{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5625,
  "eval_steps": 500,
  "global_step": 250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.6,
      "completions/max_length": 199.6,
      "completions/max_terminated_length": 134.0,
      "completions/mean_length": 171.9,
      "completions/mean_terminated_length": 122.36666870117188,
      "completions/min_length": 113.0,
      "completions/min_terminated_length": 113.0,
      "entropy": 0.06970996516756714,
      "epoch": 0.03125,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 3.630038261413574,
      "kl": 0.00014932112862879875,
      "learning_rate": 4.92e-06,
      "loss": 0.029165178537368774,
      "num_tokens": 15758.0,
      "reward": -0.31389998495578764,
      "reward_std": 0.2122000053524971,
      "rewards/reward_func/mean": -0.31389998495578764,
      "rewards/reward_func/std": 0.21219999492168426,
      "step": 5,
      "step_time": 14.728857926794444,
      "tools/call_frequency": 3.45,
      "tools/failure_frequency": 0.21573015451431274
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 168.6,
      "completions/max_terminated_length": 168.6,
      "completions/mean_length": 148.3,
      "completions/mean_terminated_length": 148.3,
      "completions/min_length": 129.6,
      "completions/min_terminated_length": 129.6,
      "entropy": 0.042718362715095284,
      "epoch": 0.0625,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 3.325033187866211,
      "kl": 0.037860750965774057,
      "learning_rate": 4.8200000000000004e-06,
      "loss": -0.011221970617771148,
      "num_tokens": 31053.0,
      "reward": 0.2989000082015991,
      "reward_std": 0.4415143087506294,
      "rewards/reward_func/mean": 0.2989000082015991,
      "rewards/reward_func/std": 0.4415143221616745,
      "step": 10,
      "step_time": 9.975367512006779,
      "tools/call_frequency": 2.5,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 152.8,
      "completions/max_terminated_length": 152.8,
      "completions/mean_length": 131.3,
      "completions/mean_terminated_length": 131.3,
      "completions/min_length": 113.6,
      "completions/min_terminated_length": 113.6,
      "entropy": 0.016039706021547317,
      "epoch": 0.09375,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 1.1289054155349731,
      "kl": 0.06640795171260834,
      "learning_rate": 4.7200000000000005e-06,
      "loss": 0.04752160608768463,
      "num_tokens": 45857.0,
      "reward": 1.1023000121116637,
      "reward_std": 0.4320605039596558,
      "rewards/reward_func/mean": 1.1023000121116637,
      "rewards/reward_func/std": 0.43206052780151366,
      "step": 15,
      "step_time": 8.620344271202338,
      "tools/call_frequency": 2.35,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.1,
      "completions/max_length": 172.0,
      "completions/max_terminated_length": 172.0,
      "completions/mean_length": 136.55,
      "completions/mean_terminated_length": 137.2500030517578,
      "completions/min_length": 111.4,
      "completions/min_terminated_length": 111.4,
      "entropy": 0.027425602450966834,
      "epoch": 0.125,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 0.8991426229476929,
      "kl": 0.09577701878733932,
      "learning_rate": 4.620000000000001e-06,
      "loss": -0.1201351523399353,
      "num_tokens": 60826.0,
      "reward": 0.7200000047683716,
      "reward_std": 0.3419178485870361,
      "rewards/reward_func/mean": 0.7200000047683716,
      "rewards/reward_func/std": 0.3419178485870361,
      "step": 20,
      "step_time": 11.403528443601681,
      "tools/call_frequency": 2.05,
      "tools/failure_frequency": 0.026666668057441712
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.35,
      "completions/max_length": 203.2,
      "completions/max_terminated_length": 161.2,
      "completions/mean_length": 188.6,
      "completions/mean_terminated_length": 150.6666687011719,
      "completions/min_length": 173.4,
      "completions/min_terminated_length": 139.6,
      "entropy": 0.033282498246990144,
      "epoch": 0.15625,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 2.041987657546997,
      "kl": 0.051508421916514634,
      "learning_rate": 4.520000000000001e-06,
      "loss": 0.03198407888412476,
      "num_tokens": 76838.0,
      "reward": 1.2669333696365357,
      "reward_std": 0.3234894543886185,
      "rewards/reward_func/mean": 1.2669333696365357,
      "rewards/reward_func/std": 0.32348946332931516,
      "step": 25,
      "step_time": 13.736867211584467,
      "tools/call_frequency": 4.4,
      "tools/failure_frequency": 0.14583333432674409
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.65,
      "completions/max_length": 209.4,
      "completions/max_terminated_length": 160.8,
      "completions/mean_length": 192.45,
      "completions/mean_terminated_length": 148.3,
      "completions/min_length": 169.2,
      "completions/min_terminated_length": 135.8,
      "entropy": 0.04025774166220799,
      "epoch": 0.1875,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 1.6383038759231567,
      "kl": 0.09242036554496735,
      "learning_rate": 4.42e-06,
      "loss": -0.03659022152423859,
      "num_tokens": 93054.0,
      "reward": 1.0333500146865844,
      "reward_std": 0.38981522917747496,
      "rewards/reward_func/mean": 1.0333500146865844,
      "rewards/reward_func/std": 0.389815217256546,
      "step": 30,
      "step_time": 14.735964270806289,
      "tools/call_frequency": 3.85,
      "tools/failure_frequency": 0.023529411852359773
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.7,
      "completions/max_length": 207.4,
      "completions/max_terminated_length": 168.8,
      "completions/mean_length": 196.4,
      "completions/mean_terminated_length": 166.7,
      "completions/min_length": 191.2,
      "completions/min_terminated_length": 164.6,
      "entropy": 0.02646293715806678,
      "epoch": 0.21875,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 0.6842532157897949,
      "kl": 0.09354882184416055,
      "learning_rate": 4.32e-06,
      "loss": 0.014650090038776398,
      "num_tokens": 109141.0,
      "reward": 1.0134333491325378,
      "reward_std": 0.28623148798942566,
      "rewards/reward_func/mean": 1.0134333491325378,
      "rewards/reward_func/std": 0.2862314820289612,
      "step": 35,
      "step_time": 14.25194917320332,
      "tools/call_frequency": 3.95,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.7,
      "completions/max_length": 228.8,
      "completions/max_terminated_length": 163.6,
      "completions/mean_length": 207.8,
      "completions/mean_terminated_length": 162.3,
      "completions/min_length": 188.8,
      "completions/min_terminated_length": 161.0,
      "entropy": 0.049686831969302146,
      "epoch": 0.25,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 2.386836528778076,
      "kl": 0.12552661653608083,
      "learning_rate": 4.22e-06,
      "loss": 0.023246073722839357,
      "num_tokens": 125712.0,
      "reward": 0.9764333426952362,
      "reward_std": 0.3545127585530281,
      "rewards/reward_func/mean": 0.9764333426952362,
      "rewards/reward_func/std": 0.35451277494430544,
      "step": 40,
      "step_time": 16.735324517198023,
      "tools/call_frequency": 3.45,
      "tools/failure_frequency": 0.02857142984867096
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.8,
      "completions/max_length": 231.4,
      "completions/max_terminated_length": 127.2,
      "completions/mean_length": 209.4,
      "completions/mean_terminated_length": 124.2,
      "completions/min_length": 187.2,
      "completions/min_terminated_length": 121.2,
      "entropy": 0.14096241008955984,
      "epoch": 0.28125,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 5.072839260101318,
      "kl": 0.10897002797573804,
      "learning_rate": 4.12e-06,
      "loss": 0.05337468385696411,
      "num_tokens": 142131.0,
      "reward": 1.0291000008583069,
      "reward_std": 0.5297403573989868,
      "rewards/reward_func/mean": 1.0291000008583069,
      "rewards/reward_func/std": 0.5297403573989868,
      "step": 45,
      "step_time": 17.371078941601446,
      "tools/call_frequency": 3.4,
      "tools/failure_frequency": 0.01428571492433548
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.6,
      "completions/max_length": 237.2,
      "completions/max_terminated_length": 172.2,
      "completions/mean_length": 201.7,
      "completions/mean_terminated_length": 160.46666870117187,
      "completions/min_length": 186.0,
      "completions/min_terminated_length": 150.8,
      "entropy": 0.1540619947016239,
      "epoch": 0.3125,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 5.2555952072143555,
      "kl": 0.17082785218954086,
      "learning_rate": 4.0200000000000005e-06,
      "loss": 0.06733548641204834,
      "num_tokens": 158431.0,
      "reward": 0.8427666783332824,
      "reward_std": 0.6860074520111084,
      "rewards/reward_func/mean": 0.8427666783332824,
      "rewards/reward_func/std": 0.6860074281692505,
      "step": 50,
      "step_time": 17.60776922639343,
      "tools/call_frequency": 3.6,
      "tools/failure_frequency": 0.027619048953056335
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.55,
      "completions/max_length": 209.2,
      "completions/max_terminated_length": 209.2,
      "completions/mean_length": 198.0,
      "completions/mean_terminated_length": 206.43333435058594,
      "completions/min_length": 189.4,
      "completions/min_terminated_length": 203.8,
      "entropy": 0.11417091116309167,
      "epoch": 0.34375,
      "frac_reward_zero_std": 0.8,
      "grad_norm": 0.1612984985113144,
      "kl": 0.14481508396565915,
      "learning_rate": 3.920000000000001e-06,
      "loss": -0.0013940947130322457,
      "num_tokens": 174665.0,
      "reward": 1.337833333015442,
      "reward_std": 0.04058598577976227,
      "rewards/reward_func/mean": 1.337833333015442,
      "rewards/reward_func/std": 0.04058598577976227,
      "step": 55,
      "step_time": 13.894916865596315,
      "tools/call_frequency": 3.9,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.35,
      "completions/max_length": 210.6,
      "completions/max_terminated_length": 210.0,
      "completions/mean_length": 189.7,
      "completions/mean_terminated_length": 184.93333435058594,
      "completions/min_length": 152.0,
      "completions/min_terminated_length": 153.4,
      "entropy": 0.18207021439447998,
      "epoch": 0.375,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 7.57163667678833,
      "kl": 0.2769763808697462,
      "learning_rate": 3.820000000000001e-06,
      "loss": -0.08738029599189759,
      "num_tokens": 190974.0,
      "reward": 0.9539999723434448,
      "reward_std": 0.24900673925876618,
      "rewards/reward_func/mean": 0.9539999723434448,
      "rewards/reward_func/std": 0.2490067459642887,
      "step": 60,
      "step_time": 13.735741792595945,
      "tools/call_frequency": 3.35,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.55,
      "completions/max_length": 213.6,
      "completions/max_terminated_length": 186.6,
      "completions/mean_length": 196.15,
      "completions/mean_terminated_length": 181.2,
      "completions/min_length": 173.2,
      "completions/min_terminated_length": 177.0,
      "entropy": 0.18931779703125357,
      "epoch": 0.40625,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 0.3368631601333618,
      "kl": 0.19928277991712093,
      "learning_rate": 3.7200000000000004e-06,
      "loss": -0.03082091510295868,
      "num_tokens": 207221.0,
      "reward": 1.1948333382606506,
      "reward_std": 0.3531351625919342,
      "rewards/reward_func/mean": 1.1948333382606506,
      "rewards/reward_func/std": 0.3531351566314697,
      "step": 65,
      "step_time": 14.853071747999639,
      "tools/call_frequency": 3.45,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.45,
      "completions/max_length": 232.4,
      "completions/max_terminated_length": 231.4,
      "completions/mean_length": 214.8,
      "completions/mean_terminated_length": 221.10000305175782,
      "completions/min_length": 199.4,
      "completions/min_terminated_length": 211.6,
      "entropy": 0.20331259737722576,
      "epoch": 0.4375,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 3.155299663543701,
      "kl": 0.21616111248731612,
      "learning_rate": 3.62e-06,
      "loss": -0.014388753473758698,
      "num_tokens": 223949.0,
      "reward": 1.187999999523163,
      "reward_std": 0.06400000005960464,
      "rewards/reward_func/mean": 1.187999999523163,
      "rewards/reward_func/std": 0.06399999856948853,
      "step": 70,
      "step_time": 15.7972018689994,
      "tools/call_frequency": 3.3,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.75,
      "completions/max_length": 228.0,
      "completions/max_terminated_length": 180.2,
      "completions/mean_length": 217.2,
      "completions/mean_terminated_length": 174.9,
      "completions/min_length": 208.2,
      "completions/min_terminated_length": 169.6,
      "entropy": 0.09874274502508343,
      "epoch": 0.46875,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 0.1496252417564392,
      "kl": 0.19251887053251265,
      "learning_rate": 3.52e-06,
      "loss": 0.0129203662276268,
      "num_tokens": 240663.0,
      "reward": 1.166100013256073,
      "reward_std": 0.27513332962989806,
      "rewards/reward_func/mean": 1.166100013256073,
      "rewards/reward_func/std": 0.275133341550827,
      "step": 75,
      "step_time": 15.653593644002104,
      "tools/call_frequency": 3.15,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.4,
      "completions/max_length": 250.6,
      "completions/max_terminated_length": 235.4,
      "completions/mean_length": 217.2,
      "completions/mean_terminated_length": 209.23333740234375,
      "completions/min_length": 169.8,
      "completions/min_terminated_length": 174.0,
      "entropy": 0.18624852728098631,
      "epoch": 0.5,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 3.9008662700653076,
      "kl": 0.19779104925692081,
      "learning_rate": 3.4200000000000007e-06,
      "loss": -0.060715597867965695,
      "num_tokens": 257232.0,
      "reward": 1.094600009918213,
      "reward_std": 0.533681058883667,
      "rewards/reward_func/mean": 1.094600009918213,
      "rewards/reward_func/std": 0.5336810708045959,
      "step": 80,
      "step_time": 16.87674882839783,
      "tools/call_frequency": 2.7,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.8,
      "completions/max_length": 263.2,
      "completions/max_terminated_length": 149.0,
      "completions/mean_length": 238.9,
      "completions/mean_terminated_length": 138.2,
      "completions/min_length": 220.0,
      "completions/min_terminated_length": 127.4,
      "entropy": 0.06761846686713398,
      "epoch": 0.53125,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 0.0583312027156353,
      "kl": 0.16298045124858618,
      "learning_rate": 3.3200000000000004e-06,
      "loss": 0.0317715585231781,
      "num_tokens": 274377.0,
      "reward": 1.168333351612091,
      "reward_std": 0.21399999260902405,
      "rewards/reward_func/mean": 1.168333351612091,
      "rewards/reward_func/std": 0.214000004529953,
      "step": 85,
      "step_time": 19.253501980405417,
      "tools/call_frequency": 2.45,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.9,
      "completions/max_length": 252.0,
      "completions/max_terminated_length": 79.6,
      "completions/mean_length": 229.35,
      "completions/mean_terminated_length": 79.6,
      "completions/min_length": 212.6,
      "completions/min_terminated_length": 79.6,
      "entropy": 0.04304317501373589,
      "epoch": 0.5625,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 0.05765737593173981,
      "kl": 0.1589741975069046,
      "learning_rate": 3.2200000000000005e-06,
      "loss": -0.009884151071310044,
      "num_tokens": 291640.0,
      "reward": 1.0771000266075135,
      "reward_std": 0.2571271777153015,
      "rewards/reward_func/mean": 1.0771000266075135,
      "rewards/reward_func/std": 0.257127183675766,
      "step": 90,
      "step_time": 19.810263851404308,
      "tools/call_frequency": 2.7,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.75,
      "completions/max_length": 252.2,
      "completions/max_terminated_length": 155.6,
      "completions/mean_length": 227.45,
      "completions/mean_terminated_length": 154.7,
      "completions/min_length": 202.6,
      "completions/min_terminated_length": 153.8,
      "entropy": 0.03856636304408312,
      "epoch": 0.59375,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 2.2899415493011475,
      "kl": 0.18391469195485116,
      "learning_rate": 3.12e-06,
      "loss": 0.012278559803962707,
      "num_tokens": 308671.0,
      "reward": 0.9493666887283325,
      "reward_std": 0.3057107627391815,
      "rewards/reward_func/mean": 0.9493666887283325,
      "rewards/reward_func/std": 0.3057107746601105,
      "step": 95,
      "step_time": 18.270148772597896,
      "tools/call_frequency": 2.75,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.65,
      "completions/max_length": 229.0,
      "completions/max_terminated_length": 128.6,
      "completions/mean_length": 210.05,
      "completions/mean_terminated_length": 121.23333435058593,
      "completions/min_length": 191.4,
      "completions/min_terminated_length": 114.6,
      "entropy": 0.03718785918317735,
      "epoch": 0.625,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 1.4016427993774414,
      "kl": 0.19276840873062612,
      "learning_rate": 3.0200000000000003e-06,
      "loss": -0.02043401300907135,
      "num_tokens": 325246.0,
      "reward": 0.9758000135421753,
      "reward_std": 0.439729905128479,
      "rewards/reward_func/mean": 0.9758000135421753,
      "rewards/reward_func/std": 0.439729905128479,
      "step": 100,
      "step_time": 16.536685503809714,
      "tools/call_frequency": 3.4,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.45,
      "completions/max_length": 231.0,
      "completions/max_terminated_length": 175.6,
      "completions/mean_length": 209.7,
      "completions/mean_terminated_length": 164.73333435058595,
      "completions/min_length": 197.2,
      "completions/min_terminated_length": 156.2,
      "entropy": 0.0890876273624599,
      "epoch": 0.65625,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 1.857412338256836,
      "kl": 0.20793221928179265,
      "learning_rate": 2.92e-06,
      "loss": 0.010671529173851012,
      "num_tokens": 341743.0,
      "reward": 1.244200015068054,
      "reward_std": 0.25437753796577456,
      "rewards/reward_func/mean": 1.244200015068054,
      "rewards/reward_func/std": 0.25437754988670347,
      "step": 105,
      "step_time": 14.550393618003,
      "tools/call_frequency": 3.4,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.6,
      "completions/max_length": 228.0,
      "completions/max_terminated_length": 215.0,
      "completions/mean_length": 211.05,
      "completions/mean_terminated_length": 204.6,
      "completions/min_length": 194.2,
      "completions/min_terminated_length": 194.2,
      "entropy": 0.09650332322344184,
      "epoch": 0.6875,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.5915409922599792,
      "kl": 0.1943995427340269,
      "learning_rate": 2.82e-06,
      "loss": -0.007803649455308914,
      "num_tokens": 358474.0,
      "reward": 0.9213667035102844,
      "reward_std": 0.48010437488555907,
      "rewards/reward_func/mean": 0.9213667035102844,
      "rewards/reward_func/std": 0.480104398727417,
      "step": 110,
      "step_time": 15.29034832160105,
      "tools/call_frequency": 3.3,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.5,
      "completions/max_length": 244.8,
      "completions/max_terminated_length": 244.8,
      "completions/mean_length": 224.4,
      "completions/mean_terminated_length": 232.93333435058594,
      "completions/min_length": 204.4,
      "completions/min_terminated_length": 219.8,
      "entropy": 0.06257005939260125,
      "epoch": 0.71875,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 0.09643584489822388,
      "kl": 0.18671961799263953,
      "learning_rate": 2.7200000000000002e-06,
      "loss": 0.0009367348626255989,
      "num_tokens": 375512.0,
      "reward": 0.9198000192642212,
      "reward_std": 0.41239041090011597,
      "rewards/reward_func/mean": 0.9198000192642212,
      "rewards/reward_func/std": 0.41239042282104493,
      "step": 115,
      "step_time": 16.68962257000094,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.5,
      "completions/max_length": 238.4,
      "completions/max_terminated_length": 194.2,
      "completions/mean_length": 221.6,
      "completions/mean_terminated_length": 185.86666870117188,
      "completions/min_length": 202.6,
      "completions/min_terminated_length": 174.8,
      "entropy": 0.19831047160550952,
      "epoch": 0.75,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 0.06480103731155396,
      "kl": 0.2127195455133915,
      "learning_rate": 2.6200000000000003e-06,
      "loss": -0.002893347479403019,
      "num_tokens": 392259.0,
      "reward": 1.1177000164985658,
      "reward_std": 0.34459384679794314,
      "rewards/reward_func/mean": 1.1177000164985658,
      "rewards/reward_func/std": 0.34459385871887205,
      "step": 120,
      "step_time": 15.74592421480629,
      "tools/call_frequency": 3.1,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.55,
      "completions/max_length": 249.6,
      "completions/max_terminated_length": 248.4,
      "completions/mean_length": 225.9,
      "completions/mean_terminated_length": 238.83333435058594,
      "completions/min_length": 207.6,
      "completions/min_terminated_length": 229.0,
      "entropy": 0.12472135615535081,
      "epoch": 0.78125,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 1.3778189420700073,
      "kl": 0.22096077986061574,
      "learning_rate": 2.52e-06,
      "loss": 0.018771570920944215,
      "num_tokens": 409108.0,
      "reward": 0.6021333426237107,
      "reward_std": 0.6133833765983582,
      "rewards/reward_func/mean": 0.6021333426237107,
      "rewards/reward_func/std": 0.6133833885192871,
      "step": 125,
      "step_time": 17.135429813191877,
      "tools/call_frequency": 3.0,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.6,
      "completions/max_length": 242.2,
      "completions/max_terminated_length": 214.8,
      "completions/mean_length": 219.0,
      "completions/mean_terminated_length": 209.9,
      "completions/min_length": 193.2,
      "completions/min_terminated_length": 203.6,
      "entropy": 0.15268438071943818,
      "epoch": 0.8125,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 2.5108072757720947,
      "kl": 0.2319534882903099,
      "learning_rate": 2.42e-06,
      "loss": -0.013115590810775757,
      "num_tokens": 425748.0,
      "reward": 0.9733999967575073,
      "reward_std": 0.6780932426452637,
      "rewards/reward_func/mean": 0.9733999967575073,
      "rewards/reward_func/std": 0.6780932545661926,
      "step": 130,
      "step_time": 16.86904642219888,
      "tools/call_frequency": 2.9,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.55,
      "completions/max_length": 238.2,
      "completions/max_terminated_length": 194.0,
      "completions/mean_length": 225.05,
      "completions/mean_terminated_length": 187.35,
      "completions/min_length": 212.8,
      "completions/min_terminated_length": 179.6,
      "entropy": 0.11236475300975143,
      "epoch": 0.84375,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 2.316483736038208,
      "kl": 0.18593905940651895,
      "learning_rate": 2.3200000000000002e-06,
      "loss": 0.003791916370391846,
      "num_tokens": 442384.0,
      "reward": 1.3600000143051147,
      "reward_std": 0.12066323161125184,
      "rewards/reward_func/mean": 1.3600000143051147,
      "rewards/reward_func/std": 0.12066323161125184,
      "step": 135,
      "step_time": 16.34641739600629,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.65,
      "completions/max_length": 242.6,
      "completions/max_terminated_length": 192.6,
      "completions/mean_length": 225.85,
      "completions/mean_terminated_length": 187.1,
      "completions/min_length": 212.6,
      "completions/min_terminated_length": 181.6,
      "entropy": 0.29790011094883084,
      "epoch": 0.875,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 4.7559309005737305,
      "kl": 0.25753427743911744,
      "learning_rate": 2.2200000000000003e-06,
      "loss": 0.003042304515838623,
      "num_tokens": 459098.0,
      "reward": 0.8764000177383423,
      "reward_std": 0.408263224363327,
      "rewards/reward_func/mean": 0.8764000177383423,
      "rewards/reward_func/std": 0.408263236284256,
      "step": 140,
      "step_time": 17.673848899203584,
      "tools/call_frequency": 2.9,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.75,
      "completions/max_length": 234.8,
      "completions/max_terminated_length": 140.4,
      "completions/mean_length": 218.5,
      "completions/mean_terminated_length": 135.4,
      "completions/min_length": 204.4,
      "completions/min_terminated_length": 130.4,
      "entropy": 0.49127169298008083,
      "epoch": 0.90625,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 0.07470813393592834,
      "kl": 0.2544811189174652,
      "learning_rate": 2.12e-06,
      "loss": 0.011104442924261094,
      "num_tokens": 475747.0,
      "reward": 1.2128000020980836,
      "reward_std": 0.3457141280174255,
      "rewards/reward_func/mean": 1.2128000020980836,
      "rewards/reward_func/std": 0.3457141280174255,
      "step": 145,
      "step_time": 17.512910800203098,
      "tools/call_frequency": 3.1,
      "tools/failure_frequency": 0.01428571492433548
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.6,
      "completions/max_length": 238.4,
      "completions/max_terminated_length": 187.0,
      "completions/mean_length": 219.55,
      "completions/mean_terminated_length": 176.0,
      "completions/min_length": 203.0,
      "completions/min_terminated_length": 163.4,
      "entropy": 0.3977989808190614,
      "epoch": 0.9375,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 1.4591357707977295,
      "kl": 0.26335713379085063,
      "learning_rate": 2.02e-06,
      "loss": 0.013294479250907898,
      "num_tokens": 492437.0,
      "reward": 0.9657333374023438,
      "reward_std": 0.5567267656326294,
      "rewards/reward_func/mean": 0.9657333374023438,
      "rewards/reward_func/std": 0.5567267417907715,
      "step": 150,
      "step_time": 17.259096857812256,
      "tools/call_frequency": 3.15,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.45,
      "completions/max_length": 232.2,
      "completions/max_terminated_length": 222.4,
      "completions/mean_length": 209.05,
      "completions/mean_terminated_length": 207.03333435058593,
      "completions/min_length": 176.2,
      "completions/min_terminated_length": 193.2,
      "entropy": 0.5087961174082011,
      "epoch": 0.96875,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 3.283621311187744,
      "kl": 0.28082513697445394,
      "learning_rate": 1.9200000000000003e-06,
      "loss": -0.025146692991256714,
      "num_tokens": 508782.0,
      "reward": 1.0452000081539154,
      "reward_std": 0.44076991081237793,
      "rewards/reward_func/mean": 1.0452000081539154,
      "rewards/reward_func/std": 0.4407699227333069,
      "step": 155,
      "step_time": 16.901877696395967,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.75,
      "completions/max_length": 237.6,
      "completions/max_terminated_length": 182.8,
      "completions/mean_length": 219.25,
      "completions/mean_terminated_length": 182.6,
      "completions/min_length": 204.8,
      "completions/min_terminated_length": 182.4,
      "entropy": 0.6035067357588559,
      "epoch": 1.0,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 3.8756589889526367,
      "kl": 0.27896949015557765,
      "learning_rate": 1.8200000000000002e-06,
      "loss": 0.018515169620513916,
      "num_tokens": 525445.0,
      "reward": 0.8603333592414856,
      "reward_std": 0.4929263710975647,
      "rewards/reward_func/mean": 0.8603333592414856,
      "rewards/reward_func/std": 0.4929263830184937,
      "step": 160,
      "step_time": 18.22537782279833,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.8,
      "completions/max_length": 226.4,
      "completions/max_terminated_length": 132.6,
      "completions/mean_length": 208.4,
      "completions/mean_terminated_length": 113.8,
      "completions/min_length": 178.0,
      "completions/min_terminated_length": 95.0,
      "entropy": 0.81973907337524,
      "epoch": 1.03125,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 0.23878704011440277,
      "kl": 0.28968340829014777,
      "learning_rate": 1.72e-06,
      "loss": -0.01732647567987442,
      "num_tokens": 541725.0,
      "reward": 0.801800012588501,
      "reward_std": 0.3781105220317841,
      "rewards/reward_func/mean": 0.801800012588501,
      "rewards/reward_func/std": 0.3781105220317841,
      "step": 165,
      "step_time": 17.06264049640158,
      "tools/call_frequency": 3.0,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.7,
      "completions/max_length": 234.2,
      "completions/max_terminated_length": 156.0,
      "completions/mean_length": 210.4,
      "completions/mean_terminated_length": 148.5,
      "completions/min_length": 176.8,
      "completions/min_terminated_length": 141.0,
      "entropy": 0.689805658115074,
      "epoch": 1.0625,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 4.456592082977295,
      "kl": 0.23804183304309845,
      "learning_rate": 1.6200000000000002e-06,
      "loss": -0.05976427793502807,
      "num_tokens": 558245.0,
      "reward": 0.8562000155448913,
      "reward_std": 0.633456540107727,
      "rewards/reward_func/mean": 0.8562000155448913,
      "rewards/reward_func/std": 0.633456540107727,
      "step": 170,
      "step_time": 18.225926614992204,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.018181818723678588
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.6,
      "completions/max_length": 226.2,
      "completions/max_terminated_length": 168.2,
      "completions/mean_length": 210.85,
      "completions/mean_terminated_length": 160.45,
      "completions/min_length": 196.2,
      "completions/min_terminated_length": 155.4,
      "entropy": 0.5438663156237453,
      "epoch": 1.09375,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 4.7369794845581055,
      "kl": 0.27658827155828475,
      "learning_rate": 1.52e-06,
      "loss": 0.017269280552864075,
      "num_tokens": 574485.0,
      "reward": 1.1061000227928162,
      "reward_std": 0.29246323108673095,
      "rewards/reward_func/mean": 1.1061000227928162,
      "rewards/reward_func/std": 0.29246323108673095,
      "step": 175,
      "step_time": 15.657095247798134,
      "tools/call_frequency": 3.25,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.7,
      "completions/max_length": 227.6,
      "completions/max_terminated_length": 153.8,
      "completions/mean_length": 210.05,
      "completions/mean_terminated_length": 150.3,
      "completions/min_length": 182.4,
      "completions/min_terminated_length": 146.8,
      "entropy": 0.45502517083659766,
      "epoch": 1.125,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 2.233670473098755,
      "kl": 0.24543451368808747,
      "learning_rate": 1.42e-06,
      "loss": -0.03468523025512695,
      "num_tokens": 591329.0,
      "reward": 0.9866333484649659,
      "reward_std": 0.43823108077049255,
      "rewards/reward_func/mean": 0.9866333484649659,
      "rewards/reward_func/std": 0.43823106288909913,
      "step": 180,
      "step_time": 16.63430393260496,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.018181818723678588
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.45,
      "completions/max_length": 238.8,
      "completions/max_terminated_length": 235.8,
      "completions/mean_length": 214.35,
      "completions/mean_terminated_length": 215.93333740234374,
      "completions/min_length": 179.6,
      "completions/min_terminated_length": 192.4,
      "entropy": 0.3232035285793245,
      "epoch": 1.15625,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 0.042836207896471024,
      "kl": 0.2335926942527294,
      "learning_rate": 1.32e-06,
      "loss": -0.053522664308547976,
      "num_tokens": 607959.0,
      "reward": 1.1267000198364259,
      "reward_std": 0.5779333353042603,
      "rewards/reward_func/mean": 1.1267000198364259,
      "rewards/reward_func/std": 0.5779333412647247,
      "step": 185,
      "step_time": 16.85379633680277,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.55,
      "completions/max_length": 234.8,
      "completions/max_terminated_length": 162.2,
      "completions/mean_length": 214.2,
      "completions/mean_terminated_length": 153.75,
      "completions/min_length": 188.8,
      "completions/min_terminated_length": 147.4,
      "entropy": 0.6248208525590598,
      "epoch": 1.1875,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 3.5707623958587646,
      "kl": 0.26286336220800877,
      "learning_rate": 1.2200000000000002e-06,
      "loss": -0.04159345626831055,
      "num_tokens": 624606.0,
      "reward": 0.7690666794776917,
      "reward_std": 0.5251105308532715,
      "rewards/reward_func/mean": 0.7690666794776917,
      "rewards/reward_func/std": 0.5251105427742004,
      "step": 190,
      "step_time": 15.895430696196854,
      "tools/call_frequency": 2.95,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.75,
      "completions/max_length": 231.2,
      "completions/max_terminated_length": 140.8,
      "completions/mean_length": 212.5,
      "completions/mean_terminated_length": 130.53333435058593,
      "completions/min_length": 182.6,
      "completions/min_terminated_length": 113.6,
      "entropy": 1.0036215352360158,
      "epoch": 1.21875,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 0.07126332074403763,
      "kl": 0.2945852160453796,
      "learning_rate": 1.12e-06,
      "loss": 0.08146535158157349,
      "num_tokens": 641178.0,
      "reward": 0.7420000076293946,
      "reward_std": 0.2832352787256241,
      "rewards/reward_func/mean": 0.7420000076293946,
      "rewards/reward_func/std": 0.28323529064655306,
      "step": 195,
      "step_time": 16.828464100402197,
      "tools/call_frequency": 2.9,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.8,
      "completions/max_length": 228.2,
      "completions/max_terminated_length": 139.8,
      "completions/mean_length": 217.85,
      "completions/mean_terminated_length": 138.8,
      "completions/min_length": 205.4,
      "completions/min_terminated_length": 137.8,
      "entropy": 0.7422479507047683,
      "epoch": 1.25,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 4.2798051834106445,
      "kl": 0.3032657243311405,
      "learning_rate": 1.02e-06,
      "loss": 0.017710180580616,
      "num_tokens": 657673.0,
      "reward": 0.8391000092029571,
      "reward_std": 0.7726872444152832,
      "rewards/reward_func/mean": 0.8391000092029571,
      "rewards/reward_func/std": 0.7726872682571411,
      "step": 200,
      "step_time": 17.576660793198972,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.75,
      "completions/max_length": 221.8,
      "completions/max_terminated_length": 146.4,
      "completions/mean_length": 207.7,
      "completions/mean_terminated_length": 143.7,
      "completions/min_length": 176.2,
      "completions/min_terminated_length": 141.0,
      "entropy": 0.28722639731131494,
      "epoch": 1.28125,
      "frac_reward_zero_std": 0.8,
      "grad_norm": 0.05100347474217415,
      "kl": 0.23604922145605087,
      "learning_rate": 9.200000000000001e-07,
      "loss": -0.04504288733005524,
      "num_tokens": 673877.0,
      "reward": 1.34099999666214,
      "reward_std": 0.09399999976158142,
      "rewards/reward_func/mean": 1.34099999666214,
      "rewards/reward_func/std": 0.09399999976158142,
      "step": 205,
      "step_time": 17.11068033759657,
      "tools/call_frequency": 2.95,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.75,
      "completions/max_length": 238.2,
      "completions/max_terminated_length": 176.8,
      "completions/mean_length": 221.0,
      "completions/mean_terminated_length": 175.9,
      "completions/min_length": 206.6,
      "completions/min_terminated_length": 175.0,
      "entropy": 0.7828503699507564,
      "epoch": 1.3125,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 3.9798154830932617,
      "kl": 0.2740111470222473,
      "learning_rate": 8.200000000000001e-07,
      "loss": 0.008357369899749756,
      "num_tokens": 690789.0,
      "reward": 1.0448000192642213,
      "reward_std": 0.33592903017997744,
      "rewards/reward_func/mean": 1.0448000192642213,
      "rewards/reward_func/std": 0.33592904210090635,
      "step": 210,
      "step_time": 16.50285193720192,
      "tools/call_frequency": 3.0,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.75,
      "completions/max_length": 234.6,
      "completions/max_terminated_length": 103.8,
      "completions/mean_length": 204.35,
      "completions/mean_terminated_length": 91.8,
      "completions/min_length": 156.4,
      "completions/min_terminated_length": 79.8,
      "entropy": 0.8003528213594109,
      "epoch": 1.34375,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 4.159558296203613,
      "kl": 0.2852329473942518,
      "learning_rate": 7.2e-07,
      "loss": -0.08619436025619506,
      "num_tokens": 707222.0,
      "reward": 1.0130333423614502,
      "reward_std": 0.6909683525562287,
      "rewards/reward_func/mean": 1.0130333423614502,
      "rewards/reward_func/std": 0.6909683525562287,
      "step": 215,
      "step_time": 17.172416884609266,
      "tools/call_frequency": 2.75,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.7,
      "completions/max_length": 227.6,
      "completions/max_terminated_length": 194.0,
      "completions/mean_length": 210.55,
      "completions/mean_terminated_length": 191.4,
      "completions/min_length": 179.4,
      "completions/min_terminated_length": 188.8,
      "entropy": 0.5576354823075235,
      "epoch": 1.375,
      "frac_reward_zero_std": 0.4,
      "grad_norm": 3.594050168991089,
      "kl": 0.278278523683548,
      "learning_rate": 6.200000000000001e-07,
      "loss": -0.04620848298072815,
      "num_tokens": 723693.0,
      "reward": 1.0568666815757752,
      "reward_std": 0.4007272839546204,
      "rewards/reward_func/mean": 1.0568666815757752,
      "rewards/reward_func/std": 0.4007272839546204,
      "step": 220,
      "step_time": 16.7619311846036,
      "tools/call_frequency": 2.9,
      "tools/failure_frequency": 0.01666666716337204
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.45,
      "completions/max_length": 238.8,
      "completions/max_terminated_length": 201.4,
      "completions/mean_length": 202.6,
      "completions/mean_terminated_length": 174.96666870117187,
      "completions/min_length": 148.0,
      "completions/min_terminated_length": 148.0,
      "entropy": 0.8566485311836004,
      "epoch": 1.40625,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 4.667319297790527,
      "kl": 0.2987564954906702,
      "learning_rate": 5.2e-07,
      "loss": -0.07293931245803834,
      "num_tokens": 740154.0,
      "reward": 0.9942000150680542,
      "reward_std": 0.6604385733604431,
      "rewards/reward_func/mean": 0.9942000150680542,
      "rewards/reward_func/std": 0.6604385554790497,
      "step": 225,
      "step_time": 16.59093914159166,
      "tools/call_frequency": 2.9,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.8,
      "completions/max_length": 230.2,
      "completions/max_terminated_length": 118.0,
      "completions/mean_length": 205.65,
      "completions/mean_terminated_length": 118.0,
      "completions/min_length": 155.0,
      "completions/min_terminated_length": 118.0,
      "entropy": 0.6852329360786825,
      "epoch": 1.4375,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 0.13374075293540955,
      "kl": 0.27822317034006117,
      "learning_rate": 4.2000000000000006e-07,
      "loss": -0.07140348553657531,
      "num_tokens": 756746.0,
      "reward": 1.137933337688446,
      "reward_std": 0.24845077395439147,
      "rewards/reward_func/mean": 1.137933337688446,
      "rewards/reward_func/std": 0.24845077395439147,
      "step": 230,
      "step_time": 16.51037070681341,
      "tools/call_frequency": 2.65,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.5,
      "completions/max_length": 229.8,
      "completions/max_terminated_length": 215.8,
      "completions/mean_length": 214.75,
      "completions/mean_terminated_length": 210.7,
      "completions/min_length": 204.4,
      "completions/min_terminated_length": 206.6,
      "entropy": 0.8697535959538072,
      "epoch": 1.46875,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 2.334895133972168,
      "kl": 0.2648452676832676,
      "learning_rate": 3.2e-07,
      "loss": 0.014858978986740112,
      "num_tokens": 773619.0,
      "reward": 0.7043333411216736,
      "reward_std": 0.402424693107605,
      "rewards/reward_func/mean": 0.7043333411216736,
      "rewards/reward_func/std": 0.402424693107605,
      "step": 235,
      "step_time": 16.278191728607634,
      "tools/call_frequency": 3.2,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.7,
      "completions/max_length": 234.0,
      "completions/max_terminated_length": 148.6,
      "completions/mean_length": 207.45,
      "completions/mean_terminated_length": 142.8,
      "completions/min_length": 165.0,
      "completions/min_terminated_length": 137.0,
      "entropy": 1.037268871301785,
      "epoch": 1.5,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 5.430893421173096,
      "kl": 0.2893723286688328,
      "learning_rate": 2.2e-07,
      "loss": -0.0661659598350525,
      "num_tokens": 789973.0,
      "reward": 0.7955000281333924,
      "reward_std": 0.6784387767314911,
      "rewards/reward_func/mean": 0.7955000281333924,
      "rewards/reward_func/std": 0.6784387946128845,
      "step": 240,
      "step_time": 17.391802603405086,
      "tools/call_frequency": 3.05,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.8,
      "completions/max_length": 231.2,
      "completions/max_terminated_length": 138.6,
      "completions/mean_length": 219.2,
      "completions/mean_terminated_length": 136.3,
      "completions/min_length": 211.2,
      "completions/min_terminated_length": 134.0,
      "entropy": 0.47988032912835477,
      "epoch": 1.53125,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 0.18447190523147583,
      "kl": 0.23510150127112867,
      "learning_rate": 1.2000000000000002e-07,
      "loss": 0.005726324021816253,
      "num_tokens": 806865.0,
      "reward": 1.1196000099182128,
      "reward_std": 0.15212990045547486,
      "rewards/reward_func/mean": 1.1196000099182128,
      "rewards/reward_func/std": 0.15212990045547486,
      "step": 245,
      "step_time": 17.48798955640523,
      "tools/call_frequency": 3.0,
      "tools/failure_frequency": 0.0
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.7,
      "completions/max_length": 224.4,
      "completions/max_terminated_length": 170.0,
      "completions/mean_length": 210.15,
      "completions/mean_terminated_length": 167.0,
      "completions/min_length": 194.4,
      "completions/min_terminated_length": 164.0,
      "entropy": 0.7179721655789763,
      "epoch": 1.5625,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 4.354219436645508,
      "kl": 0.26825664229691026,
      "learning_rate": 2e-08,
      "loss": 0.0007836699485778808,
      "num_tokens": 823319.0,
      "reward": 0.9574000000953674,
      "reward_std": 0.6371999979019165,
      "rewards/reward_func/mean": 0.9574000000953674,
      "rewards/reward_func/std": 0.6371999979019165,
      "step": 250,
      "step_time": 17.161368461008532,
      "tools/call_frequency": 3.3,
      "tools/failure_frequency": 0.0
    }
  ],
  "logging_steps": 5,
  "max_steps": 250,
  "num_input_tokens_seen": 823319,
  "num_train_epochs": 2,
  "save_steps": 125,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}