
+      "learning_rate": 3e-06,
+      "loss": -108.803,
+      "step": 1736
+    },
+    {
+      "epoch": 0.1544,
+      "grad_norm": 657.9226684570312,
+      "learning_rate": 3e-06,
+      "loss": -56.9832,
+      "step": 1737
+    },
+    {
+      "epoch": 0.1544888888888889,
+      "grad_norm": 604.657958984375,
+      "learning_rate": 3e-06,
+      "loss": -96.6592,
+      "step": 1738
+    },
+    {
+      "epoch": 0.1545777777777778,
+      "grad_norm": 491.42047119140625,
+      "learning_rate": 3e-06,
+      "loss": -87.8488,
+      "step": 1739
+    },
+    {
+      "epoch": 0.15466666666666667,
+      "grad_norm": 542.5538940429688,
+      "learning_rate": 3e-06,
+      "loss": -52.5061,
+      "step": 1740
+    },
+    {
+      "completion_length": 241.3125,
+      "epoch": 0.15475555555555556,
+      "grad_norm": 257.529541015625,
+      "learning_rate": 3e-06,
+      "loss": -24.5069,
+      "reward": 1.1250000596046448,
+      "reward_std": 0.23116152733564377,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 1741,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.15484444444444445,
+      "grad_norm": 272.9305725097656,
+      "learning_rate": 3e-06,
+      "loss": -31.0682,
+      "step": 1742
+    },
+    {
+      "epoch": 0.15493333333333334,
+      "grad_norm": 241.25071716308594,
+      "learning_rate": 3e-06,
+      "loss": -32.0895,
+      "step": 1743
+    },
+    {
+      "epoch": 0.15502222222222223,
+      "grad_norm": 208.89321899414062,
+      "learning_rate": 3e-06,
+      "loss": -28.8566,
+      "step": 1744
+    },
+    {
+      "epoch": 0.15511111111111112,
+      "grad_norm": 186.21788024902344,
+      "learning_rate": 3e-06,
+      "loss": -25.5258,
+      "step": 1745
+    },
+    {
+      "epoch": 0.1552,
+      "grad_norm": 208.84288024902344,
+      "learning_rate": 3e-06,
+      "loss": -36.6331,
+      "step": 1746
+    },
+    {
+      "epoch": 0.1552888888888889,
+      "grad_norm": 273.28900146484375,
+      "learning_rate": 3e-06,
+      "loss": -26.8255,
+      "step": 1747
+    },
+    {
+      "epoch": 0.15537777777777778,
+      "grad_norm": 255.3370361328125,
+      "learning_rate": 3e-06,
+      "loss": -35.3528,
+      "step": 1748
+    },
+    {
+      "epoch": 0.15546666666666667,
+      "grad_norm": 265.7087097167969,
+      "learning_rate": 3e-06,
+      "loss": -37.0957,
+      "step": 1749
+    },
+    {
+      "epoch": 0.15555555555555556,
+      "grad_norm": 174.1486358642578,
+      "learning_rate": 3e-06,
+      "loss": -31.8935,
+      "step": 1750
+    },
+    {
+      "epoch": 0.15564444444444445,
+      "grad_norm": 208.65518188476562,
+      "learning_rate": 3e-06,
+      "loss": -29.0371,
+      "step": 1751
+    },
+    {
+      "epoch": 0.15573333333333333,
+      "grad_norm": 172.7078857421875,
+      "learning_rate": 3e-06,
+      "loss": -39.3675,
+      "step": 1752
+    },
+    {
+      "completion_length": 252.39583587646484,
+      "epoch": 0.15582222222222222,
+      "grad_norm": 451.1358947753906,
+      "learning_rate": 3e-06,
+      "loss": -66.2963,
+      "reward": 1.3541667461395264,
+      "reward_std": 0.6070848852396011,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.6666666716337204,
+      "step": 1753,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.1559111111111111,
+      "grad_norm": 447.5511169433594,
+      "learning_rate": 3e-06,
+      "loss": -65.7659,
+      "step": 1754
+    },
+    {
+      "epoch": 0.156,
+      "grad_norm": 481.50335693359375,
+      "learning_rate": 3e-06,
+      "loss": -67.9006,
+      "step": 1755
+    },
+    {
+      "epoch": 0.1560888888888889,
+      "grad_norm": 418.3782043457031,
+      "learning_rate": 3e-06,
+      "loss": -59.6285,
+      "step": 1756
+    },
+    {
+      "epoch": 0.15617777777777778,
+      "grad_norm": 487.20574951171875,
+      "learning_rate": 3e-06,
+      "loss": -82.0834,
+      "step": 1757
+    },
+    {
+      "epoch": 0.15626666666666666,
+      "grad_norm": 454.06463623046875,
+      "learning_rate": 3e-06,
+      "loss": -57.3851,
+      "step": 1758
+    },
+    {
+      "epoch": 0.15635555555555555,
+      "grad_norm": 408.6988830566406,
+      "learning_rate": 3e-06,
+      "loss": -69.432,
+      "step": 1759
+    },
+    {
+      "epoch": 0.15644444444444444,
+      "grad_norm": 399.9183349609375,
+      "learning_rate": 3e-06,
+      "loss": -71.3301,
+      "step": 1760
+    },
+    {
+      "epoch": 0.15653333333333333,
+      "grad_norm": 577.1817626953125,
+      "learning_rate": 3e-06,
+      "loss": -74.2775,
+      "step": 1761
+    },
+    {
+      "epoch": 0.15662222222222222,
+      "grad_norm": 413.5326843261719,
+      "learning_rate": 3e-06,
+      "loss": -63.2282,
+      "step": 1762
+    },
+    {
+      "epoch": 0.1567111111111111,
+      "grad_norm": 498.8305358886719,
+      "learning_rate": 3e-06,
+      "loss": -88.1602,
+      "step": 1763
+    },
+    {
+      "epoch": 0.1568,
+      "grad_norm": 530.10595703125,
+      "learning_rate": 3e-06,
+      "loss": -64.1951,
+      "step": 1764
+    },
+    {
+      "completion_length": 247.00000762939453,
+      "epoch": 0.15688888888888888,
+      "grad_norm": 433.1192321777344,
+      "learning_rate": 3e-06,
+      "loss": -27.1999,
+      "reward": 1.6458333730697632,
+      "reward_std": 0.3602609783411026,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.9583333432674408,
+      "step": 1765,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.15697777777777777,
+      "grad_norm": 350.172119140625,
+      "learning_rate": 3e-06,
+      "loss": -32.7846,
+      "step": 1766
+    },
+    {
+      "epoch": 0.15706666666666666,
+      "grad_norm": 443.9689636230469,
+      "learning_rate": 3e-06,
+      "loss": -25.8718,
+      "step": 1767
+    },
+    {
+      "epoch": 0.15715555555555555,
+      "grad_norm": 491.6800537109375,
+      "learning_rate": 3e-06,
+      "loss": -21.288,
+      "step": 1768
+    },
+    {
+      "epoch": 0.15724444444444444,
+      "grad_norm": 490.1741943359375,
+      "learning_rate": 3e-06,
+      "loss": -30.8959,
+      "step": 1769
+    },
+    {
+      "epoch": 0.15733333333333333,
+      "grad_norm": 403.3340759277344,
+      "learning_rate": 3e-06,
+      "loss": -17.7178,
+      "step": 1770
+    },
+    {
+      "epoch": 0.1574222222222222,
+      "grad_norm": 434.2870178222656,
+      "learning_rate": 3e-06,
+      "loss": -28.6675,
+      "step": 1771
+    },
+    {
+      "epoch": 0.1575111111111111,
+      "grad_norm": 390.14208984375,
+      "learning_rate": 3e-06,
+      "loss": -35.2631,
+      "step": 1772
+    },
+    {
+      "epoch": 0.1576,
+      "grad_norm": 545.4449462890625,
+      "learning_rate": 3e-06,
+      "loss": -27.2021,
+      "step": 1773
+    },
+    {
+      "epoch": 0.15768888888888888,
+      "grad_norm": 409.76416015625,
+      "learning_rate": 3e-06,
+      "loss": -25.2074,
+      "step": 1774
+    },
+    {
+      "epoch": 0.15777777777777777,
+      "grad_norm": 1407.31787109375,
+      "learning_rate": 3e-06,
+      "loss": -32.1991,
+      "step": 1775
+    },
+    {
+      "epoch": 0.15786666666666666,
+      "grad_norm": 409.153076171875,
+      "learning_rate": 3e-06,
+      "loss": -21.3904,
+      "step": 1776
+    },
+    {
+      "completion_length": 255.6666717529297,
+      "epoch": 0.15795555555555554,
+      "grad_norm": 250.8477020263672,
+      "learning_rate": 3e-06,
+      "loss": -20.1606,
+      "reward": 1.0625000596046448,
+      "reward_std": 0.23116152733564377,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.3750000111758709,
+      "step": 1777,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.15804444444444443,
+      "grad_norm": 295.9480895996094,
+      "learning_rate": 3e-06,
+      "loss": -4.3906,
+      "step": 1778
+    },
+    {
+      "epoch": 0.15813333333333332,
+      "grad_norm": 264.98590087890625,
+      "learning_rate": 3e-06,
+      "loss": 1.1149,
+      "step": 1779
+    },
+    {
+      "epoch": 0.1582222222222222,
+      "grad_norm": 282.2425537109375,
+      "learning_rate": 3e-06,
+      "loss": -13.9569,
+      "step": 1780
+    },
+    {
+      "epoch": 0.15831111111111112,
+      "grad_norm": 269.65191650390625,
+      "learning_rate": 3e-06,
+      "loss": -12.0395,
+      "step": 1781
+    },
+    {
+      "epoch": 0.1584,
+      "grad_norm": 262.14825439453125,
+      "learning_rate": 3e-06,
+      "loss": -17.0425,
+      "step": 1782
+    },
+    {
+      "epoch": 0.1584888888888889,
+      "grad_norm": 226.9910888671875,
+      "learning_rate": 3e-06,
+      "loss": -22.3367,
+      "step": 1783
+    },
+    {
+      "epoch": 0.1585777777777778,
+      "grad_norm": 268.9870300292969,
+      "learning_rate": 3e-06,
+      "loss": -7.3806,
+      "step": 1784
+    },
+    {
+      "epoch": 0.15866666666666668,
+      "grad_norm": 252.59866333007812,
+      "learning_rate": 3e-06,
+      "loss": -2.0587,
+      "step": 1785
+    },
+    {
+      "epoch": 0.15875555555555557,
+      "grad_norm": 285.8102111816406,
+      "learning_rate": 3e-06,
+      "loss": -17.0219,
+      "step": 1786
+    },
+    {
+      "epoch": 0.15884444444444445,
+      "grad_norm": 281.6475830078125,
+      "learning_rate": 3e-06,
+      "loss": -16.6806,
+      "step": 1787
+    },
+    {
+      "epoch": 0.15893333333333334,
+      "grad_norm": 218.4907989501953,
+      "learning_rate": 3e-06,
+      "loss": -22.657,
+      "step": 1788
+    },
+    {
+      "completion_length": 252.0625,
+      "epoch": 0.15902222222222223,
+      "grad_norm": 365.0342102050781,
+      "learning_rate": 3e-06,
+      "loss": 4.7637,
+      "reward": 1.0416666865348816,
+      "reward_std": 0.23899271339178085,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.4166666679084301,
+      "step": 1789,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.15911111111111112,
+      "grad_norm": 306.0011291503906,
+      "learning_rate": 3e-06,
+      "loss": -12.1321,
+      "step": 1790
+    },
+    {
+      "epoch": 0.1592,
+      "grad_norm": 385.5239562988281,
+      "learning_rate": 3e-06,
+      "loss": -11.9155,
+      "step": 1791
+    },
+    {
+      "epoch": 0.1592888888888889,
+      "grad_norm": 502.8174743652344,
+      "learning_rate": 3e-06,
+      "loss": -9.7161,
+      "step": 1792
+    },
+    {
+      "epoch": 0.15937777777777778,
+      "grad_norm": 286.40557861328125,
+      "learning_rate": 3e-06,
+      "loss": -0.1257,
+      "step": 1793
+    },
+    {
+      "epoch": 0.15946666666666667,
+      "grad_norm": 250.90745544433594,
+      "learning_rate": 3e-06,
+      "loss": -5.5968,
+      "step": 1794
+    },
+    {
+      "epoch": 0.15955555555555556,
+      "grad_norm": 337.0714416503906,
+      "learning_rate": 3e-06,
+      "loss": 1.3961,
+      "step": 1795
+    },
+    {
+      "epoch": 0.15964444444444445,
+      "grad_norm": 279.7541198730469,
+      "learning_rate": 3e-06,
+      "loss": -15.7637,
+      "step": 1796
+    },
+    {
+      "epoch": 0.15973333333333334,
+      "grad_norm": 293.27703857421875,
+      "learning_rate": 3e-06,
+      "loss": -16.9533,
+      "step": 1797
+    },
+    {
+      "epoch": 0.15982222222222223,
+      "grad_norm": 398.4286193847656,
+      "learning_rate": 3e-06,
+      "loss": -14.4282,
+      "step": 1798
+    },
+    {
+      "epoch": 0.15991111111111111,
+      "grad_norm": 314.15338134765625,
+      "learning_rate": 3e-06,
+      "loss": -2.8244,
+      "step": 1799
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 302.6134338378906,
+      "learning_rate": 3e-06,
+      "loss": -11.5238,
+      "step": 1800
+    },
+    {
+      "completion_length": 251.93750762939453,
+      "epoch": 0.1600888888888889,
+      "grad_norm": 598.98095703125,
+      "learning_rate": 3e-06,
+      "loss": -25.8615,
+      "reward": 1.5,
+      "reward_std": 0.20412415266036987,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 1801,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.16017777777777778,
+      "grad_norm": 618.0178833007812,
+      "learning_rate": 3e-06,
+      "loss": -21.1122,
+      "step": 1802
+    },
+    {
+      "epoch": 0.16026666666666667,
+      "grad_norm": 541.5601806640625,
+      "learning_rate": 3e-06,
+      "loss": -11.0087,
+      "step": 1803
+    },
+    {
+      "epoch": 0.16035555555555556,
+      "grad_norm": 470.309814453125,
+      "learning_rate": 3e-06,
+      "loss": -22.2437,
+      "step": 1804
+    },
+    {
+      "epoch": 0.16044444444444445,
+      "grad_norm": 413.4715270996094,
+      "learning_rate": 3e-06,
+      "loss": -33.0321,
+      "step": 1805
+    },
+    {
+      "epoch": 0.16053333333333333,
+      "grad_norm": 376.2085266113281,
+      "learning_rate": 3e-06,
+      "loss": -25.3322,
+      "step": 1806
+    },
+    {
+      "epoch": 0.16062222222222222,
+      "grad_norm": 322.1478576660156,
+      "learning_rate": 3e-06,
+      "loss": -37.8067,
+      "step": 1807
+    },
+    {
+      "epoch": 0.1607111111111111,
+      "grad_norm": 272.68951416015625,
+      "learning_rate": 3e-06,
+      "loss": -34.8264,
+      "step": 1808
+    },
+    {
+      "epoch": 0.1608,
+      "grad_norm": 365.9797058105469,
+      "learning_rate": 3e-06,
+      "loss": -24.8187,
+      "step": 1809
+    },
+    {
+      "epoch": 0.1608888888888889,
+      "grad_norm": 314.506591796875,
+      "learning_rate": 3e-06,
+      "loss": -32.5667,
+      "step": 1810
+    },
+    {
+      "epoch": 0.16097777777777778,
+      "grad_norm": 410.4781494140625,
+      "learning_rate": 3e-06,
+      "loss": -43.7721,
+      "step": 1811
+    },
+    {
+      "epoch": 0.16106666666666666,
+      "grad_norm": 323.0709228515625,
+      "learning_rate": 3e-06,
+      "loss": -32.8399,
+      "step": 1812
+    },
+    {
+      "completion_length": 253.14583587646484,
+      "epoch": 0.16115555555555555,
+      "grad_norm": 490.2536315917969,
+      "learning_rate": 3e-06,
+      "loss": -24.4492,
+      "reward": 1.4375000596046448,
+      "reward_std": 0.3680921420454979,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.7500000149011612,
+      "step": 1813,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.16124444444444444,
+      "grad_norm": 376.43231201171875,
+      "learning_rate": 3e-06,
+      "loss": -22.8531,
+      "step": 1814
+    },
+    {
+      "epoch": 0.16133333333333333,
+      "grad_norm": 370.8981018066406,
+      "learning_rate": 3e-06,
+      "loss": -6.1589,
+      "step": 1815
+    },
+    {
+      "epoch": 0.16142222222222222,
+      "grad_norm": 350.6385498046875,
+      "learning_rate": 3e-06,
+      "loss": -17.2146,
+      "step": 1816
+    },
+    {
+      "epoch": 0.1615111111111111,
+      "grad_norm": 413.9906311035156,
+      "learning_rate": 3e-06,
+      "loss": -13.1823,
+      "step": 1817
+    },
+    {
+      "epoch": 0.1616,
+      "grad_norm": 511.4176940917969,
+      "learning_rate": 3e-06,
+      "loss": -23.6257,
+      "step": 1818
+    },
+    {
+      "epoch": 0.16168888888888888,
+      "grad_norm": 425.5303039550781,
+      "learning_rate": 3e-06,
+      "loss": -29.0499,
+      "step": 1819
+    },
+    {
+      "epoch": 0.16177777777777777,
+      "grad_norm": 329.032958984375,
+      "learning_rate": 3e-06,
+      "loss": -28.0294,
+      "step": 1820
+    },
+    {
+      "epoch": 0.16186666666666666,
+      "grad_norm": 482.425537109375,
+      "learning_rate": 3e-06,
+      "loss": -8.7538,
+      "step": 1821
+    },
+    {
+      "epoch": 0.16195555555555555,
+      "grad_norm": 422.88494873046875,
+      "learning_rate": 3e-06,
+      "loss": -22.7747,
+      "step": 1822
+    },
+    {
+      "epoch": 0.16204444444444444,
+      "grad_norm": 394.9844055175781,
+      "learning_rate": 3e-06,
+      "loss": -18.7768,
+      "step": 1823
+    },
+    {
+      "epoch": 0.16213333333333332,
+      "grad_norm": 449.5504455566406,
+      "learning_rate": 3e-06,
+      "loss": -25.7761,
+      "step": 1824
+    },
+    {
+      "completion_length": 240.8541717529297,
+      "epoch": 0.1622222222222222,
+      "grad_norm": 474.4821472167969,
+      "learning_rate": 3e-06,
+      "loss": 5.9184,
+      "reward": 1.5416666865348816,
+      "reward_std": 0.3602609857916832,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.7916666865348816,
+      "step": 1825,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.1623111111111111,
+      "grad_norm": 394.9033203125,
+      "learning_rate": 3e-06,
+      "loss": -8.6183,
+      "step": 1826
+    },
+    {
+      "epoch": 0.1624,
+      "grad_norm": 585.3305053710938,
+      "learning_rate": 3e-06,
+      "loss": 17.2736,
+      "step": 1827
+    },
+    {
+      "epoch": 0.16248888888888888,
+      "grad_norm": 411.5712585449219,
+      "learning_rate": 3e-06,
+      "loss": 2.0057,
+      "step": 1828
+    },
+    {
+      "epoch": 0.16257777777777777,
+      "grad_norm": 326.8497619628906,
+      "learning_rate": 3e-06,
+      "loss": 5.208,
+      "step": 1829
+    },
+    {
+      "epoch": 0.16266666666666665,
+      "grad_norm": 621.4788208007812,
+      "learning_rate": 3e-06,
+      "loss": -6.1221,
+      "step": 1830
+    },
+    {
+      "epoch": 0.16275555555555554,
+      "grad_norm": 628.4518432617188,
+      "learning_rate": 3e-06,
+      "loss": 3.4633,
+      "step": 1831
+    },
+    {
+      "epoch": 0.16284444444444446,
+      "grad_norm": 389.485595703125,
+      "learning_rate": 3e-06,
+      "loss": -14.53,
+      "step": 1832
+    },
+    {
+      "epoch": 0.16293333333333335,
+      "grad_norm": 760.7333374023438,
+      "learning_rate": 3e-06,
+      "loss": 11.3702,
+      "step": 1833
+    },
+    {
+      "epoch": 0.16302222222222224,
+      "grad_norm": 327.4570617675781,
+      "learning_rate": 3e-06,
+      "loss": -4.0738,
+      "step": 1834
+    },
+    {
+      "epoch": 0.16311111111111112,
+      "grad_norm": 325.6021728515625,
+      "learning_rate": 3e-06,
+      "loss": -0.9882,
+      "step": 1835
+    },
+    {
+      "epoch": 0.1632,
+      "grad_norm": 434.9088439941406,
+      "learning_rate": 3e-06,
+      "loss": -14.5048,
+      "step": 1836
+    },
+    {
+      "completion_length": 252.3541717529297,
+      "epoch": 0.1632888888888889,
+      "grad_norm": 364.32318115234375,
+      "learning_rate": 3e-06,
+      "loss": 5.0981,
+      "reward": 1.520833432674408,
+      "reward_std": 0.20412413775920868,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.8333333544433117,
+      "step": 1837,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.1633777777777778,
+      "grad_norm": 327.6516418457031,
+      "learning_rate": 3e-06,
+      "loss": 13.1354,
+      "step": 1838
+    },
+    {
+      "epoch": 0.16346666666666668,
+      "grad_norm": 327.8387145996094,
+      "learning_rate": 3e-06,
+      "loss": 7.7064,
+      "step": 1839
+    },
+    {
+      "epoch": 0.16355555555555557,
+      "grad_norm": 483.0592346191406,
+      "learning_rate": 3e-06,
+      "loss": 9.8312,
+      "step": 1840
+    },
+    {
+      "epoch": 0.16364444444444445,
+      "grad_norm": 409.86724853515625,
+      "learning_rate": 3e-06,
+      "loss": 13.0568,
+      "step": 1841
+    },
+    {
+      "epoch": 0.16373333333333334,
+      "grad_norm": 282.73626708984375,
+      "learning_rate": 3e-06,
+      "loss": 3.5377,
+      "step": 1842
+    },
+    {
+      "epoch": 0.16382222222222223,
+      "grad_norm": 262.7396240234375,
+      "learning_rate": 3e-06,
+      "loss": -0.2709,
+      "step": 1843
+    },
+    {
+      "epoch": 0.16391111111111112,
+      "grad_norm": 375.62359619140625,
+      "learning_rate": 3e-06,
+      "loss": 8.4106,
+      "step": 1844
+    },
+    {
+      "epoch": 0.164,
+      "grad_norm": 295.78814697265625,
+      "learning_rate": 3e-06,
+      "loss": 2.2675,
+      "step": 1845
+    },
+    {
+      "epoch": 0.1640888888888889,
+      "grad_norm": 555.635009765625,
+      "learning_rate": 3e-06,
+      "loss": 1.7638,
+      "step": 1846
+    },
+    {
+      "epoch": 0.16417777777777778,
+      "grad_norm": 362.09722900390625,
+      "learning_rate": 3e-06,
+      "loss": 2.102,
+      "step": 1847
+    },
+    {
+      "epoch": 0.16426666666666667,
+      "grad_norm": 349.70440673828125,
+      "learning_rate": 3e-06,
+      "loss": -2.454,
+      "step": 1848
+    },
+    {
+      "completion_length": 249.08333587646484,
+      "epoch": 0.16435555555555556,
+      "grad_norm": 437.3484802246094,
+      "learning_rate": 3e-06,
+      "loss": -29.0477,
+      "reward": 1.4166666865348816,
+      "reward_std": 0.4701542556285858,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.7916666567325592,
+      "step": 1849,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.16444444444444445,
+      "grad_norm": 638.8463745117188,
+      "learning_rate": 3e-06,
+      "loss": -47.733,
+      "step": 1850
+    },
+    {
+      "epoch": 0.16453333333333334,
+      "grad_norm": 460.33624267578125,
+      "learning_rate": 3e-06,
+      "loss": -34.4307,
+      "step": 1851
+    },
+    {
+      "epoch": 0.16462222222222223,
+      "grad_norm": 491.926025390625,
+      "learning_rate": 3e-06,
+      "loss": -52.2731,
+      "step": 1852
+    },
+    {
+      "epoch": 0.1647111111111111,
+      "grad_norm": 607.3854370117188,
+      "learning_rate": 3e-06,
+      "loss": -47.2157,
+      "step": 1853
+    },
+    {
+      "epoch": 0.1648,
+      "grad_norm": 512.1332397460938,
+      "learning_rate": 3e-06,
+      "loss": -50.0155,
+      "step": 1854
+    },
+    {
+      "epoch": 0.1648888888888889,
+      "grad_norm": 403.7186279296875,
+      "learning_rate": 3e-06,
+      "loss": -33.8899,
+      "step": 1855
+    },
+    {
+      "epoch": 0.16497777777777778,
+      "grad_norm": 423.3454284667969,
+      "learning_rate": 3e-06,
+      "loss": -51.1046,
+      "step": 1856
+    },
+    {
+      "epoch": 0.16506666666666667,
+      "grad_norm": 615.0731811523438,
+      "learning_rate": 3e-06,
+      "loss": -36.3091,
+      "step": 1857
+    },
+    {
+      "epoch": 0.16515555555555556,
+      "grad_norm": 557.7341918945312,
+      "learning_rate": 3e-06,
+      "loss": -58.0938,
+      "step": 1858
+    },
+    {
+      "epoch": 0.16524444444444444,
+      "grad_norm": 575.3082885742188,
+      "learning_rate": 3e-06,
+      "loss": -56.4051,
+      "step": 1859
+    },
+    {
+      "epoch": 0.16533333333333333,
+      "grad_norm": 530.2061157226562,
+      "learning_rate": 3e-06,
+      "loss": -57.1385,
+      "step": 1860
+    },
+    {
+      "completion_length": 229.4791717529297,
+      "epoch": 0.16542222222222222,
+      "grad_norm": 290.0047912597656,
+      "learning_rate": 3e-06,
+      "loss": -26.4189,
+      "reward": 1.4375000596046448,
+      "reward_std": 0.3332236111164093,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 1861,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.1655111111111111,
+      "grad_norm": 343.6731262207031,
+      "learning_rate": 3e-06,
+      "loss": -30.5266,
+      "step": 1862
+    },
+    {
+      "epoch": 0.1656,
+      "grad_norm": 321.04400634765625,
+      "learning_rate": 3e-06,
+      "loss": -26.9367,
+      "step": 1863
+    },
+    {
+      "epoch": 0.16568888888888889,
+      "grad_norm": 423.7695007324219,
+      "learning_rate": 3e-06,
+      "loss": -20.3331,
+      "step": 1864
+    },
+    {
+      "epoch": 0.16577777777777777,
+      "grad_norm": 301.9991760253906,
+      "learning_rate": 3e-06,
+      "loss": -22.0128,
+      "step": 1865
+    },
+    {
+      "epoch": 0.16586666666666666,
+      "grad_norm": 266.446533203125,
+      "learning_rate": 3e-06,
+      "loss": -13.6451,
+      "step": 1866
+    },
+    {
+      "epoch": 0.16595555555555555,
+      "grad_norm": 289.8598937988281,
+      "learning_rate": 3e-06,
+      "loss": -29.9844,
+      "step": 1867
+    },
+    {
+      "epoch": 0.16604444444444444,
+      "grad_norm": 474.86016845703125,
+      "learning_rate": 3e-06,
+      "loss": -32.2196,
+      "step": 1868
+    },
+    {
+      "epoch": 0.16613333333333333,
+      "grad_norm": 314.1653137207031,
+      "learning_rate": 3e-06,
+      "loss": -29.7642,
+      "step": 1869
+    },
+    {
+      "epoch": 0.16622222222222222,
+      "grad_norm": 362.4693603515625,
+      "learning_rate": 3e-06,
+      "loss": -24.4635,
+      "step": 1870
+    },
+    {
+      "epoch": 0.1663111111111111,
+      "grad_norm": 312.52569580078125,
+      "learning_rate": 3e-06,
+      "loss": -24.009,
+      "step": 1871
+    },
+    {
+      "epoch": 0.1664,
+      "grad_norm": 259.9459228515625,
+      "learning_rate": 3e-06,
+      "loss": -18.4052,
+      "step": 1872
+    },
+    {
+      "completion_length": 250.8541717529297,
+      "epoch": 0.16648888888888888,
+      "grad_norm": 374.1226501464844,
+      "learning_rate": 3e-06,
+      "loss": -4.5437,
+      "reward": 1.6666667461395264,
+      "reward_std": 0.20412414520978928,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.9166666865348816,
+      "step": 1873,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.16657777777777777,
+      "grad_norm": 501.4902038574219,
+      "learning_rate": 3e-06,
+      "loss": -16.285,
+      "step": 1874
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 511.9213562011719,
+      "learning_rate": 3e-06,
+      "loss": -5.965,
+      "step": 1875
+    },
+    {
+      "epoch": 0.16675555555555555,
+      "grad_norm": 377.8996276855469,
+      "learning_rate": 3e-06,
+      "loss": -11.742,
+      "step": 1876
+    },
+    {
+      "epoch": 0.16684444444444443,
+      "grad_norm": 275.3067321777344,
+      "learning_rate": 3e-06,
+      "loss": -2.8953,
+      "step": 1877
+    },
+    {
+      "epoch": 0.16693333333333332,
+      "grad_norm": 311.0989990234375,
+      "learning_rate": 3e-06,
+      "loss": 3.1096,
+      "step": 1878
+    },
+    {
+      "epoch": 0.1670222222222222,
+      "grad_norm": 408.4767761230469,
+      "learning_rate": 3e-06,
+      "loss": -5.1197,
+      "step": 1879
+    },
+    {
+      "epoch": 0.1671111111111111,
+      "grad_norm": 411.1174621582031,
+      "learning_rate": 3e-06,
+      "loss": -20.4177,
+      "step": 1880
+    },
+    {
+      "epoch": 0.1672,
+      "grad_norm": 432.32159423828125,
+      "learning_rate": 3e-06,
+      "loss": -14.0135,
+      "step": 1881
+    },
+    {
+      "epoch": 0.16728888888888888,
+      "grad_norm": 637.0897216796875,
+      "learning_rate": 3e-06,
+      "loss": -17.1271,
+      "step": 1882
+    },
+    {
+      "epoch": 0.16737777777777776,
+      "grad_norm": 277.6168212890625,
+      "learning_rate": 3e-06,
+      "loss": -7.087,
+      "step": 1883
+    },
+    {
+      "epoch": 0.16746666666666668,
+      "grad_norm": 364.540283203125,
+      "learning_rate": 3e-06,
+      "loss": 0.8844,
+      "step": 1884
+    },
+    {
+      "completion_length": 236.375,
+      "epoch": 0.16755555555555557,
+      "grad_norm": 518.0326538085938,
+      "learning_rate": 3e-06,
+      "loss": 7.3195,
+      "reward": 1.5208333730697632,
+      "reward_std": 0.3332235887646675,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.8333333283662796,
+      "step": 1885,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.16764444444444446,
+      "grad_norm": 477.8092346191406,
+      "learning_rate": 3e-06,
+      "loss": 9.9487,
+      "step": 1886
+    },
+    {
+      "epoch": 0.16773333333333335,
+      "grad_norm": 573.36083984375,
+      "learning_rate": 3e-06,
+      "loss": -5.14,
+      "step": 1887
+    },
+    {
+      "epoch": 0.16782222222222223,
+      "grad_norm": 552.39599609375,
+      "learning_rate": 3e-06,
+      "loss": 8.7078,
+      "step": 1888
+    },
+    {
+      "epoch": 0.16791111111111112,
+      "grad_norm": 585.54296875,
+      "learning_rate": 3e-06,
+      "loss": 1.6238,
+      "step": 1889
+    },
+    {
+      "epoch": 0.168,
+      "grad_norm": 431.30364990234375,
+      "learning_rate": 3e-06,
+      "loss": 2.3952,
+      "step": 1890
+    },
+    {
+      "epoch": 0.1680888888888889,
+      "grad_norm": 878.414306640625,
+      "learning_rate": 3e-06,
+      "loss": 3.0619,
+      "step": 1891
+    },
+    {
+      "epoch": 0.1681777777777778,
+      "grad_norm": 494.9808044433594,
+      "learning_rate": 3e-06,
+      "loss": 5.923,
+      "step": 1892
+    },
+    {
+      "epoch": 0.16826666666666668,
+      "grad_norm": 434.52093505859375,
+      "learning_rate": 3e-06,
+      "loss": -13.3846,
+      "step": 1893
+    },
+    {
+      "epoch": 0.16835555555555556,
+      "grad_norm": 355.6122131347656,
+      "learning_rate": 3e-06,
+      "loss": 1.8179,
+      "step": 1894
+    },
+    {
+      "epoch": 0.16844444444444445,
+      "grad_norm": 613.9844970703125,
+      "learning_rate": 3e-06,
+      "loss": -2.4103,
+      "step": 1895
+    },
+    {
+      "epoch": 0.16853333333333334,
+      "grad_norm": 337.2001953125,
+      "learning_rate": 3e-06,
+      "loss": -8.9448,
+      "step": 1896
+    },
+    {
+      "completion_length": 253.52084350585938,
+      "epoch": 0.16862222222222223,
+      "grad_norm": 439.8132629394531,
+      "learning_rate": 3e-06,
+      "loss": -50.7881,
+      "reward": 1.291666716337204,
+      "reward_std": 0.5373477265238762,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.6666666716337204,
+      "step": 1897,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.16871111111111112,
+      "grad_norm": 414.8184509277344,
+      "learning_rate": 3e-06,
+      "loss": -54.9974,
+      "step": 1898
+    },
+    {
+      "epoch": 0.1688,
+      "grad_norm": 449.9412536621094,
+      "learning_rate": 3e-06,
+      "loss": -85.2961,
+      "step": 1899
+    },
+    {
+      "epoch": 0.1688888888888889,
+      "grad_norm": 588.9105224609375,
+      "learning_rate": 3e-06,
+      "loss": -72.8215,
+      "step": 1900
+    },
+    {
+      "epoch": 0.16897777777777778,
+      "grad_norm": 511.78790283203125,
+      "learning_rate": 3e-06,
+      "loss": -88.0678,
+      "step": 1901
+    },
+    {
+      "epoch": 0.16906666666666667,
+      "grad_norm": 514.567138671875,
+      "learning_rate": 3e-06,
+      "loss": -90.2521,
+      "step": 1902
+    },
+    {
+      "epoch": 0.16915555555555556,
+      "grad_norm": 394.65826416015625,
+      "learning_rate": 3e-06,
+      "loss": -56.8307,
+      "step": 1903
+    },
+    {
+      "epoch": 0.16924444444444445,
+      "grad_norm": 463.7818603515625,
+      "learning_rate": 3e-06,
+      "loss": -62.5249,
+      "step": 1904
+    },
+    {
+      "epoch": 0.16933333333333334,
+      "grad_norm": 579.9658813476562,
+      "learning_rate": 3e-06,
+      "loss": -90.807,
+      "step": 1905
+    },
+    {
+      "epoch": 0.16942222222222222,
+      "grad_norm": 578.1275024414062,
+      "learning_rate": 3e-06,
+      "loss": -82.3221,
+      "step": 1906
+    },
+    {
+      "epoch": 0.1695111111111111,
+      "grad_norm": 405.3597412109375,
+      "learning_rate": 3e-06,
+      "loss": -100.0349,
+      "step": 1907
+    },
+    {
+      "epoch": 0.1696,
+      "grad_norm": 545.5985107421875,
+      "learning_rate": 3e-06,
+      "loss": -101.4231,
+      "step": 1908
+    },
+    {
+      "completion_length": 233.64584350585938,
+      "epoch": 0.1696888888888889,
+      "grad_norm": 459.9978942871094,
+      "learning_rate": 3e-06,
+      "loss": -49.3001,
+      "reward": 1.7916667461395264,
+      "reward_std": 0.4701542258262634,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.0416666865348816,
+      "step": 1909,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.16977777777777778,
+      "grad_norm": 383.19635009765625,
+      "learning_rate": 3e-06,
+      "loss": -29.5879,
+      "step": 1910
+    },
+    {
+      "epoch": 0.16986666666666667,
+      "grad_norm": 451.4841003417969,
+      "learning_rate": 3e-06,
+      "loss": -53.6264,
+      "step": 1911
+    },
+    {
+      "epoch": 0.16995555555555555,
+      "grad_norm": 429.31640625,
+      "learning_rate": 3e-06,
+      "loss": -41.3605,
+      "step": 1912
+    },
+    {
+      "epoch": 0.17004444444444444,
+      "grad_norm": 476.54168701171875,
+      "learning_rate": 3e-06,
+      "loss": -33.034,
+      "step": 1913
+    },
+    {
+      "epoch": 0.17013333333333333,
+      "grad_norm": 411.1586608886719,
+      "learning_rate": 3e-06,
+      "loss": -33.1554,
+      "step": 1914
+    },
+    {
+      "epoch": 0.17022222222222222,
+      "grad_norm": 504.1165771484375,
+      "learning_rate": 3e-06,
+      "loss": -54.2275,
+      "step": 1915
+    },
+    {
+      "epoch": 0.1703111111111111,
+      "grad_norm": 430.4921569824219,
+      "learning_rate": 3e-06,
+      "loss": -36.929,
+      "step": 1916
+    },
+    {
+      "epoch": 0.1704,
+      "grad_norm": 488.67071533203125,
+      "learning_rate": 3e-06,
+      "loss": -59.2601,
+      "step": 1917
+    },
+    {
+      "epoch": 0.17048888888888888,
+      "grad_norm": 381.2705078125,
+      "learning_rate": 3e-06,
+      "loss": -46.7398,
+      "step": 1918
+    },
+    {
+      "epoch": 0.17057777777777777,
+      "grad_norm": 439.42071533203125,
+      "learning_rate": 3e-06,
+      "loss": -40.6781,
+      "step": 1919
+    },
+    {
+      "epoch": 0.17066666666666666,
+      "grad_norm": 468.06365966796875,
+      "learning_rate": 3e-06,
+      "loss": -39.7694,
+      "step": 1920
+    },
+    {
+      "completion_length": 248.89583587646484,
+      "epoch": 0.17075555555555555,
+      "grad_norm": 389.05523681640625,
+      "learning_rate": 3e-06,
+      "loss": -15.4944,
+      "reward": 1.2291667461395264,
+      "reward_std": 0.23116152733564377,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.5416666716337204,
+      "step": 1921,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.17084444444444444,
+      "grad_norm": 437.8026428222656,
+      "learning_rate": 3e-06,
+      "loss": -1.488,
+      "step": 1922
+    },
+    {
+      "epoch": 0.17093333333333333,
+      "grad_norm": 251.8334197998047,
+      "learning_rate": 3e-06,
+      "loss": -4.1204,
+      "step": 1923
+    },
+    {
+      "epoch": 0.17102222222222221,
+      "grad_norm": 547.3317260742188,
+      "learning_rate": 3e-06,
+      "loss": -28.7346,
+      "step": 1924
+    },
+    {
+      "epoch": 0.1711111111111111,
+      "grad_norm": 289.30096435546875,
+      "learning_rate": 3e-06,
+      "loss": -16.5796,
+      "step": 1925
+    },
+    {
+      "epoch": 0.1712,
+      "grad_norm": 335.68682861328125,
+      "learning_rate": 3e-06,
+      "loss": -26.3341,
+      "step": 1926
+    },
+    {
+      "epoch": 0.17128888888888888,
+      "grad_norm": 396.2536926269531,
+      "learning_rate": 3e-06,
+      "loss": -18.4726,
+      "step": 1927
+    },
+    {
+      "epoch": 0.17137777777777777,
+      "grad_norm": 490.247802734375,
+      "learning_rate": 3e-06,
+      "loss": -8.9601,
+      "step": 1928
+    },
+    {
+      "epoch": 0.17146666666666666,
+      "grad_norm": 386.6977844238281,
+      "learning_rate": 3e-06,
+      "loss": -11.4419,
+      "step": 1929
+    },
+    {
+      "epoch": 0.17155555555555554,
+      "grad_norm": 547.1292114257812,
+      "learning_rate": 3e-06,
+      "loss": -34.8083,
+      "step": 1930
+    },
+    {
+      "epoch": 0.17164444444444443,
+      "grad_norm": 253.44793701171875,
+      "learning_rate": 3e-06,
+      "loss": -20.5172,
+      "step": 1931
+    },
+    {
+      "epoch": 0.17173333333333332,
+      "grad_norm": 536.4385375976562,
+      "learning_rate": 3e-06,
+      "loss": -35.105,
+      "step": 1932
+    },
+    {
+      "completion_length": 254.0416717529297,
+      "epoch": 0.1718222222222222,
+      "grad_norm": 513.8276977539062,
+      "learning_rate": 3e-06,
+      "loss": 3.0418,
+      "reward": 1.625,
+      "reward_std": 0.39512956142425537,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.875,
+      "step": 1933,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.1719111111111111,
+      "grad_norm": 700.8008422851562,
+      "learning_rate": 3e-06,
+      "loss": -24.1725,
+      "step": 1934
+    },
+    {
+      "epoch": 0.172,
+      "grad_norm": 640.3589477539062,
+      "learning_rate": 3e-06,
+      "loss": -18.554,
+      "step": 1935
+    },
+    {
+      "epoch": 0.1720888888888889,
+      "grad_norm": 507.8769836425781,
+      "learning_rate": 3e-06,
+      "loss": -30.509,
+      "step": 1936
+    },
+    {
+      "epoch": 0.1721777777777778,
+      "grad_norm": 414.5351257324219,
+      "learning_rate": 3e-06,
+      "loss": -39.3267,
+      "step": 1937
+    },
+    {
+      "epoch": 0.17226666666666668,
+      "grad_norm": 445.7782897949219,
+      "learning_rate": 3e-06,
+      "loss": -25.2095,
+      "step": 1938
+    },
+    {
+      "epoch": 0.17235555555555557,
+      "grad_norm": 665.8450927734375,
+      "learning_rate": 3e-06,
+      "loss": -6.8353,
+      "step": 1939
+    },
+    {
+      "epoch": 0.17244444444444446,
+      "grad_norm": 558.6971435546875,
+      "learning_rate": 3e-06,
+      "loss": -33.1913,
+      "step": 1940
+    },
+    {
+      "epoch": 0.17253333333333334,
+      "grad_norm": 500.55841064453125,
+      "learning_rate": 3e-06,
+      "loss": -30.2144,
+      "step": 1941
+    },
+    {
+      "epoch": 0.17262222222222223,
+      "grad_norm": 530.8914184570312,
+      "learning_rate": 3e-06,
+      "loss": -38.4873,
+      "step": 1942
+    },
+    {
+      "epoch": 0.17271111111111112,
+      "grad_norm": 427.3404846191406,
+      "learning_rate": 3e-06,
+      "loss": -45.8584,
+      "step": 1943
+    },
+    {
+      "epoch": 0.1728,
+      "grad_norm": 391.1449279785156,
+      "learning_rate": 3e-06,
+      "loss": -29.2224,
+      "step": 1944
+    },
+    {
+      "completion_length": 255.1041717529297,
+      "epoch": 0.1728888888888889,
+      "grad_norm": 763.403076171875,
+      "learning_rate": 3e-06,
+      "loss": -9.7223,
+      "reward": 1.8333333730697632,
+      "reward_std": 0.4971916079521179,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.0833333134651184,
+      "step": 1945,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.17297777777777779,
+      "grad_norm": 562.3991088867188,
+      "learning_rate": 3e-06,
+      "loss": -7.2214,
+      "step": 1946
+    },
+    {
+      "epoch": 0.17306666666666667,
+      "grad_norm": 686.5684814453125,
+      "learning_rate": 3e-06,
+      "loss": 2.4798,
+      "step": 1947
+    },
+    {
+      "epoch": 0.17315555555555556,
+      "grad_norm": 897.48046875,
+      "learning_rate": 3e-06,
+      "loss": -0.4508,
+      "step": 1948
+    },
+    {
+      "epoch": 0.17324444444444445,
+      "grad_norm": 439.28924560546875,
+      "learning_rate": 3e-06,
+      "loss": -15.6095,
+      "step": 1949
+    },
+    {
+      "epoch": 0.17333333333333334,
+      "grad_norm": 501.46044921875,
+      "learning_rate": 3e-06,
+      "loss": -7.5293,
+      "step": 1950
+    },
+    {
+      "epoch": 0.17342222222222223,
+      "grad_norm": 659.4769287109375,
+      "learning_rate": 3e-06,
+      "loss": -14.7485,
+      "step": 1951
+    },
+    {
+      "epoch": 0.17351111111111112,
+      "grad_norm": 528.5435180664062,
+      "learning_rate": 3e-06,
+      "loss": -14.1578,
+      "step": 1952
+    },
+    {
+      "epoch": 0.1736,
+      "grad_norm": 827.1624145507812,
+      "learning_rate": 3e-06,
+      "loss": -3.8977,
+      "step": 1953
+    },
+    {
+      "epoch": 0.1736888888888889,
+      "grad_norm": 593.5472412109375,
+      "learning_rate": 3e-06,
+      "loss": -4.4025,
+      "step": 1954
+    },
+    {
+      "epoch": 0.17377777777777778,
+      "grad_norm": 444.0652160644531,
+      "learning_rate": 3e-06,
+      "loss": -18.2012,
+      "step": 1955
+    },
+    {
+      "epoch": 0.17386666666666667,
+      "grad_norm": 478.5816345214844,
+      "learning_rate": 3e-06,
+      "loss": -13.0243,
+      "step": 1956
+    },
+    {
+      "completion_length": 253.25000762939453,
+      "epoch": 0.17395555555555556,
+      "grad_norm": 375.89666748046875,
+      "learning_rate": 3e-06,
+      "loss": -12.1653,
+      "reward": 1.2291666865348816,
+      "reward_std": 0.10206207633018494,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.5416666567325592,
+      "step": 1957,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.17404444444444445,
+      "grad_norm": 902.545654296875,
+      "learning_rate": 3e-06,
+      "loss": -25.8535,
+      "step": 1958
+    },
+    {
+      "epoch": 0.17413333333333333,
+      "grad_norm": 629.4489135742188,
+      "learning_rate": 3e-06,
+      "loss": -22.1681,
+      "step": 1959
+    },
+    {
+      "epoch": 0.17422222222222222,
+      "grad_norm": 359.41644287109375,
+      "learning_rate": 3e-06,
+      "loss": -19.6096,
+      "step": 1960
+    },
+    {
+      "epoch": 0.1743111111111111,
+      "grad_norm": 356.7577819824219,
+      "learning_rate": 3e-06,
+      "loss": -12.0564,
+      "step": 1961
+    },
+    {
+      "epoch": 0.1744,
+      "grad_norm": 236.30433654785156,
+      "learning_rate": 3e-06,
+      "loss": -25.8836,
+      "step": 1962
+    },
+    {
+      "epoch": 0.1744888888888889,
+      "grad_norm": 340.1228332519531,
+      "learning_rate": 3e-06,
+      "loss": -14.5898,
+      "step": 1963
+    },
+    {
+      "epoch": 0.17457777777777778,
+      "grad_norm": 186.7640838623047,
+      "learning_rate": 3e-06,
+      "loss": -29.7451,
+      "step": 1964
+    },
+    {
+      "epoch": 0.17466666666666666,
+      "grad_norm": 411.6400451660156,
+      "learning_rate": 3e-06,
+      "loss": -20.199,
+      "step": 1965
+    },
+    {
+      "epoch": 0.17475555555555555,
+      "grad_norm": 319.7075500488281,
+      "learning_rate": 3e-06,
+      "loss": -22.8842,
+      "step": 1966
+    },
+    {
+      "epoch": 0.17484444444444444,
+      "grad_norm": 417.3407287597656,
+      "learning_rate": 3e-06,
+      "loss": -16.5058,
+      "step": 1967
+    },
+    {
+      "epoch": 0.17493333333333333,
+      "grad_norm": 243.57456970214844,
+      "learning_rate": 3e-06,
+      "loss": -31.0802,
+      "step": 1968
+    },
+    {
+      "completion_length": 249.93750762939453,
+      "epoch": 0.17502222222222222,
+      "grad_norm": 276.6730041503906,
+      "learning_rate": 3e-06,
+      "loss": -29.1006,
+      "reward": 1.3958333730697632,
+      "reward_std": 0.10206206887960434,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.7083333432674408,
+      "step": 1969,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.1751111111111111,
+      "grad_norm": 699.74658203125,
+      "learning_rate": 3e-06,
+      "loss": -14.0976,
+      "step": 1970
+    },
+    {
+      "epoch": 0.1752,
+      "grad_norm": 471.5838623046875,
+      "learning_rate": 3e-06,
+      "loss": -20.9071,
+      "step": 1971
+    },
+    {
+      "epoch": 0.17528888888888888,
+      "grad_norm": 472.9868469238281,
+      "learning_rate": 3e-06,
+      "loss": -25.4522,
+      "step": 1972
+    },
+    {
+      "epoch": 0.17537777777777777,
+      "grad_norm": 254.69761657714844,
+      "learning_rate": 3e-06,
+      "loss": -29.7094,
+      "step": 1973
+    },
+    {
+      "epoch": 0.17546666666666666,
+      "grad_norm": 248.12869262695312,
+      "learning_rate": 3e-06,
+      "loss": -28.2889,
+      "step": 1974
+    },
+    {
+      "epoch": 0.17555555555555555,
+      "grad_norm": 236.1974639892578,
+      "learning_rate": 3e-06,
+      "loss": -32.1523,
+      "step": 1975
+    },
+    {
+      "epoch": 0.17564444444444444,
+      "grad_norm": 648.0961303710938,
+      "learning_rate": 3e-06,
+      "loss": -16.4427,
+      "step": 1976
+    },
+    {
+      "epoch": 0.17573333333333332,
+      "grad_norm": 378.50634765625,
+      "learning_rate": 3e-06,
+      "loss": -27.5418,
+      "step": 1977
+    },
+    {
+      "epoch": 0.1758222222222222,
+      "grad_norm": 439.670654296875,
+      "learning_rate": 3e-06,
+      "loss": -30.8163,
+      "step": 1978
+    },
+    {
+      "epoch": 0.1759111111111111,
+      "grad_norm": 216.43161010742188,
+      "learning_rate": 3e-06,
+      "loss": -34.2176,
+      "step": 1979
+    },
+    {
+      "epoch": 0.176,
+      "grad_norm": 242.0641632080078,
+      "learning_rate": 3e-06,
+      "loss": -31.2334,
+      "step": 1980
+    },
+    {
+      "completion_length": 248.2916717529297,
+      "epoch": 0.17608888888888888,
+      "grad_norm": 431.9725646972656,
+      "learning_rate": 3e-06,
+      "loss": -90.337,
+      "reward": 1.0208333730697632,
+      "reward_std": 0.37592336535453796,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5625,
+      "rewards/correctness_reward_func_math": 0.4583333432674408,
+      "step": 1981,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.17617777777777777,
+      "grad_norm": 421.96124267578125,
+      "learning_rate": 3e-06,
+      "loss": -74.4792,
+      "step": 1982
+    },
+    {
+      "epoch": 0.17626666666666665,
+      "grad_norm": 376.8736877441406,
+      "learning_rate": 3e-06,
+      "loss": -66.9567,
+      "step": 1983
+    },
+    {
+      "epoch": 0.17635555555555554,
+      "grad_norm": 489.1451110839844,
+      "learning_rate": 3e-06,
+      "loss": -61.4308,
+      "step": 1984
+    },
+    {
+      "epoch": 0.17644444444444443,
+      "grad_norm": 597.60498046875,
+      "learning_rate": 3e-06,
+      "loss": -58.1928,
+      "step": 1985
+    },
+    {
+      "epoch": 0.17653333333333332,
+      "grad_norm": 466.6503601074219,
+      "learning_rate": 3e-06,
+      "loss": -63.9999,
+      "step": 1986
+    },
+    {
+      "epoch": 0.17662222222222224,
+      "grad_norm": 420.9534606933594,
+      "learning_rate": 3e-06,
+      "loss": -98.6249,
+      "step": 1987
+    },
+    {
+      "epoch": 0.17671111111111112,
+      "grad_norm": 428.0660400390625,
+      "learning_rate": 3e-06,
+      "loss": -80.7811,
+      "step": 1988
+    },
+    {
+      "epoch": 0.1768,
+      "grad_norm": 381.4080505371094,
+      "learning_rate": 3e-06,
+      "loss": -75.1067,
+      "step": 1989
+    },
+    {
+      "epoch": 0.1768888888888889,
+      "grad_norm": 451.9751892089844,
+      "learning_rate": 3e-06,
+      "loss": -66.6592,
+      "step": 1990
+    },
+    {
+      "epoch": 0.1769777777777778,
+      "grad_norm": 526.015625,
+      "learning_rate": 3e-06,
+      "loss": -69.5291,
+      "step": 1991
+    },
+    {
+      "epoch": 0.17706666666666668,
+      "grad_norm": 532.5322265625,
+      "learning_rate": 3e-06,
+      "loss": -72.9259,
+      "step": 1992
+    },
+    {
+      "completion_length": 254.4166717529297,
+      "epoch": 0.17715555555555557,
+      "grad_norm": 836.5994873046875,
+      "learning_rate": 3e-06,
+      "loss": -56.0218,
+      "reward": 1.0625000298023224,
+      "reward_std": 0.6184598803520203,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.3749999925494194,
+      "step": 1993,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.17724444444444445,
+      "grad_norm": 828.1771240234375,
+      "learning_rate": 3e-06,
+      "loss": 7.2773,
+      "step": 1994
+    },
+    {
+      "epoch": 0.17733333333333334,
+      "grad_norm": 832.0421752929688,
+      "learning_rate": 3e-06,
+      "loss": -26.279,
+      "step": 1995
+    },
+    {
+      "epoch": 0.17742222222222223,
+      "grad_norm": 842.8486938476562,
+      "learning_rate": 3e-06,
+      "loss": -26.3621,
+      "step": 1996
+    },
+    {
+      "epoch": 0.17751111111111112,
+      "grad_norm": 641.175048828125,
+      "learning_rate": 3e-06,
+      "loss": -35.8716,
+      "step": 1997
+    },
+    {
+      "epoch": 0.1776,
+      "grad_norm": 621.6319580078125,
+      "learning_rate": 3e-06,
+      "loss": -46.1934,
+      "step": 1998
+    },
+    {
+      "epoch": 0.1776888888888889,
+      "grad_norm": 908.200927734375,
+      "learning_rate": 3e-06,
+      "loss": -65.1173,
+      "step": 1999
+    },
+    {
+      "epoch": 0.17777777777777778,
+      "grad_norm": 875.9900512695312,
+      "learning_rate": 3e-06,
+      "loss": 0.5691,
+      "step": 2000
+    },
+    {
+      "epoch": 0.17786666666666667,
+      "grad_norm": 801.6747436523438,
+      "learning_rate": 3e-06,
+      "loss": -35.2798,
+      "step": 2001
+    },
+    {
+      "epoch": 0.17795555555555556,
+      "grad_norm": 942.287353515625,
+      "learning_rate": 3e-06,
+      "loss": -34.8283,
+      "step": 2002
+    },
+    {
+      "epoch": 0.17804444444444445,
+      "grad_norm": 616.319091796875,
+      "learning_rate": 3e-06,
+      "loss": -39.5028,
+      "step": 2003
+    },
+    {
+      "epoch": 0.17813333333333334,
+      "grad_norm": 901.8914184570312,
+      "learning_rate": 3e-06,
+      "loss": -52.1208,
+      "step": 2004
+    },
+    {
+      "completion_length": 248.33333587646484,
+      "epoch": 0.17822222222222223,
+      "grad_norm": 706.6128540039062,
+      "learning_rate": 3e-06,
+      "loss": -29.2574,
+      "reward": 1.5416666865348816,
+      "reward_std": 0.6341222822666168,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.7916666865348816,
+      "step": 2005,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.17831111111111111,
+      "grad_norm": 781.2979736328125,
+      "learning_rate": 3e-06,
+      "loss": -8.4374,
+      "step": 2006
+    },
+    {
+      "epoch": 0.1784,
+      "grad_norm": 799.8029174804688,
+      "learning_rate": 3e-06,
+      "loss": -62.7724,
+      "step": 2007
+    },
+    {
+      "epoch": 0.1784888888888889,
+      "grad_norm": 783.2274780273438,
+      "learning_rate": 3e-06,
+      "loss": -32.8635,
+      "step": 2008
+    },
+    {
+      "epoch": 0.17857777777777778,
+      "grad_norm": 657.0900268554688,
+      "learning_rate": 3e-06,
+      "loss": -6.8832,
+      "step": 2009
+    },
+    {
+      "epoch": 0.17866666666666667,
+      "grad_norm": 736.504638671875,
+      "learning_rate": 3e-06,
+      "loss": -34.957,
+      "step": 2010
+    },
+    {
+      "epoch": 0.17875555555555556,
+      "grad_norm": 723.53173828125,
+      "learning_rate": 3e-06,
+      "loss": -38.0641,
+      "step": 2011
+    },
+    {
+      "epoch": 0.17884444444444444,
+      "grad_norm": 789.1529541015625,
+      "learning_rate": 3e-06,
+      "loss": -14.8114,
+      "step": 2012
+    },
+    {
+      "epoch": 0.17893333333333333,
+      "grad_norm": 596.5152587890625,
+      "learning_rate": 3e-06,
+      "loss": -74.1531,
+      "step": 2013
+    },
+    {
+      "epoch": 0.17902222222222222,
+      "grad_norm": 761.294189453125,
+      "learning_rate": 3e-06,
+      "loss": -44.1236,
+      "step": 2014
+    },
+    {
+      "epoch": 0.1791111111111111,
+      "grad_norm": 669.9348754882812,
+      "learning_rate": 3e-06,
+      "loss": -13.3884,
+      "step": 2015
+    },
+    {
+      "epoch": 0.1792,
+      "grad_norm": 624.5435180664062,
+      "learning_rate": 3e-06,
+      "loss": -43.6932,
+      "step": 2016
+    },
+    {
+      "completion_length": 248.64584350585938,
+      "epoch": 0.1792888888888889,
+      "grad_norm": 277.96478271484375,
+      "learning_rate": 3e-06,
+      "loss": 27.6711,
+      "reward": 1.0625,
+      "reward_std": 0.23116151988506317,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 2017,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.17937777777777777,
+      "grad_norm": 346.2447814941406,
+      "learning_rate": 3e-06,
+      "loss": 23.3867,
+      "step": 2018
+    },
+    {
+      "epoch": 0.17946666666666666,
+      "grad_norm": 250.9003448486328,
+      "learning_rate": 3e-06,
+      "loss": 24.0388,
+      "step": 2019
+    },
+    {
+      "epoch": 0.17955555555555555,
+      "grad_norm": 308.0636291503906,
+      "learning_rate": 3e-06,
+      "loss": 22.8199,
+      "step": 2020
+    },
+    {
+      "epoch": 0.17964444444444444,
+      "grad_norm": 356.0393371582031,
+      "learning_rate": 3e-06,
+      "loss": 17.5979,
+      "step": 2021
+    },
+    {
+      "epoch": 0.17973333333333333,
+      "grad_norm": 350.8787841796875,
+      "learning_rate": 3e-06,
+      "loss": 21.2881,
+      "step": 2022
+    },
+    {
+      "epoch": 0.17982222222222222,
+      "grad_norm": 290.56292724609375,
+      "learning_rate": 3e-06,
+      "loss": 24.3076,
+      "step": 2023
+    },
+    {
+      "epoch": 0.1799111111111111,
+      "grad_norm": 291.3890075683594,
+      "learning_rate": 3e-06,
+      "loss": 19.0537,
+      "step": 2024
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 249.847412109375,
+      "learning_rate": 3e-06,
+      "loss": 20.9012,
+      "step": 2025
+    },
+    {
+      "epoch": 0.18008888888888888,
+      "grad_norm": 343.4595031738281,
+      "learning_rate": 3e-06,
+      "loss": 17.5784,
+      "step": 2026
+    },
+    {
+      "epoch": 0.18017777777777777,
+      "grad_norm": 551.070068359375,
+      "learning_rate": 3e-06,
+      "loss": 11.5671,
+      "step": 2027
+    },
+    {
+      "epoch": 0.18026666666666666,
+      "grad_norm": 353.6826171875,
+      "learning_rate": 3e-06,
+      "loss": 13.8267,
+      "step": 2028
+    },
+    {
+      "completion_length": 252.8541717529297,
+      "epoch": 0.18035555555555555,
+      "grad_norm": 691.8173828125,
+      "learning_rate": 3e-06,
+      "loss": 19.1562,
+      "reward": 1.3958333730697632,
+      "reward_std": 0.505022794008255,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.7083333134651184,
+      "step": 2029,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.18044444444444444,
+      "grad_norm": 769.9490356445312,
+      "learning_rate": 3e-06,
+      "loss": -6.2182,
+      "step": 2030
+    },
+    {
+      "epoch": 0.18053333333333332,
+      "grad_norm": 746.6094360351562,
+      "learning_rate": 3e-06,
+      "loss": 26.2743,
+      "step": 2031
+    },
+    {
+      "epoch": 0.1806222222222222,
+      "grad_norm": 538.3868408203125,
+      "learning_rate": 3e-06,
+      "loss": 41.1944,
+      "step": 2032
+    },
+    {
+      "epoch": 0.1807111111111111,
+      "grad_norm": 1304.790771484375,
+      "learning_rate": 3e-06,
+      "loss": -12.3747,
+      "step": 2033
+    },
+    {
+      "epoch": 0.1808,
+      "grad_norm": 551.8775634765625,
+      "learning_rate": 3e-06,
+      "loss": 18.321,
+      "step": 2034
+    },
+    {
+      "epoch": 0.18088888888888888,
+      "grad_norm": 639.1527709960938,
+      "learning_rate": 3e-06,
+      "loss": 9.5878,
+      "step": 2035
+    },
+    {
+      "epoch": 0.18097777777777777,
+      "grad_norm": 659.2324829101562,
+      "learning_rate": 3e-06,
+      "loss": -12.4579,
+      "step": 2036
+    },
+    {
+      "epoch": 0.18106666666666665,
+      "grad_norm": 709.0194702148438,
+      "learning_rate": 3e-06,
+      "loss": 14.4976,
+      "step": 2037
+    },
+    {
+      "epoch": 0.18115555555555554,
+      "grad_norm": 522.2162475585938,
+      "learning_rate": 3e-06,
+      "loss": 31.9036,
+      "step": 2038
+    },
+    {
+      "epoch": 0.18124444444444446,
+      "grad_norm": 605.5569458007812,
+      "learning_rate": 3e-06,
+      "loss": -16.3849,
+      "step": 2039
+    },
+    {
+      "epoch": 0.18133333333333335,
+      "grad_norm": 586.4031982421875,
+      "learning_rate": 3e-06,
+      "loss": 7.3009,
+      "step": 2040
+    },
+    {
+      "completion_length": 255.25,
+      "epoch": 0.18142222222222223,
+      "grad_norm": 294.46368408203125,
+      "learning_rate": 3e-06,
+      "loss": -46.1645,
+      "reward": 1.1666666865348816,
+      "reward_std": 0.26603010296821594,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.5416666567325592,
+      "step": 2041,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.18151111111111112,
+      "grad_norm": 281.5460510253906,
+      "learning_rate": 3e-06,
+      "loss": -45.9692,
+      "step": 2042
+    },
+    {
+      "epoch": 0.1816,
+      "grad_norm": 426.70758056640625,
+      "learning_rate": 3e-06,
+      "loss": -59.0595,
+      "step": 2043
+    },
+    {
+      "epoch": 0.1816888888888889,
+      "grad_norm": 379.1589660644531,
+      "learning_rate": 3e-06,
+      "loss": -48.9399,
+      "step": 2044
+    },
+    {
+      "epoch": 0.1817777777777778,
+      "grad_norm": 372.3189697265625,
+      "learning_rate": 3e-06,
+      "loss": -59.5334,
+      "step": 2045
+    },
+    {
+      "epoch": 0.18186666666666668,
+      "grad_norm": 316.384765625,
+      "learning_rate": 3e-06,
+      "loss": -42.7701,
+      "step": 2046
+    },
+    {
+      "epoch": 0.18195555555555556,
+      "grad_norm": 333.133056640625,
+      "learning_rate": 3e-06,
+      "loss": -52.4876,
+      "step": 2047
+    },
+    {
+      "epoch": 0.18204444444444445,
+      "grad_norm": 302.69488525390625,
+      "learning_rate": 3e-06,
+      "loss": -50.558,
+      "step": 2048
+    },
+    {
+      "epoch": 0.18213333333333334,
+      "grad_norm": 297.52264404296875,
+      "learning_rate": 3e-06,
+      "loss": -65.7042,
+      "step": 2049
+    },
+    {
+      "epoch": 0.18222222222222223,
+      "grad_norm": 390.93719482421875,
+      "learning_rate": 3e-06,
+      "loss": -56.6311,
+      "step": 2050
+    },
+    {
+      "epoch": 0.18231111111111112,
+      "grad_norm": 510.16064453125,
+      "learning_rate": 3e-06,
+      "loss": -66.4786,
+      "step": 2051
+    },
+    {
+      "epoch": 0.1824,
+      "grad_norm": 386.4976501464844,
+      "learning_rate": 3e-06,
+      "loss": -55.7261,
+      "step": 2052
+    },
+    {
+      "completion_length": 231.77083587646484,
+      "epoch": 0.1824888888888889,
+      "grad_norm": 214.66683959960938,
+      "learning_rate": 3e-06,
+      "loss": -13.3666,
+      "reward": 1.0208333730697632,
+      "reward_std": 0.10206207633018494,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5625,
+      "rewards/correctness_reward_func_math": 0.4583333283662796,
+      "step": 2053,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.18257777777777778,
+      "grad_norm": 241.9335174560547,
+      "learning_rate": 3e-06,
+      "loss": -16.4601,
+      "step": 2054
+    },
+    {
+      "epoch": 0.18266666666666667,
+      "grad_norm": 227.05535888671875,
+      "learning_rate": 3e-06,
+      "loss": -15.4358,
+      "step": 2055
+    },
+    {
+      "epoch": 0.18275555555555556,
+      "grad_norm": 280.6861267089844,
+      "learning_rate": 3e-06,
+      "loss": -19.7707,
+      "step": 2056
+    },
+    {
+      "epoch": 0.18284444444444445,
+      "grad_norm": 211.3414306640625,
+      "learning_rate": 3e-06,
+      "loss": -11.4767,
+      "step": 2057
+    },
+    {
+      "epoch": 0.18293333333333334,
+      "grad_norm": 332.11248779296875,
+      "learning_rate": 3e-06,
+      "loss": -11.748,
+      "step": 2058
+    },
+    {
+      "epoch": 0.18302222222222223,
+      "grad_norm": 210.33470153808594,
+      "learning_rate": 3e-06,
+      "loss": -16.0377,
+      "step": 2059
+    },
+    {
+      "epoch": 0.1831111111111111,
+      "grad_norm": 230.14593505859375,
+      "learning_rate": 3e-06,
+      "loss": -19.6348,
+      "step": 2060
+    },
+    {
+      "epoch": 0.1832,
+      "grad_norm": 215.13331604003906,
+      "learning_rate": 3e-06,
+      "loss": -19.3665,
+      "step": 2061
+    },
+    {
+      "epoch": 0.1832888888888889,
+      "grad_norm": 401.6134338378906,
+      "learning_rate": 3e-06,
+      "loss": -22.9875,
+      "step": 2062
+    },
+    {
+      "epoch": 0.18337777777777778,
+      "grad_norm": 223.34193420410156,
+      "learning_rate": 3e-06,
+      "loss": -16.2301,
+      "step": 2063
+    },
+    {
+      "epoch": 0.18346666666666667,
+      "grad_norm": 241.48159790039062,
+      "learning_rate": 3e-06,
+      "loss": -17.0173,
+      "step": 2064
+    },
+    {
+      "completion_length": 250.12500762939453,
+      "epoch": 0.18355555555555556,
+      "grad_norm": 783.0624389648438,
+      "learning_rate": 3e-06,
+      "loss": -7.0804,
+      "reward": 1.2500000596046448,
+      "reward_std": 0.3332235962152481,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.5000000111758709,
+      "step": 2065,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.18364444444444444,
+      "grad_norm": 456.5368957519531,
+      "learning_rate": 3e-06,
+      "loss": -5.7073,
+      "step": 2066
+    },
+    {
+      "epoch": 0.18373333333333333,
+      "grad_norm": 801.0426635742188,
+      "learning_rate": 3e-06,
+      "loss": 24.9016,
+      "step": 2067
+    },
+    {
+      "epoch": 0.18382222222222222,
+      "grad_norm": 452.8072204589844,
+      "learning_rate": 3e-06,
+      "loss": -16.3536,
+      "step": 2068
+    },
+    {
+      "epoch": 0.1839111111111111,
+      "grad_norm": 573.7277221679688,
+      "learning_rate": 3e-06,
+      "loss": -14.2969,
+      "step": 2069
+    },
+    {
+      "epoch": 0.184,
+      "grad_norm": 490.78375244140625,
+      "learning_rate": 3e-06,
+      "loss": -29.7209,
+      "step": 2070
+    },
+    {
+      "epoch": 0.18408888888888889,
+      "grad_norm": 967.42578125,
+      "learning_rate": 3e-06,
+      "loss": -11.7546,
+      "step": 2071
+    },
+    {
+      "epoch": 0.18417777777777777,
+      "grad_norm": 446.62945556640625,
+      "learning_rate": 3e-06,
+      "loss": -11.8012,
+      "step": 2072
+    },
+    {
+      "epoch": 0.18426666666666666,
+      "grad_norm": 586.0079345703125,
+      "learning_rate": 3e-06,
+      "loss": 16.1318,
+      "step": 2073
+    },
+    {
+      "epoch": 0.18435555555555555,
+      "grad_norm": 483.7483825683594,
+      "learning_rate": 3e-06,
+      "loss": -22.5055,
+      "step": 2074
+    },
+    {
+      "epoch": 0.18444444444444444,
+      "grad_norm": 552.7943725585938,
+      "learning_rate": 3e-06,
+      "loss": -23.8956,
+      "step": 2075
+    },
+    {
+      "epoch": 0.18453333333333333,
+      "grad_norm": 516.2462768554688,
+      "learning_rate": 3e-06,
+      "loss": -36.5617,
+      "step": 2076
+    },
+    {
+      "completion_length": 245.45833587646484,
+      "epoch": 0.18462222222222222,
+      "grad_norm": 526.9989013671875,
+      "learning_rate": 3e-06,
+      "loss": -1.3188,
+      "reward": 1.8541667461395264,
+      "reward_std": 0.3872983753681183,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 1.1666666269302368,
+      "step": 2077,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.1847111111111111,
+      "grad_norm": 695.4895629882812,
+      "learning_rate": 3e-06,
+      "loss": -7.7195,
+      "step": 2078
+    },
+    {
+      "epoch": 0.1848,
+      "grad_norm": 732.2682495117188,
+      "learning_rate": 3e-06,
+      "loss": -3.051,
+      "step": 2079
+    },
+    {
+      "epoch": 0.18488888888888888,
+      "grad_norm": 717.4251098632812,
+      "learning_rate": 3e-06,
+      "loss": 6.4156,
+      "step": 2080
+    },
+    {
+      "epoch": 0.18497777777777777,
+      "grad_norm": 588.271484375,
+      "learning_rate": 3e-06,
+      "loss": -1.8704,
+      "step": 2081
+    },
+    {
+      "epoch": 0.18506666666666666,
+      "grad_norm": 563.7857055664062,
+      "learning_rate": 3e-06,
+      "loss": 5.8266,
+      "step": 2082
+    },
+    {
+      "epoch": 0.18515555555555555,
+      "grad_norm": 523.8809814453125,
+      "learning_rate": 3e-06,
+      "loss": -6.2269,
+      "step": 2083
+    },
+    {
+      "epoch": 0.18524444444444443,
+      "grad_norm": 566.2587890625,
+      "learning_rate": 3e-06,
+      "loss": -11.1423,
+      "step": 2084
+    },
+    {
+      "epoch": 0.18533333333333332,
+      "grad_norm": 575.639892578125,
+      "learning_rate": 3e-06,
+      "loss": -8.9986,
+      "step": 2085
+    },
+    {
+      "epoch": 0.1854222222222222,
+      "grad_norm": 607.2640380859375,
+      "learning_rate": 3e-06,
+      "loss": 0.5698,
+      "step": 2086
+    },
+    {
+      "epoch": 0.1855111111111111,
+      "grad_norm": 685.4660034179688,
+      "learning_rate": 3e-06,
+      "loss": -13.8418,
+      "step": 2087
+    },
+    {
+      "epoch": 0.1856,
+      "grad_norm": 1059.2657470703125,
+      "learning_rate": 3e-06,
+      "loss": -3.093,
+      "step": 2088
+    },
+    {
+      "completion_length": 243.375,
+      "epoch": 0.18568888888888888,
+      "grad_norm": 533.3504638671875,
+      "learning_rate": 3e-06,
+      "loss": -2.4458,
+      "reward": 1.7916667461395264,
+      "reward_std": 0.43528565764427185,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.0416666567325592,
+      "step": 2089,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.18577777777777776,
+      "grad_norm": 466.76678466796875,
+      "learning_rate": 3e-06,
+      "loss": -25.5447,
+      "step": 2090
+    },
+    {
+      "epoch": 0.18586666666666668,
+      "grad_norm": 482.4854736328125,
+      "learning_rate": 3e-06,
+      "loss": -11.9751,
+      "step": 2091
+    },
+    {
+      "epoch": 0.18595555555555557,
+      "grad_norm": 617.836669921875,
+      "learning_rate": 3e-06,
+      "loss": -16.149,
+      "step": 2092
+    },
+    {
+      "epoch": 0.18604444444444446,
+      "grad_norm": 806.7719116210938,
+      "learning_rate": 3e-06,
+      "loss": -12.1531,
+      "step": 2093
+    },
+    {
+      "epoch": 0.18613333333333335,
+      "grad_norm": 436.9642333984375,
+      "learning_rate": 3e-06,
+      "loss": -20.6401,
+      "step": 2094
+    },
+    {
+      "epoch": 0.18622222222222223,
+      "grad_norm": 533.0576171875,
+      "learning_rate": 3e-06,
+      "loss": -6.39,
+      "step": 2095
+    },
+    {
+      "epoch": 0.18631111111111112,
+      "grad_norm": 578.9844360351562,
+      "learning_rate": 3e-06,
+      "loss": -28.3945,
+      "step": 2096
+    },
+    {
+      "epoch": 0.1864,
+      "grad_norm": 518.856201171875,
+      "learning_rate": 3e-06,
+      "loss": -18.442,
+      "step": 2097
+    },
+    {
+      "epoch": 0.1864888888888889,
+      "grad_norm": 681.1263427734375,
+      "learning_rate": 3e-06,
+      "loss": -22.9392,
+      "step": 2098
+    },
+    {
+      "epoch": 0.1865777777777778,
+      "grad_norm": 759.8504028320312,
+      "learning_rate": 3e-06,
+      "loss": -19.4906,
+      "step": 2099
+    },
+    {
+      "epoch": 0.18666666666666668,
+      "grad_norm": 497.5691833496094,
+      "learning_rate": 3e-06,
+      "loss": -25.688,
+      "step": 2100
+    },
+    {
+      "completion_length": 247.7291717529297,
+      "epoch": 0.18675555555555556,
+      "grad_norm": 766.0548706054688,
+      "learning_rate": 3e-06,
+      "loss": 9.0542,
+      "reward": 1.145833358168602,
+      "reward_std": 0.20412415266036987,
+      "rewards/boxed_and_answer_tags_format_reward": 0.5625,
+      "rewards/correctness_reward_func_math": 0.5833333544433117,
+      "step": 2101,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.18684444444444445,
+      "grad_norm": 274.5552062988281,
+      "learning_rate": 3e-06,
+      "loss": -14.4338,
+      "step": 2102
+    },
+    {
+      "epoch": 0.18693333333333334,
+      "grad_norm": 357.243896484375,
+      "learning_rate": 3e-06,
+      "loss": -3.3275,
+      "step": 2103
+    },
+    {
+      "epoch": 0.18702222222222223,
+      "grad_norm": 349.6654968261719,
+      "learning_rate": 3e-06,
+      "loss": -3.714,
+      "step": 2104
+    },
+    {
+      "epoch": 0.18711111111111112,
+      "grad_norm": 252.651611328125,
+      "learning_rate": 3e-06,
+      "loss": 4.8234,
+      "step": 2105
+    },
+    {
+      "epoch": 0.1872,
+      "grad_norm": 422.1712951660156,
+      "learning_rate": 3e-06,
+      "loss": 2.0141,
+      "step": 2106
+    },
+    {
+      "epoch": 0.1872888888888889,
+      "grad_norm": 848.9307250976562,
+      "learning_rate": 3e-06,
+      "loss": 1.1462,
+      "step": 2107
+    },
+    {
+      "epoch": 0.18737777777777778,
+      "grad_norm": 255.5582275390625,
+      "learning_rate": 3e-06,
+      "loss": -17.6895,
+      "step": 2108
+    },
+    {
+      "epoch": 0.18746666666666667,
+      "grad_norm": 307.6992492675781,
+      "learning_rate": 3e-06,
+      "loss": -9.5895,
+      "step": 2109
+    },
+    {
+      "epoch": 0.18755555555555556,
+      "grad_norm": 277.8653259277344,
+      "learning_rate": 3e-06,
+      "loss": -9.602,
+      "step": 2110
+    },
+    {
+      "epoch": 0.18764444444444445,
+      "grad_norm": 234.34913635253906,
+      "learning_rate": 3e-06,
+      "loss": 0.8899,
+      "step": 2111
+    },
+    {
+      "epoch": 0.18773333333333334,
+      "grad_norm": 355.8470764160156,
+      "learning_rate": 3e-06,
+      "loss": -4.6677,
+      "step": 2112
+    },
+    {
+      "completion_length": 250.89583587646484,
+      "epoch": 0.18782222222222222,
+      "grad_norm": 810.5846557617188,
+      "learning_rate": 3e-06,
+      "loss": 15.7377,
+      "reward": 1.125,
+      "reward_std": 0.23116151988506317,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.375,
+      "step": 2113,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.1879111111111111,
+      "grad_norm": 509.224853515625,
+      "learning_rate": 3e-06,
+      "loss": 18.5617,
+      "step": 2114
+    },
+    {
+      "epoch": 0.188,
+      "grad_norm": 530.5845336914062,
+      "learning_rate": 3e-06,
+      "loss": 7.082,
+      "step": 2115
+    },
+    {
+      "epoch": 0.1880888888888889,
+      "grad_norm": 672.1024169921875,
+      "learning_rate": 3e-06,
+      "loss": 1.3842,
+      "step": 2116
+    },
+    {
+      "epoch": 0.18817777777777778,
+      "grad_norm": 408.4438171386719,
+      "learning_rate": 3e-06,
+      "loss": 16.7967,
+      "step": 2117
+    },
+    {
+      "epoch": 0.18826666666666667,
+      "grad_norm": 569.5597534179688,
+      "learning_rate": 3e-06,
+      "loss": 15.6526,
+      "step": 2118
+    },
+    {
+      "epoch": 0.18835555555555555,
+      "grad_norm": 613.7771606445312,
+      "learning_rate": 3e-06,
+      "loss": 11.1284,
+      "step": 2119
+    },
+    {
+      "epoch": 0.18844444444444444,
+      "grad_norm": 443.5073547363281,
+      "learning_rate": 3e-06,
+      "loss": 13.1326,
+      "step": 2120
+    },
+    {
+      "epoch": 0.18853333333333333,
+      "grad_norm": 745.5543823242188,
+      "learning_rate": 3e-06,
+      "loss": 3.177,
+      "step": 2121
+    },
+    {
+      "epoch": 0.18862222222222222,
+      "grad_norm": 776.3263549804688,
+      "learning_rate": 3e-06,
+      "loss": -4.0693,
+      "step": 2122
+    },
+    {
+      "epoch": 0.1887111111111111,
+      "grad_norm": 439.9760437011719,
+      "learning_rate": 3e-06,
+      "loss": 14.0792,
+      "step": 2123
+    },
+    {
+      "epoch": 0.1888,
+      "grad_norm": 620.5515747070312,
+      "learning_rate": 3e-06,
+      "loss": 10.3075,
+      "step": 2124
+    },
+    {
+      "completion_length": 255.89583587646484,
+      "epoch": 0.18888888888888888,
+      "grad_norm": 1261.142578125,
+      "learning_rate": 3e-06,
+      "loss": -52.8242,
+      "reward": 1.5416666865348816,
+      "reward_std": 0.7283531129360199,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.7916666567325592,
+      "step": 2125,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.18897777777777777,
+      "grad_norm": 828.8499755859375,
+      "learning_rate": 3e-06,
+      "loss": -18.1417,
+      "step": 2126
+    },
+    {
+      "epoch": 0.18906666666666666,
+      "grad_norm": 921.6665649414062,
+      "learning_rate": 3e-06,
+      "loss": -28.653,
+      "step": 2127
+    },
+    {
+      "epoch": 0.18915555555555555,
+      "grad_norm": 1188.288330078125,
+      "learning_rate": 3e-06,
+      "loss": -14.6245,
+      "step": 2128
+    },
+    {
+      "epoch": 0.18924444444444444,
+      "grad_norm": 875.7568969726562,
+      "learning_rate": 3e-06,
+      "loss": -19.076,
+      "step": 2129
+    },
+    {
+      "epoch": 0.18933333333333333,
+      "grad_norm": 935.9678955078125,
+      "learning_rate": 3e-06,
+      "loss": -30.5958,
+      "step": 2130
+    },
+    {
+      "epoch": 0.18942222222222221,
+      "grad_norm": 933.6688842773438,
+      "learning_rate": 3e-06,
+      "loss": -56.0788,
+      "step": 2131
+    },
+    {
+      "epoch": 0.1895111111111111,
+      "grad_norm": 1004.11572265625,
+      "learning_rate": 3e-06,
+      "loss": -27.5338,
+      "step": 2132
+    },
+    {
+      "epoch": 0.1896,
+      "grad_norm": 805.9441528320312,
+      "learning_rate": 3e-06,
+      "loss": -38.4037,
+      "step": 2133
+    },
+    {
+      "epoch": 0.18968888888888888,
+      "grad_norm": 1125.8046875,
+      "learning_rate": 3e-06,
+      "loss": -21.7139,
+      "step": 2134
+    },
+    {
+      "epoch": 0.18977777777777777,
+      "grad_norm": 892.211181640625,
+      "learning_rate": 3e-06,
+      "loss": -29.3145,
+      "step": 2135
+    },
+    {
+      "epoch": 0.18986666666666666,
+      "grad_norm": 895.8474731445312,
+      "learning_rate": 3e-06,
+      "loss": -38.8515,
+      "step": 2136
+    },
+    {
+      "completion_length": 241.06250762939453,
+      "epoch": 0.18995555555555554,
+      "grad_norm": 587.2543334960938,
+      "learning_rate": 3e-06,
+      "loss": -30.24,
+      "reward": 1.7083333730697632,
+      "reward_std": 0.5128540322184563,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 1.0833333134651184,
+      "step": 2137,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.19004444444444443,
+      "grad_norm": 483.7323913574219,
+      "learning_rate": 3e-06,
+      "loss": -54.7314,
+      "step": 2138
+    },
+    {
+      "epoch": 0.19013333333333332,
+      "grad_norm": 695.5252075195312,
+      "learning_rate": 3e-06,
+      "loss": -37.2405,
+      "step": 2139
+    },
+    {
+      "epoch": 0.1902222222222222,
+      "grad_norm": 659.273681640625,
+      "learning_rate": 3e-06,
+      "loss": -54.6989,
+      "step": 2140
+    },
+    {
+      "epoch": 0.1903111111111111,
+      "grad_norm": 665.203857421875,
+      "learning_rate": 3e-06,
+      "loss": -49.4977,
+      "step": 2141
+    },
+    {
+      "epoch": 0.1904,
+      "grad_norm": 570.3987426757812,
+      "learning_rate": 3e-06,
+      "loss": -29.7209,
+      "step": 2142
+    },
+    {
+      "epoch": 0.1904888888888889,
+      "grad_norm": 543.2655029296875,
+      "learning_rate": 3e-06,
+      "loss": -35.8428,
+      "step": 2143
+    },
+    {
+      "epoch": 0.1905777777777778,
+      "grad_norm": 810.827880859375,
+      "learning_rate": 3e-06,
+      "loss": -59.7728,
+      "step": 2144
+    },
+    {
+      "epoch": 0.19066666666666668,
+      "grad_norm": 702.2298583984375,
+      "learning_rate": 3e-06,
+      "loss": -49.0496,
+      "step": 2145
+    },
+    {
+      "epoch": 0.19075555555555557,
+      "grad_norm": 575.1386108398438,
+      "learning_rate": 3e-06,
+      "loss": -64.258,
+      "step": 2146
+    },
+    {
+      "epoch": 0.19084444444444446,
+      "grad_norm": 646.1061401367188,
+      "learning_rate": 3e-06,
+      "loss": -60.6597,
+      "step": 2147
+    },
+    {
+      "epoch": 0.19093333333333334,
+      "grad_norm": 583.6048583984375,
+      "learning_rate": 3e-06,
+      "loss": -42.0407,
+      "step": 2148
+    },
+    {
+      "completion_length": 230.43750762939453,
+      "epoch": 0.19102222222222223,
+      "grad_norm": 553.2525024414062,
+      "learning_rate": 3e-06,
+      "loss": 31.5824,
+      "reward": 1.2083333730697632,
+      "reward_std": 0.46232303977012634,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.5833333283662796,
+      "step": 2149,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.19111111111111112,
+      "grad_norm": 672.8676147460938,
+      "learning_rate": 3e-06,
+      "loss": 26.2286,
+      "step": 2150
+    },
+    {
+      "epoch": 0.1912,
+      "grad_norm": 594.0144653320312,
+      "learning_rate": 3e-06,
+      "loss": -2.6153,
+      "step": 2151
+    },
+    {
+      "epoch": 0.1912888888888889,
+      "grad_norm": 828.8351440429688,
+      "learning_rate": 3e-06,
+      "loss": 2.5627,
+      "step": 2152
+    },
+    {
+      "epoch": 0.19137777777777779,
+      "grad_norm": 601.149658203125,
+      "learning_rate": 3e-06,
+      "loss": -1.2538,
+      "step": 2153
+    },
+    {
+      "epoch": 0.19146666666666667,
+      "grad_norm": 685.7401123046875,
+      "learning_rate": 3e-06,
+      "loss": 8.7495,
+      "step": 2154
+    },
+    {
+      "epoch": 0.19155555555555556,
+      "grad_norm": 561.976318359375,
+      "learning_rate": 3e-06,
+      "loss": 22.0952,
+      "step": 2155
+    },
+    {
+      "epoch": 0.19164444444444445,
+      "grad_norm": 583.328369140625,
+      "learning_rate": 3e-06,
+      "loss": 15.3746,
+      "step": 2156
+    },
+    {
+      "epoch": 0.19173333333333334,
+      "grad_norm": 495.55609130859375,
+      "learning_rate": 3e-06,
+      "loss": -9.5814,
+      "step": 2157
+    },
+    {
+      "epoch": 0.19182222222222223,
+      "grad_norm": 764.2197265625,
+      "learning_rate": 3e-06,
+      "loss": -2.7963,
+      "step": 2158
+    },
+    {
+      "epoch": 0.19191111111111112,
+      "grad_norm": 2350.23779296875,
+      "learning_rate": 3e-06,
+      "loss": -8.0366,
+      "step": 2159
+    },
+    {
+      "epoch": 0.192,
+      "grad_norm": 904.4937744140625,
+      "learning_rate": 3e-06,
+      "loss": -6.0055,
+      "step": 2160
+    },
+    {
+      "completion_length": 247.93750762939453,
+      "epoch": 0.1920888888888889,
+      "grad_norm": 899.4290771484375,
+      "learning_rate": 3e-06,
+      "loss": 51.9844,
+      "reward": 1.5625000596046448,
+      "reward_std": 0.5050228163599968,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.8749999701976776,
+      "step": 2161,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.19217777777777778,
+      "grad_norm": 936.0744018554688,
+      "learning_rate": 3e-06,
+      "loss": 24.5592,
+      "step": 2162
+    },
+    {
+      "epoch": 0.19226666666666667,
+      "grad_norm": 865.0958251953125,
+      "learning_rate": 3e-06,
+      "loss": 4.2507,
+      "step": 2163
+    },
+    {
+      "epoch": 0.19235555555555556,
+      "grad_norm": 819.485107421875,
+      "learning_rate": 3e-06,
+      "loss": -17.4293,
+      "step": 2164
+    },
+    {
+      "epoch": 0.19244444444444445,
+      "grad_norm": 823.7230224609375,
+      "learning_rate": 3e-06,
+      "loss": -19.1261,
+      "step": 2165
+    },
+    {
+      "epoch": 0.19253333333333333,
+      "grad_norm": 1002.2861328125,
+      "learning_rate": 3e-06,
+      "loss": -19.0886,
+      "step": 2166
+    },
+    {
+      "epoch": 0.19262222222222222,
+      "grad_norm": 926.2667236328125,
+      "learning_rate": 3e-06,
+      "loss": 38.8715,
+      "step": 2167
+    },
+    {
+      "epoch": 0.1927111111111111,
+      "grad_norm": 890.3988037109375,
+      "learning_rate": 3e-06,
+      "loss": 16.1366,
+      "step": 2168
+    },
+    {
+      "epoch": 0.1928,
+      "grad_norm": 863.7974853515625,
+      "learning_rate": 3e-06,
+      "loss": -6.0243,
+      "step": 2169
+    },
+    {
+      "epoch": 0.1928888888888889,
+      "grad_norm": 675.4772338867188,
+      "learning_rate": 3e-06,
+      "loss": -29.941,
+      "step": 2170
+    },
+    {
+      "epoch": 0.19297777777777778,
+      "grad_norm": 795.9426879882812,
+      "learning_rate": 3e-06,
+      "loss": -37.6626,
+      "step": 2171
+    },
+    {
+      "epoch": 0.19306666666666666,
+      "grad_norm": 1173.4658203125,
+      "learning_rate": 3e-06,
+      "loss": -42.7799,
+      "step": 2172
+    },
+    {
+      "completion_length": 255.27083587646484,
+      "epoch": 0.19315555555555555,
+      "grad_norm": 1052.441650390625,
+      "learning_rate": 3e-06,
+      "loss": 17.4183,
+      "reward": 1.75,
+      "reward_std": 0.720521941781044,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.0,
+      "step": 2173,
+      "zero_std_ratio": 0.25
+    },
+    {
+      "epoch": 0.19324444444444444,
+      "grad_norm": 1221.40673828125,
+      "learning_rate": 3e-06,
+      "loss": 10.0883,
+      "step": 2174
+    },
+    {
+      "epoch": 0.19333333333333333,
+      "grad_norm": 733.4461059570312,
+      "learning_rate": 3e-06,
+      "loss": 7.5639,
+      "step": 2175
+    },
+    {
+      "epoch": 0.19342222222222222,
+      "grad_norm": 981.289794921875,
+      "learning_rate": 3e-06,
+      "loss": 32.1803,
+      "step": 2176
+    },
+    {
+      "epoch": 0.1935111111111111,
+      "grad_norm": 1169.4273681640625,
+      "learning_rate": 3e-06,
+      "loss": 38.9057,
+      "step": 2177
+    },
+    {
+      "epoch": 0.1936,
+      "grad_norm": 1102.001220703125,
+      "learning_rate": 3e-06,
+      "loss": 11.3484,
+      "step": 2178
+    },
+    {
+      "epoch": 0.19368888888888888,
+      "grad_norm": 992.0403442382812,
+      "learning_rate": 3e-06,
+      "loss": 17.4565,
+      "step": 2179
+    },
+    {
+      "epoch": 0.19377777777777777,
+      "grad_norm": 1574.7171630859375,
+      "learning_rate": 3e-06,
+      "loss": -4.7767,
+      "step": 2180
+    },
+    {
+      "epoch": 0.19386666666666666,
+      "grad_norm": 713.719482421875,
+      "learning_rate": 3e-06,
+      "loss": -2.3293,
+      "step": 2181
+    },
+    {
+      "epoch": 0.19395555555555555,
+      "grad_norm": 999.0922241210938,
+      "learning_rate": 3e-06,
+      "loss": 20.7797,
+      "step": 2182
+    },
+    {
+      "epoch": 0.19404444444444444,
+      "grad_norm": 1164.4508056640625,
+      "learning_rate": 3e-06,
+      "loss": 22.3378,
+      "step": 2183
+    },
+    {
+      "epoch": 0.19413333333333332,
+      "grad_norm": 1133.1566162109375,
+      "learning_rate": 3e-06,
+      "loss": -3.4018,
+      "step": 2184
+    },
+    {
+      "completion_length": 246.64584350585938,
+      "epoch": 0.1942222222222222,
+      "grad_norm": 668.209716796875,
+      "learning_rate": 3e-06,
+      "loss": -9.4869,
+      "reward": 1.7083333730697632,
+      "reward_std": 0.3061862215399742,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.9583333432674408,
+      "step": 2185,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.1943111111111111,
+      "grad_norm": 433.6723937988281,
+      "learning_rate": 3e-06,
+      "loss": -11.4995,
+      "step": 2186
+    },
+    {
+      "epoch": 0.1944,
+      "grad_norm": 682.7325439453125,
+      "learning_rate": 3e-06,
+      "loss": -28.6521,
+      "step": 2187
+    },
+    {
+      "epoch": 0.19448888888888888,
+      "grad_norm": 636.8472900390625,
+      "learning_rate": 3e-06,
+      "loss": -13.4065,
+      "step": 2188
+    },
+    {
+      "epoch": 0.19457777777777777,
+      "grad_norm": 447.03173828125,
+      "learning_rate": 3e-06,
+      "loss": -18.6822,
+      "step": 2189
+    },
+    {
+      "epoch": 0.19466666666666665,
+      "grad_norm": 855.8515014648438,
+      "learning_rate": 3e-06,
+      "loss": -18.9689,
+      "step": 2190
+    },
+    {
+      "epoch": 0.19475555555555554,
+      "grad_norm": 1044.5152587890625,
+      "learning_rate": 3e-06,
+      "loss": -20.745,
+      "step": 2191
+    },
+    {
+      "epoch": 0.19484444444444443,
+      "grad_norm": 422.6670837402344,
+      "learning_rate": 3e-06,
+      "loss": -16.4546,
+      "step": 2192
+    },
+    {
+      "epoch": 0.19493333333333332,
+      "grad_norm": 698.4715576171875,
+      "learning_rate": 3e-06,
+      "loss": -37.7735,
+      "step": 2193
+    },
+    {
+      "epoch": 0.19502222222222224,
+      "grad_norm": 689.1241455078125,
+      "learning_rate": 3e-06,
+      "loss": -24.1425,
+      "step": 2194
+    },
+    {
+      "epoch": 0.19511111111111112,
+      "grad_norm": 521.723876953125,
+      "learning_rate": 3e-06,
+      "loss": -25.6598,
+      "step": 2195
+    },
+    {
+      "epoch": 0.1952,
+      "grad_norm": 641.6820678710938,
+      "learning_rate": 3e-06,
+      "loss": -27.7095,
+      "step": 2196
+    },
+    {
+      "completion_length": 250.18750762939453,
+      "epoch": 0.1952888888888889,
+      "grad_norm": 695.4179077148438,
+      "learning_rate": 3e-06,
+      "loss": -23.0851,
+      "reward": 1.5625000596046448,
+      "reward_std": 0.23116153478622437,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.8750000149011612,
+      "step": 2197,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.1953777777777778,
+      "grad_norm": 606.8045654296875,
+      "learning_rate": 3e-06,
+      "loss": 1.2567,
+      "step": 2198
+    },
+    {
+      "epoch": 0.19546666666666668,
+      "grad_norm": 742.4933471679688,
+      "learning_rate": 3e-06,
+      "loss": 1.28,
+      "step": 2199
+    },
+    {
+      "epoch": 0.19555555555555557,
+      "grad_norm": 555.1710205078125,
+      "learning_rate": 3e-06,
+      "loss": -2.6916,
+      "step": 2200
+    },
+    {
+      "epoch": 0.19564444444444445,
+      "grad_norm": 658.5838012695312,
+      "learning_rate": 3e-06,
+      "loss": -3.5826,
+      "step": 2201
+    },
+    {
+      "epoch": 0.19573333333333334,
+      "grad_norm": 481.04693603515625,
+      "learning_rate": 3e-06,
+      "loss": -9.7868,
+      "step": 2202
+    },
+    {
+      "epoch": 0.19582222222222223,
+      "grad_norm": 583.1636352539062,
+      "learning_rate": 3e-06,
+      "loss": -26.7921,
+      "step": 2203
+    },
+    {
+      "epoch": 0.19591111111111112,
+      "grad_norm": 582.9187622070312,
+      "learning_rate": 3e-06,
+      "loss": -1.0781,
+      "step": 2204
+    },
+    {
+      "epoch": 0.196,
+      "grad_norm": 660.7078247070312,
+      "learning_rate": 3e-06,
+      "loss": -2.465,
+      "step": 2205
+    },
+    {
+      "epoch": 0.1960888888888889,
+      "grad_norm": 522.7738647460938,
+      "learning_rate": 3e-06,
+      "loss": -3.8803,
+      "step": 2206
+    },
+    {
+      "epoch": 0.19617777777777778,
+      "grad_norm": 1109.6839599609375,
+      "learning_rate": 3e-06,
+      "loss": -13.524,
+      "step": 2207
+    },
+    {
+      "epoch": 0.19626666666666667,
+      "grad_norm": 588.8282470703125,
+      "learning_rate": 3e-06,
+      "loss": -12.7292,
+      "step": 2208
+    },
+    {
+      "completion_length": 244.9791717529297,
+      "epoch": 0.19635555555555556,
+      "grad_norm": 403.5253601074219,
+      "learning_rate": 3e-06,
+      "loss": -24.4641,
+      "reward": 1.4895833730697632,
+      "reward_std": 0.20219219475984573,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 2209,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.19644444444444445,
+      "grad_norm": 545.7481079101562,
+      "learning_rate": 3e-06,
+      "loss": -28.5028,
+      "step": 2210
+    },
+    {
+      "epoch": 0.19653333333333334,
+      "grad_norm": 737.4202270507812,
+      "learning_rate": 3e-06,
+      "loss": -21.5566,
+      "step": 2211
+    },
+    {
+      "epoch": 0.19662222222222223,
+      "grad_norm": 415.2235412597656,
+      "learning_rate": 3e-06,
+      "loss": -19.7415,
+      "step": 2212
+    },
+    {
+      "epoch": 0.19671111111111111,
+      "grad_norm": 661.0859375,
+      "learning_rate": 3e-06,
+      "loss": -27.5604,
+      "step": 2213
+    },
+    {
+      "epoch": 0.1968,
+      "grad_norm": 688.0073852539062,
+      "learning_rate": 3e-06,
+      "loss": -25.3044,
+      "step": 2214
+    },
+    {
+      "epoch": 0.1968888888888889,
+      "grad_norm": 377.343505859375,
+      "learning_rate": 3e-06,
+      "loss": -30.7918,
+      "step": 2215
+    },
+    {
+      "epoch": 0.19697777777777778,
+      "grad_norm": 535.9216918945312,
+      "learning_rate": 3e-06,
+      "loss": -28.4559,
+      "step": 2216
+    },
+    {
+      "epoch": 0.19706666666666667,
+      "grad_norm": 631.3789672851562,
+      "learning_rate": 3e-06,
+      "loss": -28.4084,
+      "step": 2217
+    },
+    {
+      "epoch": 0.19715555555555556,
+      "grad_norm": 467.3281555175781,
+      "learning_rate": 3e-06,
+      "loss": -24.7054,
+      "step": 2218
+    },
+    {
+      "epoch": 0.19724444444444444,
+      "grad_norm": 673.6080932617188,
+      "learning_rate": 3e-06,
+      "loss": -30.1266,
+      "step": 2219
+    },
+    {
+      "epoch": 0.19733333333333333,
+      "grad_norm": 650.60009765625,
+      "learning_rate": 3e-06,
+      "loss": -30.6526,
+      "step": 2220
+    },
+    {
+      "completion_length": 251.27083587646484,
+      "epoch": 0.19742222222222222,
+      "grad_norm": 937.9761962890625,
+      "learning_rate": 3e-06,
+      "loss": -1.5455,
+      "reward": 1.9166667461395264,
+      "reward_std": 0.4779854342341423,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.1666666567325592,
+      "step": 2221,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.1975111111111111,
+      "grad_norm": 618.4127807617188,
+      "learning_rate": 3e-06,
+      "loss": 8.2676,
+      "step": 2222
+    },
+    {
+      "epoch": 0.1976,
+      "grad_norm": 681.9295654296875,
+      "learning_rate": 3e-06,
+      "loss": 5.7096,
+      "step": 2223
+    },
+    {
+      "epoch": 0.1976888888888889,
+      "grad_norm": 1006.9591064453125,
+      "learning_rate": 3e-06,
+      "loss": 52.6603,
+      "step": 2224
+    },
+    {
+      "epoch": 0.19777777777777777,
+      "grad_norm": 786.645263671875,
+      "learning_rate": 3e-06,
+      "loss": 7.371,
+      "step": 2225
+    },
+    {
+      "epoch": 0.19786666666666666,
+      "grad_norm": 779.1517333984375,
+      "learning_rate": 3e-06,
+      "loss": 15.4347,
+      "step": 2226
+    },
+    {
+      "epoch": 0.19795555555555555,
+      "grad_norm": 1029.3656005859375,
+      "learning_rate": 3e-06,
+      "loss": -12.7798,
+      "step": 2227
+    },
+    {
+      "epoch": 0.19804444444444444,
+      "grad_norm": 1035.653076171875,
+      "learning_rate": 3e-06,
+      "loss": -0.4735,
+      "step": 2228
+    },
+    {
+      "epoch": 0.19813333333333333,
+      "grad_norm": 829.162841796875,
+      "learning_rate": 3e-06,
+      "loss": -2.8614,
+      "step": 2229
+    },
+    {
+      "epoch": 0.19822222222222222,
+      "grad_norm": 873.4863891601562,
+      "learning_rate": 3e-06,
+      "loss": 30.9012,
+      "step": 2230
+    },
+    {
+      "epoch": 0.1983111111111111,
+      "grad_norm": 649.9591674804688,
+      "learning_rate": 3e-06,
+      "loss": -6.5108,
+      "step": 2231
+    },
+    {
+      "epoch": 0.1984,
+      "grad_norm": 805.36328125,
+      "learning_rate": 3e-06,
+      "loss": 2.5386,
+      "step": 2232
+    },
+    {
+      "completion_length": 246.1041717529297,
+      "epoch": 0.19848888888888888,
+      "grad_norm": 895.270263671875,
+      "learning_rate": 3e-06,
+      "loss": -57.1749,
+      "reward": 1.2500000596046448,
+      "reward_std": 0.4779854342341423,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.5000000111758709,
+      "step": 2233,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.19857777777777777,
+      "grad_norm": 765.6522827148438,
+      "learning_rate": 3e-06,
+      "loss": -48.4423,
+      "step": 2234
+    },
+    {
+      "epoch": 0.19866666666666666,
+      "grad_norm": 755.2459716796875,
+      "learning_rate": 3e-06,
+      "loss": -29.8722,
+      "step": 2235
+    },
+    {
+      "epoch": 0.19875555555555555,
+      "grad_norm": 685.5021362304688,
+      "learning_rate": 3e-06,
+      "loss": -19.4638,
+      "step": 2236
+    },
+    {
+      "epoch": 0.19884444444444443,
+      "grad_norm": 832.52880859375,
+      "learning_rate": 3e-06,
+      "loss": -41.564,
+      "step": 2237
+    },
+    {
+      "epoch": 0.19893333333333332,
+      "grad_norm": 681.293701171875,
+      "learning_rate": 3e-06,
+      "loss": -56.8457,
+      "step": 2238
+    },
+    {
+      "epoch": 0.1990222222222222,
+      "grad_norm": 733.9737548828125,
+      "learning_rate": 3e-06,
+      "loss": -71.0826,
+      "step": 2239
+    },
+    {
+      "epoch": 0.1991111111111111,
+      "grad_norm": 759.0897216796875,
+      "learning_rate": 3e-06,
+      "loss": -59.8898,
+      "step": 2240
+    },
+    {
+      "epoch": 0.1992,
+      "grad_norm": 752.1569213867188,
+      "learning_rate": 3e-06,
+      "loss": -35.6279,
+      "step": 2241
+    },
+    {
+      "epoch": 0.19928888888888888,
+      "grad_norm": 793.9288330078125,
+      "learning_rate": 3e-06,
+      "loss": -31.1288,
+      "step": 2242
+    },
+    {
+      "epoch": 0.19937777777777776,
+      "grad_norm": 875.0328979492188,
+      "learning_rate": 3e-06,
+      "loss": -54.7059,
+      "step": 2243
+    },
+    {
+      "epoch": 0.19946666666666665,
+      "grad_norm": 735.8705444335938,
+      "learning_rate": 3e-06,
+      "loss": -64.119,
+      "step": 2244
+    },
+    {
+      "completion_length": 245.1875,
+      "epoch": 0.19955555555555557,
+      "grad_norm": 691.626953125,
+      "learning_rate": 3e-06,
+      "loss": 33.6045,
+      "reward": 1.0729166865348816,
+      "reward_std": 0.41281384229660034,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.3333333432674408,
+      "step": 2245,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.19964444444444446,
+      "grad_norm": 847.1795654296875,
+      "learning_rate": 3e-06,
+      "loss": 36.038,
+      "step": 2246
+    },
+    {
+      "epoch": 0.19973333333333335,
+      "grad_norm": 502.91473388671875,
+      "learning_rate": 3e-06,
+      "loss": 22.1079,
+      "step": 2247
+    },
+    {
+      "epoch": 0.19982222222222223,
+      "grad_norm": 577.439697265625,
+      "learning_rate": 3e-06,
+      "loss": 33.3352,
+      "step": 2248
+    },
+    {
+      "epoch": 0.19991111111111112,
+      "grad_norm": 830.4883422851562,
+      "learning_rate": 3e-06,
+      "loss": 41.7401,
+      "step": 2249
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 769.1807861328125,
+      "learning_rate": 3e-06,
+      "loss": 37.7994,
+      "step": 2250
+    },
+    {
+      "epoch": 0.2000888888888889,
+      "grad_norm": 636.1270141601562,
+      "learning_rate": 3e-06,
+      "loss": 27.5784,
+      "step": 2251
+    },
+    {
+      "epoch": 0.2001777777777778,
+      "grad_norm": 761.4356079101562,
+      "learning_rate": 3e-06,
+      "loss": 27.7986,
+      "step": 2252
+    },
+    {
+      "epoch": 0.20026666666666668,
+      "grad_norm": 530.1929931640625,
+      "learning_rate": 3e-06,
+      "loss": 17.7146,
+      "step": 2253
+    },
+    {
+      "epoch": 0.20035555555555556,
+      "grad_norm": 564.9805908203125,
+      "learning_rate": 3e-06,
+      "loss": 27.1709,
+      "step": 2254
+    },
+    {
+      "epoch": 0.20044444444444445,
+      "grad_norm": 600.7700805664062,
+      "learning_rate": 3e-06,
+      "loss": 37.6539,
+      "step": 2255
+    },
+    {
+      "epoch": 0.20053333333333334,
+      "grad_norm": 710.3458862304688,
+      "learning_rate": 3e-06,
+      "loss": 30.2361,
+      "step": 2256
+    },
+    {
+      "completion_length": 249.9791717529297,
+      "epoch": 0.20062222222222223,
+      "grad_norm": 404.9451904296875,
+      "learning_rate": 3e-06,
+      "loss": -93.8459,
+      "reward": 1.1458333730697632,
+      "reward_std": 0.23116151988506317,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.4583333283662796,
+      "step": 2257,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.20071111111111112,
+      "grad_norm": 394.6553649902344,
+      "learning_rate": 3e-06,
+      "loss": -76.6418,
+      "step": 2258
+    },
+    {
+      "epoch": 0.2008,
+      "grad_norm": 366.0687255859375,
+      "learning_rate": 3e-06,
+      "loss": -73.8687,
+      "step": 2259
+    },
+    {
+      "epoch": 0.2008888888888889,
+      "grad_norm": 464.67388916015625,
+      "learning_rate": 3e-06,
+      "loss": -82.5149,
+      "step": 2260
+    },
+    {
+      "epoch": 0.20097777777777778,
+      "grad_norm": 437.32470703125,
+      "learning_rate": 3e-06,
+      "loss": -62.3202,
+      "step": 2261
+    },
+    {
+      "epoch": 0.20106666666666667,
+      "grad_norm": 884.9143676757812,
+      "learning_rate": 3e-06,
+      "loss": -76.1318,
+      "step": 2262
+    },
+    {
+      "epoch": 0.20115555555555556,
+      "grad_norm": 431.22882080078125,
+      "learning_rate": 3e-06,
+      "loss": -101.0349,
+      "step": 2263
+    },
+    {
+      "epoch": 0.20124444444444445,
+      "grad_norm": 407.47344970703125,
+      "learning_rate": 3e-06,
+      "loss": -83.3404,
+      "step": 2264
+    },
+    {
+      "epoch": 0.20133333333333334,
+      "grad_norm": 450.6275634765625,
+      "learning_rate": 3e-06,
+      "loss": -81.2565,
+      "step": 2265
+    },
+    {
+      "epoch": 0.20142222222222222,
+      "grad_norm": 446.26715087890625,
+      "learning_rate": 3e-06,
+      "loss": -92.9891,
+      "step": 2266
+    },
+    {
+      "epoch": 0.2015111111111111,
+      "grad_norm": 421.58514404296875,
+      "learning_rate": 3e-06,
+      "loss": -72.7071,
+      "step": 2267
+    },
+    {
+      "epoch": 0.2016,
+      "grad_norm": 499.9620361328125,
+      "learning_rate": 3e-06,
+      "loss": -86.6616,
+      "step": 2268
+    },
+    {
+      "completion_length": 251.43750762939453,
+      "epoch": 0.2016888888888889,
+      "grad_norm": 1807.4974365234375,
+      "learning_rate": 3e-06,
+      "loss": -2.4199,
+      "reward": 1.4375,
+      "reward_std": 0.46232306957244873,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 2269,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.20177777777777778,
+      "grad_norm": 1014.031005859375,
+      "learning_rate": 3e-06,
+      "loss": -11.0913,
+      "step": 2270
+    },
+    {
+      "epoch": 0.20186666666666667,
+      "grad_norm": 969.154541015625,
+      "learning_rate": 3e-06,
+      "loss": -45.8343,
+      "step": 2271
+    },
+    {
+      "epoch": 0.20195555555555555,
+      "grad_norm": 1514.218994140625,
+      "learning_rate": 3e-06,
+      "loss": -6.767,
+      "step": 2272
+    },
+    {
+      "epoch": 0.20204444444444444,
+      "grad_norm": 1047.0552978515625,
+      "learning_rate": 3e-06,
+      "loss": -41.9022,
+      "step": 2273
+    },
+    {
+      "epoch": 0.20213333333333333,
+      "grad_norm": 1069.608642578125,
+      "learning_rate": 3e-06,
+      "loss": -32.74,
+      "step": 2274
+    },
+    {
+      "epoch": 0.20222222222222222,
+      "grad_norm": 1186.3797607421875,
+      "learning_rate": 3e-06,
+      "loss": -5.627,
+      "step": 2275
+    },
+    {
+      "epoch": 0.2023111111111111,
+      "grad_norm": 1353.1217041015625,
+      "learning_rate": 3e-06,
+      "loss": -20.7514,
+      "step": 2276
+    },
+    {
+      "epoch": 0.2024,
+      "grad_norm": 973.7822875976562,
+      "learning_rate": 3e-06,
+      "loss": -57.7389,
+      "step": 2277
+    },
+    {
+      "epoch": 0.20248888888888888,
+      "grad_norm": 1122.6533203125,
+      "learning_rate": 3e-06,
+      "loss": -19.221,
+      "step": 2278
+    },
+    {
+      "epoch": 0.20257777777777777,
+      "grad_norm": 1007.2978515625,
+      "learning_rate": 3e-06,
+      "loss": -49.3046,
+      "step": 2279
+    },
+    {
+      "epoch": 0.20266666666666666,
+      "grad_norm": 1087.270751953125,
+      "learning_rate": 3e-06,
+      "loss": -46.9498,
+      "step": 2280
+    },
+    {
+      "completion_length": 248.0,
+      "epoch": 0.20275555555555555,
+      "grad_norm": 684.866455078125,
+      "learning_rate": 3e-06,
+      "loss": -24.7332,
+      "reward": 1.0416666865348816,
+      "reward_std": 0.40296074748039246,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 0.4166666716337204,
+      "step": 2281,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.20284444444444444,
+      "grad_norm": 685.3329467773438,
+      "learning_rate": 3e-06,
+      "loss": 10.4014,
+      "step": 2282
+    },
+    {
+      "epoch": 0.20293333333333333,
+      "grad_norm": 828.5563354492188,
+      "learning_rate": 3e-06,
+      "loss": -17.5656,
+      "step": 2283
+    },
+    {
+      "epoch": 0.20302222222222222,
+      "grad_norm": 797.2943115234375,
+      "learning_rate": 3e-06,
+      "loss": -24.5962,
+      "step": 2284
+    },
+    {
+      "epoch": 0.2031111111111111,
+      "grad_norm": 778.4367065429688,
+      "learning_rate": 3e-06,
+      "loss": 13.921,
+      "step": 2285
+    },
+    {
+      "epoch": 0.2032,
+      "grad_norm": 694.8433837890625,
+      "learning_rate": 3e-06,
+      "loss": -1.7194,
+      "step": 2286
+    },
+    {
+      "epoch": 0.20328888888888888,
+      "grad_norm": 808.0003662109375,
+      "learning_rate": 3e-06,
+      "loss": -30.4908,
+      "step": 2287
+    },
+    {
+      "epoch": 0.20337777777777777,
+      "grad_norm": 724.6696166992188,
+      "learning_rate": 3e-06,
+      "loss": 2.6068,
+      "step": 2288
+    },
+    {
+      "epoch": 0.20346666666666666,
+      "grad_norm": 830.2708129882812,
+      "learning_rate": 3e-06,
+      "loss": -28.4917,
+      "step": 2289
+    },
+    {
+      "epoch": 0.20355555555555555,
+      "grad_norm": 785.9896850585938,
+      "learning_rate": 3e-06,
+      "loss": -31.1686,
+      "step": 2290
+    },
+    {
+      "epoch": 0.20364444444444443,
+      "grad_norm": 851.0347290039062,
+      "learning_rate": 3e-06,
+      "loss": 0.2481,
+      "step": 2291
+    },
+    {
+      "epoch": 0.20373333333333332,
+      "grad_norm": 774.1303100585938,
+      "learning_rate": 3e-06,
+      "loss": -6.2776,
+      "step": 2292
+    },
+    {
+      "completion_length": 246.0625,
+      "epoch": 0.2038222222222222,
+      "grad_norm": 635.4057006835938,
+      "learning_rate": 3e-06,
+      "loss": 4.3122,
+      "reward": 1.5104167461395264,
+      "reward_std": 0.34120412170886993,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 0.8333333283662796,
+      "step": 2293,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.2039111111111111,
+      "grad_norm": 653.86328125,
+      "learning_rate": 3e-06,
+      "loss": 12.7972,
+      "step": 2294
+    },
+    {
+      "epoch": 0.204,
+      "grad_norm": 610.3839721679688,
+      "learning_rate": 3e-06,
+      "loss": -2.9267,
+      "step": 2295
+    },
+    {
+      "epoch": 0.20408888888888888,
+      "grad_norm": 714.1361083984375,
+      "learning_rate": 3e-06,
+      "loss": -23.5832,
+      "step": 2296
+    },
+    {
+      "epoch": 0.2041777777777778,
+      "grad_norm": 718.49658203125,
+      "learning_rate": 3e-06,
+      "loss": -4.627,
+      "step": 2297
+    },
+    {
+      "epoch": 0.20426666666666668,
+      "grad_norm": 696.8006591796875,
+      "learning_rate": 3e-06,
+      "loss": -11.8357,
+      "step": 2298
+    },
+    {
+      "epoch": 0.20435555555555557,
+      "grad_norm": 814.3312377929688,
+      "learning_rate": 3e-06,
+      "loss": -1.5097,
+      "step": 2299
+    },
+    {
+      "epoch": 0.20444444444444446,
+      "grad_norm": 556.5509643554688,
+      "learning_rate": 3e-06,
+      "loss": 10.6618,
+      "step": 2300
+    },
+    {
+      "epoch": 0.20453333333333334,
+      "grad_norm": 667.0651245117188,
+      "learning_rate": 3e-06,
+      "loss": -6.0333,
+      "step": 2301
+    },
+    {
+      "epoch": 0.20462222222222223,
+      "grad_norm": 694.1640625,
+      "learning_rate": 3e-06,
+      "loss": -24.5284,
+      "step": 2302
+    },
+    {
+      "epoch": 0.20471111111111112,
+      "grad_norm": 923.972900390625,
+      "learning_rate": 3e-06,
+      "loss": -11.2173,
+      "step": 2303
+    },
+    {
+      "epoch": 0.2048,
+      "grad_norm": 659.1800537109375,
+      "learning_rate": 3e-06,
+      "loss": -16.7782,
+      "step": 2304
+    },
+    {
+      "completion_length": 254.4375,
+      "epoch": 0.2048888888888889,
+      "grad_norm": 971.201171875,
+      "learning_rate": 3e-06,
+      "loss": -6.4646,
+      "reward": 2.125,
+      "reward_std": 0.43528565764427185,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.375,
+      "step": 2305,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.2049777777777778,
+      "grad_norm": 1799.6265869140625,
+      "learning_rate": 3e-06,
+      "loss": 13.6479,
+      "step": 2306
+    },
+    {
+      "epoch": 0.20506666666666667,
+      "grad_norm": 1432.3759765625,
+      "learning_rate": 3e-06,
+      "loss": -12.8131,
+      "step": 2307
+    },
+    {
+      "epoch": 0.20515555555555556,
+      "grad_norm": 1100.3665771484375,
+      "learning_rate": 3e-06,
+      "loss": 19.3096,
+      "step": 2308
+    },
+    {
+      "epoch": 0.20524444444444445,
+      "grad_norm": 954.0089111328125,
+      "learning_rate": 3e-06,
+      "loss": -5.624,
+      "step": 2309
+    },
+    {
+      "epoch": 0.20533333333333334,
+      "grad_norm": 1022.4109497070312,
+      "learning_rate": 3e-06,
+      "loss": -26.2045,
+      "step": 2310
+    },
+    {
+      "epoch": 0.20542222222222223,
+      "grad_norm": 1009.416015625,
+      "learning_rate": 3e-06,
+      "loss": -9.7483,
+      "step": 2311
+    },
+    {
+      "epoch": 0.20551111111111112,
+      "grad_norm": 1540.92333984375,
+      "learning_rate": 3e-06,
+      "loss": 0.92,
+      "step": 2312
+    },
+    {
+      "epoch": 0.2056,
+      "grad_norm": 1027.509765625,
+      "learning_rate": 3e-06,
+      "loss": -24.657,
+      "step": 2313
+    },
+    {
+      "epoch": 0.2056888888888889,
+      "grad_norm": 948.5579833984375,
+      "learning_rate": 3e-06,
+      "loss": 8.6437,
+      "step": 2314
+    },
+    {
+      "epoch": 0.20577777777777778,
+      "grad_norm": 919.4548950195312,
+      "learning_rate": 3e-06,
+      "loss": -21.9334,
+      "step": 2315
+    },
+    {
+      "epoch": 0.20586666666666667,
+      "grad_norm": 979.2217407226562,
+      "learning_rate": 3e-06,
+      "loss": -36.1305,
+      "step": 2316
+    },
+    {
+      "completion_length": 253.52083587646484,
+      "epoch": 0.20595555555555556,
+      "grad_norm": 382.2681579589844,
+      "learning_rate": 3e-06,
+      "loss": 13.32,
+      "reward": 0.9791666865348816,
+      "reward_std": 0.10206206887960434,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.2916666679084301,
+      "step": 2317,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.20604444444444445,
+      "grad_norm": 585.9896850585938,
+      "learning_rate": 3e-06,
+      "loss": 14.567,
+      "step": 2318
+    },
+    {
+      "epoch": 0.20613333333333334,
+      "grad_norm": 433.8167419433594,
+      "learning_rate": 3e-06,
+      "loss": 0.179,
+      "step": 2319
+    },
+    {
+      "epoch": 0.20622222222222222,
+      "grad_norm": 473.40289306640625,
+      "learning_rate": 3e-06,
+      "loss": -8.1342,
+      "step": 2320
+    },
+    {
+      "epoch": 0.2063111111111111,
+      "grad_norm": 454.9488220214844,
+      "learning_rate": 3e-06,
+      "loss": 6.4596,
+      "step": 2321
+    },
+    {
+      "epoch": 0.2064,
+      "grad_norm": 545.6451416015625,
+      "learning_rate": 3e-06,
+      "loss": -0.0272,
+      "step": 2322
+    },
+    {
+      "epoch": 0.2064888888888889,
+      "grad_norm": 393.2840576171875,
+      "learning_rate": 3e-06,
+      "loss": 10.1428,
+      "step": 2323
+    },
+    {
+      "epoch": 0.20657777777777778,
+      "grad_norm": 624.9196166992188,
+      "learning_rate": 3e-06,
+      "loss": 10.519,
+      "step": 2324
+    },
+    {
+      "epoch": 0.20666666666666667,
+      "grad_norm": 423.4202880859375,
+      "learning_rate": 3e-06,
+      "loss": -6.3045,
+      "step": 2325
+    },
+    {
+      "epoch": 0.20675555555555555,
+      "grad_norm": 396.0754699707031,
+      "learning_rate": 3e-06,
+      "loss": -11.096,
+      "step": 2326
+    },
+    {
+      "epoch": 0.20684444444444444,
+      "grad_norm": 583.927734375,
+      "learning_rate": 3e-06,
+      "loss": 2.5989,
+      "step": 2327
+    },
+    {
+      "epoch": 0.20693333333333333,
+      "grad_norm": 597.1937255859375,
+      "learning_rate": 3e-06,
+      "loss": -8.4196,
+      "step": 2328
+    },
+    {
+      "completion_length": 242.62500762939453,
+      "epoch": 0.20702222222222222,
+      "grad_norm": 800.9533081054688,
+      "learning_rate": 3e-06,
+      "loss": -33.827,
+      "reward": 1.4375000596046448,
+      "reward_std": 0.46232303977012634,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 2329,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.2071111111111111,
+      "grad_norm": 696.8334350585938,
+      "learning_rate": 3e-06,
+      "loss": -20.2565,
+      "step": 2330
+    },
+    {
+      "epoch": 0.2072,
+      "grad_norm": 728.7371826171875,
+      "learning_rate": 3e-06,
+      "loss": -55.6491,
+      "step": 2331
+    },
+    {
+      "epoch": 0.20728888888888888,
+      "grad_norm": 845.1964111328125,
+      "learning_rate": 3e-06,
+      "loss": -34.6344,
+      "step": 2332
+    },
+    {
+      "epoch": 0.20737777777777777,
+      "grad_norm": 797.7058715820312,
+      "learning_rate": 3e-06,
+      "loss": -38.3381,
+      "step": 2333
+    },
+    {
+      "epoch": 0.20746666666666666,
+      "grad_norm": 815.6392211914062,
+      "learning_rate": 3e-06,
+      "loss": -39.5545,
+      "step": 2334
+    },
+    {
+      "epoch": 0.20755555555555555,
+      "grad_norm": 824.5341796875,
+      "learning_rate": 3e-06,
+      "loss": -42.9804,
+      "step": 2335
+    },
+    {
+      "epoch": 0.20764444444444444,
+      "grad_norm": 839.4075927734375,
+      "learning_rate": 3e-06,
+      "loss": -31.0013,
+      "step": 2336
+    },
+    {
+      "epoch": 0.20773333333333333,
+      "grad_norm": 808.272705078125,
+      "learning_rate": 3e-06,
+      "loss": -63.1393,
+      "step": 2337
+    },
+    {
+      "epoch": 0.2078222222222222,
+      "grad_norm": 937.5029296875,
+      "learning_rate": 3e-06,
+      "loss": -49.1489,
+      "step": 2338
+    },
+    {
+      "epoch": 0.2079111111111111,
+      "grad_norm": 852.005859375,
+      "learning_rate": 3e-06,
+      "loss": -50.4064,
+      "step": 2339
+    },
+    {
+      "epoch": 0.208,
+      "grad_norm": 897.2970581054688,
+      "learning_rate": 3e-06,
+      "loss": -52.8892,
+      "step": 2340
+    },
+    {
+      "completion_length": 222.70834350585938,
+      "epoch": 0.20808888888888888,
+      "grad_norm": 757.0648803710938,
+      "learning_rate": 3e-06,
+      "loss": 35.128,
+      "reward": 1.7083333730697632,
+      "reward_std": 0.3332235887646675,
+      "rewards/boxed_and_answer_tags_format_reward": 0.625,
+      "rewards/correctness_reward_func_math": 1.0833333134651184,
+      "step": 2341,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.20817777777777777,
+      "grad_norm": 974.6918334960938,
+      "learning_rate": 3e-06,
+      "loss": 26.4346,
+      "step": 2342
+    },
+    {
+      "epoch": 0.20826666666666666,
+      "grad_norm": 960.5851440429688,
+      "learning_rate": 3e-06,
+      "loss": 18.2117,
+      "step": 2343
+    },
+    {
+      "epoch": 0.20835555555555554,
+      "grad_norm": 748.3045654296875,
+      "learning_rate": 3e-06,
+      "loss": 19.8586,
+      "step": 2344
+    },
+    {
+      "epoch": 0.20844444444444443,
+      "grad_norm": 900.9320068359375,
+      "learning_rate": 3e-06,
+      "loss": 9.7996,
+      "step": 2345
+    },
+    {
+      "epoch": 0.20853333333333332,
+      "grad_norm": 769.8694458007812,
+      "learning_rate": 3e-06,
+      "loss": 25.5797,
+      "step": 2346
+    },
+    {
+      "epoch": 0.2086222222222222,
+      "grad_norm": 734.0509033203125,
+      "learning_rate": 3e-06,
+      "loss": 27.9682,
+      "step": 2347
+    },
+    {
+      "epoch": 0.2087111111111111,
+      "grad_norm": 869.8523559570312,
+      "learning_rate": 3e-06,
+      "loss": 21.5981,
+      "step": 2348
+    },
+    {
+      "epoch": 0.2088,
+      "grad_norm": 1000.9803466796875,
+      "learning_rate": 3e-06,
+      "loss": 4.3671,
+      "step": 2349
+    },
+    {
+      "epoch": 0.2088888888888889,
+      "grad_norm": 766.8132934570312,
+      "learning_rate": 3e-06,
+      "loss": 10.1062,
+      "step": 2350
+    },
+    {
+      "epoch": 0.2089777777777778,
+      "grad_norm": 1140.35986328125,
+      "learning_rate": 3e-06,
+      "loss": -0.0209,
+      "step": 2351
+    },
+    {
+      "epoch": 0.20906666666666668,
+      "grad_norm": 769.6608276367188,
+      "learning_rate": 3e-06,
+      "loss": 17.1774,
+      "step": 2352
+    },
+    {
+      "completion_length": 235.12500762939453,
+      "epoch": 0.20915555555555557,
+      "grad_norm": 1358.110595703125,
+      "learning_rate": 3e-06,
+      "loss": -29.3384,
+      "reward": 1.5625000596046448,
+      "reward_std": 0.599253699183464,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.8750000298023224,
+      "step": 2353,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.20924444444444446,
+      "grad_norm": 1078.0203857421875,
+      "learning_rate": 3e-06,
+      "loss": -4.6919,
+      "step": 2354
+    },
+    {
+      "epoch": 0.20933333333333334,
+      "grad_norm": 1239.9210205078125,
+      "learning_rate": 3e-06,
+      "loss": -58.9311,
+      "step": 2355
+    },
+    {
+      "epoch": 0.20942222222222223,
+      "grad_norm": 1190.312255859375,
+      "learning_rate": 3e-06,
+      "loss": -41.3338,
+      "step": 2356
+    },
+    {
+      "epoch": 0.20951111111111112,
+      "grad_norm": 1177.6978759765625,
+      "learning_rate": 3e-06,
+      "loss": -77.211,
+      "step": 2357
+    },
+    {
+      "epoch": 0.2096,
+      "grad_norm": 1262.98876953125,
+      "learning_rate": 3e-06,
+      "loss": -30.6502,
+      "step": 2358
+    },
+    {
+      "epoch": 0.2096888888888889,
+      "grad_norm": 1281.507568359375,
+      "learning_rate": 3e-06,
+      "loss": -46.3909,
+      "step": 2359
+    },
+    {
+      "epoch": 0.20977777777777779,
+      "grad_norm": 1242.6148681640625,
+      "learning_rate": 3e-06,
+      "loss": -20.0434,
+      "step": 2360
+    },
+    {
+      "epoch": 0.20986666666666667,
+      "grad_norm": 1216.9324951171875,
+      "learning_rate": 3e-06,
+      "loss": -75.5012,
+      "step": 2361
+    },
+    {
+      "epoch": 0.20995555555555556,
+      "grad_norm": 1148.396240234375,
+      "learning_rate": 3e-06,
+      "loss": -60.3074,
+      "step": 2362
+    },
+    {
+      "epoch": 0.21004444444444445,
+      "grad_norm": 1038.05224609375,
+      "learning_rate": 3e-06,
+      "loss": -90.1416,
+      "step": 2363
+    },
+    {
+      "epoch": 0.21013333333333334,
+      "grad_norm": 1481.82470703125,
+      "learning_rate": 3e-06,
+      "loss": -41.4112,
+      "step": 2364
+    },
+    {
+      "completion_length": 236.4375,
+      "epoch": 0.21022222222222223,
+      "grad_norm": 1174.8533935546875,
+      "learning_rate": 3e-06,
+      "loss": -15.5789,
+      "reward": 1.2500000298023224,
+      "reward_std": 0.4431168735027313,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.4999999925494194,
+      "step": 2365,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.21031111111111112,
+      "grad_norm": 3015.200927734375,
+      "learning_rate": 3e-06,
+      "loss": -27.8838,
+      "step": 2366
+    },
+    {
+      "epoch": 0.2104,
+      "grad_norm": 1439.3358154296875,
+      "learning_rate": 3e-06,
+      "loss": -38.5885,
+      "step": 2367
+    },
+    {
+      "epoch": 0.2104888888888889,
+      "grad_norm": 869.1331787109375,
+      "learning_rate": 3e-06,
+      "loss": -19.1413,
+      "step": 2368
+    },
+    {
+      "epoch": 0.21057777777777778,
+      "grad_norm": 954.2206420898438,
+      "learning_rate": 3e-06,
+      "loss": -22.8123,
+      "step": 2369
+    },
+    {
+      "epoch": 0.21066666666666667,
+      "grad_norm": 923.0101928710938,
+      "learning_rate": 3e-06,
+      "loss": -34.3305,
+      "step": 2370
+    },
+    {
+      "epoch": 0.21075555555555556,
+      "grad_norm": 1316.6534423828125,
+      "learning_rate": 3e-06,
+      "loss": -30.9986,
+      "step": 2371
+    },
+    {
+      "epoch": 0.21084444444444445,
+      "grad_norm": 1236.0667724609375,
+      "learning_rate": 3e-06,
+      "loss": -48.6076,
+      "step": 2372
+    },
+    {
+      "epoch": 0.21093333333333333,
+      "grad_norm": 1252.537109375,
+      "learning_rate": 3e-06,
+      "loss": -61.8291,
+      "step": 2373
+    },
+    {
+      "epoch": 0.21102222222222222,
+      "grad_norm": 1343.861328125,
+      "learning_rate": 3e-06,
+      "loss": -29.769,
+      "step": 2374
+    },
+    {
+      "epoch": 0.2111111111111111,
+      "grad_norm": 1069.728515625,
+      "learning_rate": 3e-06,
+      "loss": -31.3108,
+      "step": 2375
+    },
+    {
+      "epoch": 0.2112,
+      "grad_norm": 898.3480834960938,
+      "learning_rate": 3e-06,
+      "loss": -53.5648,
+      "step": 2376
+    },
+    {
+      "completion_length": 250.2916717529297,
+      "epoch": 0.2112888888888889,
+      "grad_norm": 1555.4439697265625,
+      "learning_rate": 3e-06,
+      "loss": 23.924,
+      "reward": 0.9583333432674408,
+      "reward_std": 0.43528565764427185,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.2083333358168602,
+      "step": 2377,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.21137777777777778,
+      "grad_norm": 1271.503662109375,
+      "learning_rate": 3e-06,
+      "loss": -28.6245,
+      "step": 2378
+    },
+    {
+      "epoch": 0.21146666666666666,
+      "grad_norm": 1083.2822265625,
+      "learning_rate": 3e-06,
+      "loss": -61.1598,
+      "step": 2379
+    },
+    {
+      "epoch": 0.21155555555555555,
+      "grad_norm": 950.6062622070312,
+      "learning_rate": 3e-06,
+      "loss": -19.0397,
+      "step": 2380
+    },
+    {
+      "epoch": 0.21164444444444444,
+      "grad_norm": 1111.03857421875,
+      "learning_rate": 3e-06,
+      "loss": 11.29,
+      "step": 2381
+    },
+    {
+      "epoch": 0.21173333333333333,
+      "grad_norm": 1028.947509765625,
+      "learning_rate": 3e-06,
+      "loss": -16.2803,
+      "step": 2382
+    },
+    {
+      "epoch": 0.21182222222222222,
+      "grad_norm": 1200.6107177734375,
+      "learning_rate": 3e-06,
+      "loss": 14.5257,
+      "step": 2383
+    },
+    {
+      "epoch": 0.2119111111111111,
+      "grad_norm": 1430.796630859375,
+      "learning_rate": 3e-06,
+      "loss": -34.2578,
+      "step": 2384
+    },
+    {
+      "epoch": 0.212,
+      "grad_norm": 1028.548583984375,
+      "learning_rate": 3e-06,
+      "loss": -68.5076,
+      "step": 2385
+    },
+    {
+      "epoch": 0.21208888888888888,
+      "grad_norm": 1001.2411499023438,
+      "learning_rate": 3e-06,
+      "loss": -25.6736,
+      "step": 2386
+    },
+    {
+      "epoch": 0.21217777777777777,
+      "grad_norm": 984.0308227539062,
+      "learning_rate": 3e-06,
+      "loss": -2.7987,
+      "step": 2387
+    },
+    {
+      "epoch": 0.21226666666666666,
+      "grad_norm": 965.3121337890625,
+      "learning_rate": 3e-06,
+      "loss": -28.9898,
+      "step": 2388
+    },
+    {
+      "completion_length": 229.2916717529297,
+      "epoch": 0.21235555555555555,
+      "grad_norm": 557.446533203125,
+      "learning_rate": 3e-06,
+      "loss": 21.0555,
+      "reward": 1.8437500596046448,
+      "reward_std": 0.1546149756759405,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6770833432674408,
+      "rewards/correctness_reward_func_math": 1.1666666716337204,
+      "step": 2389,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.21244444444444444,
+      "grad_norm": 575.1875610351562,
+      "learning_rate": 3e-06,
+      "loss": 23.3964,
+      "step": 2390
+    },
+    {
+      "epoch": 0.21253333333333332,
+      "grad_norm": 360.0423278808594,
+      "learning_rate": 3e-06,
+      "loss": 33.3316,
+      "step": 2391
+    },
+    {
+      "epoch": 0.2126222222222222,
+      "grad_norm": 255.54953002929688,
+      "learning_rate": 3e-06,
+      "loss": 31.5129,
+      "step": 2392
+    },
+    {
+      "epoch": 0.2127111111111111,
+      "grad_norm": 265.9867248535156,
+      "learning_rate": 3e-06,
+      "loss": 16.7137,
+      "step": 2393
+    },
+    {
+      "epoch": 0.2128,
+      "grad_norm": 356.7539978027344,
+      "learning_rate": 3e-06,
+      "loss": 29.8486,
+      "step": 2394
+    },
+    {
+      "epoch": 0.21288888888888888,
+      "grad_norm": 380.1522521972656,
+      "learning_rate": 3e-06,
+      "loss": 18.9171,
+      "step": 2395
+    },
+    {
+      "epoch": 0.21297777777777777,
+      "grad_norm": 627.226806640625,
+      "learning_rate": 3e-06,
+      "loss": 14.277,
+      "step": 2396
+    },
+    {
+      "epoch": 0.21306666666666665,
+      "grad_norm": 473.0029296875,
+      "learning_rate": 3e-06,
+      "loss": 26.1599,
+      "step": 2397
+    },
+    {
+      "epoch": 0.21315555555555554,
+      "grad_norm": 256.2850646972656,
+      "learning_rate": 3e-06,
+      "loss": 27.8036,
+      "step": 2398
+    },
+    {
+      "epoch": 0.21324444444444443,
+      "grad_norm": 294.08056640625,
+      "learning_rate": 3e-06,
+      "loss": 11.133,
+      "step": 2399
+    },
+    {
+      "epoch": 0.21333333333333335,
+      "grad_norm": 292.5602111816406,
+      "learning_rate": 3e-06,
+      "loss": 21.8162,
+      "step": 2400
+    },
+    {
+      "completion_length": 248.7291717529297,
+      "epoch": 0.21342222222222224,
+      "grad_norm": 991.9778442382812,
+      "learning_rate": 3e-06,
+      "loss": 12.964,
+      "reward": 1.5000000596046448,
+      "reward_std": 0.20412413775920868,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.75,
+      "step": 2401,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.21351111111111112,
+      "grad_norm": 644.521484375,
+      "learning_rate": 3e-06,
+      "loss": -4.3122,
+      "step": 2402
+    },
+    {
+      "epoch": 0.2136,
+      "grad_norm": 680.9287719726562,
+      "learning_rate": 3e-06,
+      "loss": 3.5692,
+      "step": 2403
+    },
+    {
+      "epoch": 0.2136888888888889,
+      "grad_norm": 824.704345703125,
+      "learning_rate": 3e-06,
+      "loss": 20.7909,
+      "step": 2404
+    },
+    {
+      "epoch": 0.2137777777777778,
+      "grad_norm": 956.5078125,
+      "learning_rate": 3e-06,
+      "loss": 10.9202,
+      "step": 2405
+    },
+    {
+      "epoch": 0.21386666666666668,
+      "grad_norm": 1738.125244140625,
+      "learning_rate": 3e-06,
+      "loss": 5.5927,
+      "step": 2406
+    },
+    {
+      "epoch": 0.21395555555555557,
+      "grad_norm": 920.435546875,
+      "learning_rate": 3e-06,
+      "loss": 2.9809,
+      "step": 2407
+    },
+    {
+      "epoch": 0.21404444444444445,
+      "grad_norm": 677.7364501953125,
+      "learning_rate": 3e-06,
+      "loss": -7.6728,
+      "step": 2408
+    },
+    {
+      "epoch": 0.21413333333333334,
+      "grad_norm": 724.0176391601562,
+      "learning_rate": 3e-06,
+      "loss": -5.1334,
+      "step": 2409
+    },
+    {
+      "epoch": 0.21422222222222223,
+      "grad_norm": 885.7861938476562,
+      "learning_rate": 3e-06,
+      "loss": 17.665,
+      "step": 2410
+    },
+    {
+      "epoch": 0.21431111111111112,
+      "grad_norm": 884.716552734375,
+      "learning_rate": 3e-06,
+      "loss": 3.5877,
+      "step": 2411
+    },
+    {
+      "epoch": 0.2144,
+      "grad_norm": 1398.6461181640625,
+      "learning_rate": 3e-06,
+      "loss": 3.2704,
+      "step": 2412
+    },
+    {
+      "completion_length": 234.20834350585938,
+      "epoch": 0.2144888888888889,
+      "grad_norm": 1070.3880615234375,
+      "learning_rate": 3e-06,
+      "loss": -4.4756,
+      "reward": 1.3750000596046448,
+      "reward_std": 0.23116153478622437,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.6249999813735485,
+      "step": 2413,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.21457777777777778,
+      "grad_norm": 1026.6502685546875,
+      "learning_rate": 3e-06,
+      "loss": 44.4837,
+      "step": 2414
+    },
+    {
+      "epoch": 0.21466666666666667,
+      "grad_norm": 772.33837890625,
+      "learning_rate": 3e-06,
+      "loss": 37.0703,
+      "step": 2415
+    },
+    {
+      "epoch": 0.21475555555555556,
+      "grad_norm": 1085.96875,
+      "learning_rate": 3e-06,
+      "loss": 30.7537,
+      "step": 2416
+    },
+    {
+      "epoch": 0.21484444444444445,
+      "grad_norm": 827.8580932617188,
+      "learning_rate": 3e-06,
+      "loss": 48.8112,
+      "step": 2417
+    },
+    {
+      "epoch": 0.21493333333333334,
+      "grad_norm": 781.8536376953125,
+      "learning_rate": 3e-06,
+      "loss": 35.331,
+      "step": 2418
+    },
+    {
+      "epoch": 0.21502222222222223,
+      "grad_norm": 830.3689575195312,
+      "learning_rate": 3e-06,
+      "loss": -6.2381,
+      "step": 2419
+    },
+    {
+      "epoch": 0.21511111111111111,
+      "grad_norm": 1895.36572265625,
+      "learning_rate": 3e-06,
+      "loss": 38.7443,
+      "step": 2420
+    },
+    {
+      "epoch": 0.2152,
+      "grad_norm": 1093.376220703125,
+      "learning_rate": 3e-06,
+      "loss": 32.4039,
+      "step": 2421
+    },
+    {
+      "epoch": 0.2152888888888889,
+      "grad_norm": 1121.695556640625,
+      "learning_rate": 3e-06,
+      "loss": 21.949,
+      "step": 2422
+    },
+    {
+      "epoch": 0.21537777777777778,
+      "grad_norm": 908.1818237304688,
+      "learning_rate": 3e-06,
+      "loss": 39.3356,
+      "step": 2423
+    },
+    {
+      "epoch": 0.21546666666666667,
+      "grad_norm": 780.5909423828125,
+      "learning_rate": 3e-06,
+      "loss": 28.4789,
+      "step": 2424
+    },
+    {
+      "completion_length": 240.37500762939453,
+      "epoch": 0.21555555555555556,
+      "grad_norm": 1291.5950927734375,
+      "learning_rate": 3e-06,
+      "loss": -48.4281,
+      "reward": 1.625,
+      "reward_std": 0.564385175704956,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.875,
+      "step": 2425,
+      "zero_std_ratio": 0.375
+    },
+    {
+      "epoch": 0.21564444444444444,
+      "grad_norm": 1240.4404296875,
+      "learning_rate": 3e-06,
+      "loss": -54.5814,
+      "step": 2426
+    },
+    {
+      "epoch": 0.21573333333333333,
+      "grad_norm": 1328.792724609375,
+      "learning_rate": 3e-06,
+      "loss": -77.3574,
+      "step": 2427
+    },
+    {
+      "epoch": 0.21582222222222222,
+      "grad_norm": 1365.1212158203125,
+      "learning_rate": 3e-06,
+      "loss": -56.5684,
+      "step": 2428
+    },
+    {
+      "epoch": 0.2159111111111111,
+      "grad_norm": 1141.714111328125,
+      "learning_rate": 3e-06,
+      "loss": -52.8116,
+      "step": 2429
+    },
+    {
+      "epoch": 0.216,
+      "grad_norm": 1254.617919921875,
+      "learning_rate": 3e-06,
+      "loss": -45.3142,
+      "step": 2430
+    },
+    {
+      "epoch": 0.21608888888888889,
+      "grad_norm": 1215.14404296875,
+      "learning_rate": 3e-06,
+      "loss": -61.5184,
+      "step": 2431
+    },
+    {
+      "epoch": 0.21617777777777777,
+      "grad_norm": 2824.7529296875,
+      "learning_rate": 3e-06,
+      "loss": -67.1598,
+      "step": 2432
+    },
+    {
+      "epoch": 0.21626666666666666,
+      "grad_norm": 1158.3133544921875,
+      "learning_rate": 3e-06,
+      "loss": -84.968,
+      "step": 2433
+    },
+    {
+      "epoch": 0.21635555555555555,
+      "grad_norm": 948.725341796875,
+      "learning_rate": 3e-06,
+      "loss": -67.6487,
+      "step": 2434
+    },
+    {
+      "epoch": 0.21644444444444444,
+      "grad_norm": 1137.3834228515625,
+      "learning_rate": 3e-06,
+      "loss": -60.4848,
+      "step": 2435
+    },
+    {
+      "epoch": 0.21653333333333333,
+      "grad_norm": 1104.3291015625,
+      "learning_rate": 3e-06,
+      "loss": -61.2857,
+      "step": 2436
+    },
+    {
+      "completion_length": 240.14583587646484,
+      "epoch": 0.21662222222222222,
+      "grad_norm": 668.9032592773438,
+      "learning_rate": 3e-06,
+      "loss": 11.3774,
+      "reward": 1.2291667461395264,
+      "reward_std": 0.23116152733564377,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.5416666716337204,
+      "step": 2437,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.2167111111111111,
+      "grad_norm": 636.1585693359375,
+      "learning_rate": 3e-06,
+      "loss": -7.5818,
+      "step": 2438
+    },
+    {
+      "epoch": 0.2168,
+      "grad_norm": 497.7554626464844,
+      "learning_rate": 3e-06,
+      "loss": 6.8708,
+      "step": 2439
+    },
+    {
+      "epoch": 0.21688888888888888,
+      "grad_norm": 553.154052734375,
+      "learning_rate": 3e-06,
+      "loss": 24.3757,
+      "step": 2440
+    },
+    {
+      "epoch": 0.21697777777777777,
+      "grad_norm": 1472.5419921875,
+      "learning_rate": 3e-06,
+      "loss": -3.4629,
+      "step": 2441
+    },
+    {
+      "epoch": 0.21706666666666666,
+      "grad_norm": 652.6502685546875,
+      "learning_rate": 3e-06,
+      "loss": 18.709,
+      "step": 2442
+    },
+    {
+      "epoch": 0.21715555555555555,
+      "grad_norm": 599.662353515625,
+      "learning_rate": 3e-06,
+      "loss": 6.6085,
+      "step": 2443
+    },
+    {
+      "epoch": 0.21724444444444443,
+      "grad_norm": 596.9681396484375,
+      "learning_rate": 3e-06,
+      "loss": -10.8602,
+      "step": 2444
+    },
+    {
+      "epoch": 0.21733333333333332,
+      "grad_norm": 487.63580322265625,
+      "learning_rate": 3e-06,
+      "loss": -1.5804,
+      "step": 2445
+    },
+    {
+      "epoch": 0.2174222222222222,
+      "grad_norm": 1273.145751953125,
+      "learning_rate": 3e-06,
+      "loss": 16.3577,
+      "step": 2446
+    },
+    {
+      "epoch": 0.2175111111111111,
+      "grad_norm": 771.951904296875,
+      "learning_rate": 3e-06,
+      "loss": -6.3023,
+      "step": 2447
+    },
+    {
+      "epoch": 0.2176,
+      "grad_norm": 712.5418701171875,
+      "learning_rate": 3e-06,
+      "loss": 16.2355,
+      "step": 2448
+    },
+    {
+      "completion_length": 248.37500762939453,
+      "epoch": 0.21768888888888888,
+      "grad_norm": 790.022216796875,
+      "learning_rate": 3e-06,
+      "loss": 7.116,
+      "reward": 2.2291667461395264,
+      "reward_std": 0.26603007316589355,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 1.5416666865348816,
+      "step": 2449,
+      "zero_std_ratio": 0.75
+    },
+    {
+      "epoch": 0.21777777777777776,
+      "grad_norm": 546.7139892578125,
+      "learning_rate": 3e-06,
+      "loss": -10.9262,
+      "step": 2450
+    },
+    {
+      "epoch": 0.21786666666666665,
+      "grad_norm": 517.8162231445312,
+      "learning_rate": 3e-06,
+      "loss": 0.3609,
+      "step": 2451
+    },
+    {
+      "epoch": 0.21795555555555557,
+      "grad_norm": 682.6878051757812,
+      "learning_rate": 3e-06,
+      "loss": 0.4365,
+      "step": 2452
+    },
+    {
+      "epoch": 0.21804444444444446,
+      "grad_norm": 607.7827758789062,
+      "learning_rate": 3e-06,
+      "loss": 10.3393,
+      "step": 2453
+    },
+    {
+      "epoch": 0.21813333333333335,
+      "grad_norm": 518.8916625976562,
+      "learning_rate": 3e-06,
+      "loss": 0.2425,
+      "step": 2454
+    },
+    {
+      "epoch": 0.21822222222222223,
+      "grad_norm": 629.7891235351562,
+      "learning_rate": 3e-06,
+      "loss": -3.1223,
+      "step": 2455
+    },
+    {
+      "epoch": 0.21831111111111112,
+      "grad_norm": 510.4330139160156,
+      "learning_rate": 3e-06,
+      "loss": -15.683,
+      "step": 2456
+    },
+    {
+      "epoch": 0.2184,
+      "grad_norm": 553.461669921875,
+      "learning_rate": 3e-06,
+      "loss": -4.2699,
+      "step": 2457
+    },
+    {
+      "epoch": 0.2184888888888889,
+      "grad_norm": 526.62109375,
+      "learning_rate": 3e-06,
+      "loss": -6.1042,
+      "step": 2458
+    },
+    {
+      "epoch": 0.2185777777777778,
+      "grad_norm": 562.5404052734375,
+      "learning_rate": 3e-06,
+      "loss": -0.4739,
+      "step": 2459
+    },
+    {
+      "epoch": 0.21866666666666668,
+      "grad_norm": 544.2666625976562,
+      "learning_rate": 3e-06,
+      "loss": -6.7032,
+      "step": 2460
+    },
+    {
+      "completion_length": 246.02083587646484,
+      "epoch": 0.21875555555555556,
+      "grad_norm": 1100.9122314453125,
+      "learning_rate": 3e-06,
+      "loss": 12.7639,
+      "reward": 1.8750001192092896,
+      "reward_std": 0.3061862289905548,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 1.1250000298023224,
+      "step": 2461,
+      "zero_std_ratio": 0.625
+    },
+    {
+      "epoch": 0.21884444444444445,
+      "grad_norm": 897.7620239257812,
+      "learning_rate": 3e-06,
+      "loss": 20.9614,
+      "step": 2462
+    },
+    {
+      "epoch": 0.21893333333333334,
+      "grad_norm": 986.5072021484375,
+      "learning_rate": 3e-06,
+      "loss": 42.7016,
+      "step": 2463
+    },
+    {
+      "epoch": 0.21902222222222223,
+      "grad_norm": 1005.2491455078125,
+      "learning_rate": 3e-06,
+      "loss": 59.914,
+      "step": 2464
+    },
+    {
+      "epoch": 0.21911111111111112,
+      "grad_norm": 723.766357421875,
+      "learning_rate": 3e-06,
+      "loss": -18.4072,
+      "step": 2465
+    },
+    {
+      "epoch": 0.2192,
+      "grad_norm": 814.1770629882812,
+      "learning_rate": 3e-06,
+      "loss": 56.8266,
+      "step": 2466
+    },
+    {
+      "epoch": 0.2192888888888889,
+      "grad_norm": 880.131591796875,
+      "learning_rate": 3e-06,
+      "loss": 0.9147,
+      "step": 2467
+    },
+    {
+      "epoch": 0.21937777777777778,
+      "grad_norm": 1002.6812744140625,
+      "learning_rate": 3e-06,
+      "loss": 10.1945,
+      "step": 2468
+    },
+    {
+      "epoch": 0.21946666666666667,
+      "grad_norm": 928.541748046875,
+      "learning_rate": 3e-06,
+      "loss": 28.8268,
+      "step": 2469
+    },
+    {
+      "epoch": 0.21955555555555556,
+      "grad_norm": 1081.3568115234375,
+      "learning_rate": 3e-06,
+      "loss": 42.3389,
+      "step": 2470
+    },
+    {
+      "epoch": 0.21964444444444445,
+      "grad_norm": 849.524658203125,
+      "learning_rate": 3e-06,
+      "loss": -28.4933,
+      "step": 2471
+    },
+    {
+      "epoch": 0.21973333333333334,
+      "grad_norm": 800.930419921875,
+      "learning_rate": 3e-06,
+      "loss": 48.021,
+      "step": 2472
+    },
+    {
+      "completion_length": 243.7291717529297,
+      "epoch": 0.21982222222222222,
+      "grad_norm": 840.5211181640625,
+      "learning_rate": 3e-06,
+      "loss": -29.1071,
+      "reward": 1.4062500596046448,
+      "reward_std": 0.47030356526374817,
+      "rewards/boxed_and_answer_tags_format_reward": 0.7395833432674408,
+      "rewards/correctness_reward_func_math": 0.6666666567325592,
+      "step": 2473,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.2199111111111111,
+      "grad_norm": 766.2994384765625,
+      "learning_rate": 3e-06,
+      "loss": -22.7254,
+      "step": 2474
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 1315.990966796875,
+      "learning_rate": 3e-06,
+      "loss": -2.9843,
+      "step": 2475
+    },
+    {
+      "epoch": 0.2200888888888889,
+      "grad_norm": 823.3301391601562,
+      "learning_rate": 3e-06,
+      "loss": -45.7856,
+      "step": 2476
+    },
+    {
+      "epoch": 0.22017777777777778,
+      "grad_norm": 969.9705810546875,
+      "learning_rate": 3e-06,
+      "loss": -26.3312,
+      "step": 2477
+    },
+    {
+      "epoch": 0.22026666666666667,
+      "grad_norm": 1052.6732177734375,
+      "learning_rate": 3e-06,
+      "loss": -6.0265,
+      "step": 2478
+    },
+    {
+      "epoch": 0.22035555555555555,
+      "grad_norm": 810.92041015625,
+      "learning_rate": 3e-06,
+      "loss": -35.8816,
+      "step": 2479
+    },
+    {
+      "epoch": 0.22044444444444444,
+      "grad_norm": 716.6881713867188,
+      "learning_rate": 3e-06,
+      "loss": -34.8377,
+      "step": 2480
+    },
+    {
+      "epoch": 0.22053333333333333,
+      "grad_norm": 1079.2554931640625,
+      "learning_rate": 3e-06,
+      "loss": -13.4226,
+      "step": 2481
+    },
+    {
+      "epoch": 0.22062222222222222,
+      "grad_norm": 770.211669921875,
+      "learning_rate": 3e-06,
+      "loss": -50.3465,
+      "step": 2482
+    },
+    {
+      "epoch": 0.2207111111111111,
+      "grad_norm": 886.2747802734375,
+      "learning_rate": 3e-06,
+      "loss": -33.0956,
+      "step": 2483
+    },
+    {
+      "epoch": 0.2208,
+      "grad_norm": 890.3179321289062,
+      "learning_rate": 3e-06,
+      "loss": -15.2544,
+      "step": 2484
+    },
+    {
+      "completion_length": 253.58333587646484,
+      "epoch": 0.22088888888888888,
+      "grad_norm": 303.0287780761719,
+      "learning_rate": 3e-06,
+      "loss": -15.111,
+      "reward": 1.2916666865348816,
+      "reward_std": 0.10206207633018494,
+      "rewards/boxed_and_answer_tags_format_reward": 0.75,
+      "rewards/correctness_reward_func_math": 0.5416666567325592,
+      "step": 2485,
+      "zero_std_ratio": 0.875
+    },
+    {
+      "epoch": 0.22097777777777777,
+      "grad_norm": 377.69757080078125,
+      "learning_rate": 3e-06,
+      "loss": -16.8297,
+      "step": 2486
+    },
+    {
+      "epoch": 0.22106666666666666,
+      "grad_norm": 388.11773681640625,
+      "learning_rate": 3e-06,
+      "loss": -14.6188,
+      "step": 2487
+    },
+    {
+      "epoch": 0.22115555555555555,
+      "grad_norm": 291.3501892089844,
+      "learning_rate": 3e-06,
+      "loss": -13.4672,
+      "step": 2488
+    },
+    {
+      "epoch": 0.22124444444444444,
+      "grad_norm": 439.7605285644531,
+      "learning_rate": 3e-06,
+      "loss": -23.8357,
+      "step": 2489
+    },
+    {
+      "epoch": 0.22133333333333333,
+      "grad_norm": 379.401611328125,
+      "learning_rate": 3e-06,
+      "loss": -9.9997,
+      "step": 2490
+    },
+    {
+      "epoch": 0.22142222222222221,
+      "grad_norm": 297.3055725097656,
+      "learning_rate": 3e-06,
+      "loss": -18.6912,
+      "step": 2491
+    },
+    {
+      "epoch": 0.2215111111111111,
+      "grad_norm": 412.62890625,
+      "learning_rate": 3e-06,
+      "loss": -20.8417,
+      "step": 2492
+    },
+    {
+      "epoch": 0.2216,
+      "grad_norm": 300.6817321777344,
+      "learning_rate": 3e-06,
+      "loss": -19.9718,
+      "step": 2493
+    },
+    {
+      "epoch": 0.22168888888888888,
+      "grad_norm": 268.96551513671875,
+      "learning_rate": 3e-06,
+      "loss": -18.9416,
+      "step": 2494
+    },
+    {
+      "epoch": 0.22177777777777777,
+      "grad_norm": 426.9893798828125,
+      "learning_rate": 3e-06,
+      "loss": -27.8923,
+      "step": 2495
+    },
+    {
+      "epoch": 0.22186666666666666,
+      "grad_norm": 381.15704345703125,
+      "learning_rate": 3e-06,
+      "loss": -19.2658,
+      "step": 2496
+    },
+    {
+      "completion_length": 249.95833587646484,
+      "epoch": 0.22195555555555554,
+      "grad_norm": 1263.931640625,
+      "learning_rate": 3e-06,
+      "loss": -12.4574,
+      "reward": 1.1875,
+      "reward_std": 0.46232303977012634,
+      "rewards/boxed_and_answer_tags_format_reward": 0.6875,
+      "rewards/correctness_reward_func_math": 0.5,
+      "step": 2497,
+      "zero_std_ratio": 0.5
+    },
+    {
+      "epoch": 0.22204444444444443,
+      "grad_norm": 963.8367919921875,
+      "learning_rate": 3e-06,
+      "loss": 15.1327,
+      "step": 2498
+    },
+    {
+      "epoch": 0.22213333333333332,
+      "grad_norm": 1354.34619140625,
+      "learning_rate": 3e-06,
+      "loss": -37.5826,
+      "step": 2499
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 1030.41650390625,
+      "learning_rate": 3e-06,
+      "loss": 3.9406,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 112500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "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": 6,
+  "trial_name": null,
+  "trial_params": null
+}