diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,7537 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.08888888888888889,
+  "eval_steps": 1000,
+  "global_step": 1000,
+  "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
+    }
+  ],
+  "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": true,
+        "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
+}