| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0012106537530266344, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "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": 0.0, | |
| "completions/max_length": 3274.0, | |
| "completions/max_terminated_length": 3274.0, | |
| "completions/mean_length": 1305.6875, | |
| "completions/mean_terminated_length": 1305.6875, | |
| "completions/min_length": 359.0, | |
| "completions/min_terminated_length": 359.0, | |
| "epoch": 6.053268765133172e-06, | |
| "grad_norm": 0.7806517512814045, | |
| "kl": 0.0191497802734375, | |
| "learning_rate": 0.0, | |
| "loss": -0.0018, | |
| "num_tokens": 28431.0, | |
| "reward": 0.10032547265291214, | |
| "reward_std": 0.21521395444869995, | |
| "rewards/correct_answer_reward_func": 0.1875, | |
| "rewards/efficient_thinking_reward_func": 5.08148193359375, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.1875, | |
| "rewards/tool_execution_reward_func": 0.9375, | |
| "step": 1 | |
| }, | |
| { | |
| "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, | |
| "epoch": 1.2106537530266343e-05, | |
| "grad_norm": 0.779905050857545, | |
| "kl": 0.0191497802734375, | |
| "learning_rate": 1.5000000000000002e-07, | |
| "loss": -0.0018, | |
| "step": 2 | |
| }, | |
| { | |
| "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, | |
| "epoch": 1.8159806295399516e-05, | |
| "grad_norm": 0.775895943887529, | |
| "kl": 0.0205230712890625, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": -0.0018, | |
| "step": 3 | |
| }, | |
| { | |
| "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, | |
| "epoch": 2.4213075060532686e-05, | |
| "grad_norm": 0.7054717897002856, | |
| "kl": 0.0189208984375, | |
| "learning_rate": 4.5e-07, | |
| "loss": -0.0018, | |
| "step": 4 | |
| }, | |
| { | |
| "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": 0.0625, | |
| "completions/max_length": 3197.0, | |
| "completions/max_terminated_length": 3197.0, | |
| "completions/mean_length": 1387.5625, | |
| "completions/mean_terminated_length": 1480.0666666666666, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 301.0, | |
| "epoch": 3.026634382566586e-05, | |
| "grad_norm": 2.8398488181877535, | |
| "kl": 0.0939483642578125, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.0005, | |
| "num_tokens": 62348.0, | |
| "reward": 0.0669252872467041, | |
| "reward_std": 0.18236754834651947, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 4.3189802169799805, | |
| "rewards/format_reward_func": 0.9474999904632568, | |
| "rewards/num_xml_reward_func": 1.0, | |
| "rewards/tool_execution_reward_func": 0.875, | |
| "step": 5 | |
| }, | |
| { | |
| "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, | |
| "epoch": 3.631961259079903e-05, | |
| "grad_norm": 0.6955667618263449, | |
| "kl": 0.013031005859375, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.0002, | |
| "step": 6 | |
| }, | |
| { | |
| "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, | |
| "epoch": 4.2372881355932206e-05, | |
| "grad_norm": 0.6839471972703409, | |
| "kl": 0.0131378173828125, | |
| "learning_rate": 9e-07, | |
| "loss": 0.0003, | |
| "step": 7 | |
| }, | |
| { | |
| "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, | |
| "epoch": 4.842615012106537e-05, | |
| "grad_norm": 0.6825085562302083, | |
| "kl": 0.0136871337890625, | |
| "learning_rate": 1.05e-06, | |
| "loss": 0.0003, | |
| "step": 8 | |
| }, | |
| { | |
| "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": 1535.0, | |
| "completions/max_terminated_length": 1535.0, | |
| "completions/mean_length": 1288.375, | |
| "completions/mean_terminated_length": 1288.375, | |
| "completions/min_length": 501.0, | |
| "completions/min_terminated_length": 501.0, | |
| "epoch": 5.4479418886198546e-05, | |
| "grad_norm": 2.1585776860330728, | |
| "kl": 0.01605224609375, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 0.0002, | |
| "num_tokens": 90542.0, | |
| "reward": 0.004091441631317139, | |
| "reward_std": 0.4142349064350128, | |
| "rewards/correct_answer_reward_func": 0.4375, | |
| "rewards/efficient_thinking_reward_func": 4.875295639038086, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.15625, | |
| "rewards/tool_execution_reward_func": 0.9375, | |
| "step": 9 | |
| }, | |
| { | |
| "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, | |
| "epoch": 6.053268765133172e-05, | |
| "grad_norm": 2.086709029627379, | |
| "kl": 0.030364990234375, | |
| "learning_rate": 1.35e-06, | |
| "loss": 0.0003, | |
| "step": 10 | |
| }, | |
| { | |
| "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, | |
| "epoch": 6.658595641646489e-05, | |
| "grad_norm": 2.055694134933805, | |
| "kl": 0.035125732421875, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.0003, | |
| "step": 11 | |
| }, | |
| { | |
| "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, | |
| "epoch": 7.263922518159807e-05, | |
| "grad_norm": 5.214306021603971, | |
| "kl": 0.24627685546875, | |
| "learning_rate": 1.65e-06, | |
| "loss": 0.0009, | |
| "step": 12 | |
| }, | |
| { | |
| "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": 3169.0, | |
| "completions/max_terminated_length": 3169.0, | |
| "completions/mean_length": 1584.5625, | |
| "completions/mean_terminated_length": 1584.5625, | |
| "completions/min_length": 1048.0, | |
| "completions/min_terminated_length": 1048.0, | |
| "epoch": 7.869249394673124e-05, | |
| "grad_norm": 3.0602846246870006, | |
| "kl": 0.2955322265625, | |
| "learning_rate": 1.8e-06, | |
| "loss": 0.0012, | |
| "num_tokens": 123515.0, | |
| "reward": 0.00020614694221876562, | |
| "reward_std": 0.00014354230370372534, | |
| "rewards/correct_answer_reward_func": 0.0, | |
| "rewards/efficient_thinking_reward_func": 4.435483932495117, | |
| "rewards/format_reward_func": 0.9916666746139526, | |
| "rewards/num_xml_reward_func": 1.03125, | |
| "rewards/tool_execution_reward_func": 1.125, | |
| "step": 13 | |
| }, | |
| { | |
| "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, | |
| "epoch": 8.474576271186441e-05, | |
| "grad_norm": 3.24773890393597, | |
| "kl": 0.262451171875, | |
| "learning_rate": 1.95e-06, | |
| "loss": 0.0012, | |
| "step": 14 | |
| }, | |
| { | |
| "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, | |
| "epoch": 9.079903147699757e-05, | |
| "grad_norm": 1.0320633234812377, | |
| "kl": 0.21142578125, | |
| "learning_rate": 2.1e-06, | |
| "loss": 0.0009, | |
| "step": 15 | |
| }, | |
| { | |
| "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, | |
| "epoch": 9.685230024213075e-05, | |
| "grad_norm": 0.6955273697626486, | |
| "kl": 0.19775390625, | |
| "learning_rate": 2.25e-06, | |
| "loss": 0.0008, | |
| "step": 16 | |
| }, | |
| { | |
| "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": 0.0625, | |
| "completions/max_length": 2131.0, | |
| "completions/max_terminated_length": 2131.0, | |
| "completions/mean_length": 1081.8125, | |
| "completions/mean_terminated_length": 1153.9333333333334, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 279.0, | |
| "epoch": 0.00010290556900726392, | |
| "grad_norm": 1.383354124641088, | |
| "kl": 0.18017578125, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.0065, | |
| "num_tokens": 152500.0, | |
| "reward": 0.14100381731987, | |
| "reward_std": 0.44302019476890564, | |
| "rewards/correct_answer_reward_func": 0.5, | |
| "rewards/efficient_thinking_reward_func": 5.360107898712158, | |
| "rewards/format_reward_func": 0.9816666841506958, | |
| "rewards/num_xml_reward_func": 1.0458333492279053, | |
| "rewards/tool_execution_reward_func": 0.75, | |
| "step": 17 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00010895883777239709, | |
| "grad_norm": 1.370303518112428, | |
| "kl": 0.256591796875, | |
| "learning_rate": 2.55e-06, | |
| "loss": 0.0067, | |
| "step": 18 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00011501210653753027, | |
| "grad_norm": 1.3030395065430471, | |
| "kl": 0.32666015625, | |
| "learning_rate": 2.7e-06, | |
| "loss": 0.0069, | |
| "step": 19 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00012106537530266344, | |
| "grad_norm": 1.5799075174968453, | |
| "kl": 0.5322265625, | |
| "learning_rate": 2.85e-06, | |
| "loss": 0.0075, | |
| "step": 20 | |
| }, | |
| { | |
| "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": 0.0625, | |
| "completions/max_length": 2074.0, | |
| "completions/max_terminated_length": 2074.0, | |
| "completions/mean_length": 1281.9375, | |
| "completions/mean_terminated_length": 1367.4, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 420.0, | |
| "epoch": 0.0001271186440677966, | |
| "grad_norm": 38.55336802862083, | |
| "kl": 2.150390625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0073, | |
| "num_tokens": 184727.0, | |
| "reward": 0.03352511301636696, | |
| "reward_std": 0.1334875077009201, | |
| "rewards/correct_answer_reward_func": 0.0625, | |
| "rewards/efficient_thinking_reward_func": 3.6303672790527344, | |
| "rewards/format_reward_func": 0.9537500143051147, | |
| "rewards/num_xml_reward_func": 0.8125, | |
| "rewards/tool_execution_reward_func": 0.9375, | |
| "step": 21 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00013317191283292979, | |
| "grad_norm": 10.045688461429338, | |
| "kl": 0.865234375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0025, | |
| "step": 22 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00013922518159806296, | |
| "grad_norm": 2.9644172349588827, | |
| "kl": 0.8720703125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0022, | |
| "step": 23 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00014527845036319613, | |
| "grad_norm": 2.3601465047763033, | |
| "kl": 0.6787109375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0017, | |
| "step": 24 | |
| }, | |
| { | |
| "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": 1914.0, | |
| "completions/max_terminated_length": 1914.0, | |
| "completions/mean_length": 1091.0625, | |
| "completions/mean_terminated_length": 1091.0625, | |
| "completions/min_length": 227.0, | |
| "completions/min_terminated_length": 227.0, | |
| "epoch": 0.0001513317191283293, | |
| "grad_norm": 9.206773673324696, | |
| "kl": 1.31640625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0148, | |
| "num_tokens": 209764.0, | |
| "reward": -0.007718075066804886, | |
| "reward_std": 0.3225381076335907, | |
| "rewards/correct_answer_reward_func": 0.3125, | |
| "rewards/efficient_thinking_reward_func": 4.085506439208984, | |
| "rewards/format_reward_func": 0.9916666746139526, | |
| "rewards/num_xml_reward_func": 0.9270833730697632, | |
| "rewards/tool_execution_reward_func": 0.6875, | |
| "step": 25 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00015738498789346248, | |
| "grad_norm": 241.05237266393604, | |
| "kl": 1.84228515625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0166, | |
| "step": 26 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00016343825665859565, | |
| "grad_norm": 1.2667482118640077, | |
| "kl": 0.6767578125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0123, | |
| "step": 27 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00016949152542372882, | |
| "grad_norm": 1.066626868681045, | |
| "kl": 0.572265625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0119, | |
| "step": 28 | |
| }, | |
| { | |
| "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": 2413.0, | |
| "completions/max_terminated_length": 2413.0, | |
| "completions/mean_length": 1022.5625, | |
| "completions/mean_terminated_length": 1022.5625, | |
| "completions/min_length": 13.0, | |
| "completions/min_terminated_length": 13.0, | |
| "epoch": 0.000175544794188862, | |
| "grad_norm": 3.376000196555004, | |
| "kl": 0.90673828125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0045, | |
| "num_tokens": 233745.0, | |
| "reward": 0.06693361699581146, | |
| "reward_std": 0.18236428499221802, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 4.435483932495117, | |
| "rewards/format_reward_func": 0.8999999761581421, | |
| "rewards/num_xml_reward_func": 1.0416666269302368, | |
| "rewards/tool_execution_reward_func": 0.6875, | |
| "step": 29 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00018159806295399514, | |
| "grad_norm": 0.3430397854754984, | |
| "kl": 0.67919921875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0051, | |
| "step": 30 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00018765133171912832, | |
| "grad_norm": 0.2377290664264501, | |
| "kl": 0.70751953125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0051, | |
| "step": 31 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0001937046004842615, | |
| "grad_norm": 0.2344163560775934, | |
| "kl": 0.751953125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0051, | |
| "step": 32 | |
| }, | |
| { | |
| "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": 2072.0, | |
| "completions/max_terminated_length": 2072.0, | |
| "completions/mean_length": 1032.0, | |
| "completions/mean_terminated_length": 1032.0, | |
| "completions/min_length": 355.0, | |
| "completions/min_terminated_length": 355.0, | |
| "epoch": 0.00019975786924939466, | |
| "grad_norm": 9.098443242784603, | |
| "kl": 1.21142578125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0035, | |
| "num_tokens": 257837.0, | |
| "reward": 0.13366317749023438, | |
| "reward_std": 0.23877617716789246, | |
| "rewards/correct_answer_reward_func": 0.25, | |
| "rewards/efficient_thinking_reward_func": 4.384251594543457, | |
| "rewards/format_reward_func": 0.987500011920929, | |
| "rewards/num_xml_reward_func": 1.0625, | |
| "rewards/tool_execution_reward_func": 0.625, | |
| "step": 33 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00020581113801452784, | |
| "grad_norm": 2.3617131203251613, | |
| "kl": 0.775390625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0014, | |
| "step": 34 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000211864406779661, | |
| "grad_norm": 0.47446261418389957, | |
| "kl": 0.59765625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0006, | |
| "step": 35 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00021791767554479418, | |
| "grad_norm": 0.34925057861597, | |
| "kl": 0.564453125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0004, | |
| "step": 36 | |
| }, | |
| { | |
| "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": 0.125, | |
| "completions/max_length": 2844.0, | |
| "completions/max_terminated_length": 2844.0, | |
| "completions/mean_length": 1202.9375, | |
| "completions/mean_terminated_length": 1374.7857142857142, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 377.0, | |
| "epoch": 0.00022397094430992736, | |
| "grad_norm": 7.330956289365437, | |
| "kl": 1.2607421875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0044, | |
| "num_tokens": 292896.0, | |
| "reward": 0.03360215947031975, | |
| "reward_std": 0.1334669440984726, | |
| "rewards/correct_answer_reward_func": 0.0625, | |
| "rewards/efficient_thinking_reward_func": 5.241935729980469, | |
| "rewards/format_reward_func": 0.9662500023841858, | |
| "rewards/num_xml_reward_func": 1.1979166269302368, | |
| "rewards/tool_execution_reward_func": 0.9375, | |
| "step": 37 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00023002421307506053, | |
| "grad_norm": 0.549579766985337, | |
| "kl": 0.7529296875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0029, | |
| "step": 38 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0002360774818401937, | |
| "grad_norm": 0.16634021782187236, | |
| "kl": 0.7275390625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0028, | |
| "step": 39 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00024213075060532688, | |
| "grad_norm": 0.1711020261656612, | |
| "kl": 0.716796875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0027, | |
| "step": 40 | |
| }, | |
| { | |
| "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": 2655.0, | |
| "completions/max_terminated_length": 2655.0, | |
| "completions/mean_length": 1266.6875, | |
| "completions/mean_terminated_length": 1266.6875, | |
| "completions/min_length": 403.0, | |
| "completions/min_terminated_length": 403.0, | |
| "epoch": 0.00024818401937046, | |
| "grad_norm": 0.6184470607386672, | |
| "kl": 0.796875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0074, | |
| "num_tokens": 320743.0, | |
| "reward": 0.18172809481620789, | |
| "reward_std": 0.28321510553359985, | |
| "rewards/correct_answer_reward_func": 0.3125, | |
| "rewards/efficient_thinking_reward_func": 6.04838752746582, | |
| "rewards/format_reward_func": 0.987500011920929, | |
| "rewards/num_xml_reward_func": 1.3541666269302368, | |
| "rewards/tool_execution_reward_func": 0.875, | |
| "step": 41 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0002542372881355932, | |
| "grad_norm": 0.4662817750620152, | |
| "kl": 0.775390625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0075, | |
| "step": 42 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00026029055690072637, | |
| "grad_norm": 0.48985485884822744, | |
| "kl": 0.7978515625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0074, | |
| "step": 43 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00026634382566585957, | |
| "grad_norm": 0.5614378811652494, | |
| "kl": 0.83984375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0073, | |
| "step": 44 | |
| }, | |
| { | |
| "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": 0.125, | |
| "completions/max_length": 3370.0, | |
| "completions/max_terminated_length": 3370.0, | |
| "completions/mean_length": 1417.3125, | |
| "completions/mean_terminated_length": 1619.7857142857142, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1053.0, | |
| "epoch": 0.0002723970944309927, | |
| "grad_norm": 57880.63241282136, | |
| "kl": 1502.919921875, | |
| "learning_rate": 3e-06, | |
| "loss": 10.1932, | |
| "num_tokens": 359232.0, | |
| "reward": 0.06695549190044403, | |
| "reward_std": 0.18235576152801514, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 5.443548679351807, | |
| "rewards/format_reward_func": 0.987500011920929, | |
| "rewards/num_xml_reward_func": 1.1510416269302368, | |
| "rewards/tool_execution_reward_func": 1.1875, | |
| "step": 45 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0002784503631961259, | |
| "grad_norm": 2894.0307559416137, | |
| "kl": 157.4794921875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.3165, | |
| "step": 46 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00028450363196125906, | |
| "grad_norm": 37.029243795631096, | |
| "kl": 3.7021484375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0144, | |
| "step": 47 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00029055690072639226, | |
| "grad_norm": 12.662485260083166, | |
| "kl": 2.4423828125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0108, | |
| "step": 48 | |
| }, | |
| { | |
| "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": 3108.0, | |
| "completions/max_terminated_length": 3108.0, | |
| "completions/mean_length": 1516.75, | |
| "completions/mean_terminated_length": 1516.75, | |
| "completions/min_length": 1020.0, | |
| "completions/min_terminated_length": 1020.0, | |
| "epoch": 0.0002966101694915254, | |
| "grad_norm": 7845.857745835188, | |
| "kl": 527.46875, | |
| "learning_rate": 3e-06, | |
| "loss": 1.7351, | |
| "num_tokens": 391080.0, | |
| "reward": 0.1431797295808792, | |
| "reward_std": 0.3620518743991852, | |
| "rewards/correct_answer_reward_func": 0.375, | |
| "rewards/efficient_thinking_reward_func": 6.04838752746582, | |
| "rewards/format_reward_func": 0.9916666746139526, | |
| "rewards/num_xml_reward_func": 1.34375, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 49 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0003026634382566586, | |
| "grad_norm": 705.0569148996761, | |
| "kl": 47.4765625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.1643, | |
| "step": 50 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00030871670702179176, | |
| "grad_norm": 19.469174009613884, | |
| "kl": 5.75390625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0042, | |
| "step": 51 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00031476997578692496, | |
| "grad_norm": 4.187483088072952, | |
| "kl": 2.0078125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0101, | |
| "step": 52 | |
| }, | |
| { | |
| "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": 3277.0, | |
| "completions/max_terminated_length": 3277.0, | |
| "completions/mean_length": 1857.125, | |
| "completions/mean_terminated_length": 1857.125, | |
| "completions/min_length": 1220.0, | |
| "completions/min_terminated_length": 1220.0, | |
| "epoch": 0.0003208232445520581, | |
| "grad_norm": 15.598878965568742, | |
| "kl": 1.259765625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0128, | |
| "num_tokens": 428414.0, | |
| "reward": 0.06696485728025436, | |
| "reward_std": 0.18235208094120026, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 5.241935729980469, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.1979166269302368, | |
| "rewards/tool_execution_reward_func": 1.0625, | |
| "step": 53 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0003268765133171913, | |
| "grad_norm": 0.46962880382282596, | |
| "kl": 0.845703125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0115, | |
| "step": 54 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00033292978208232445, | |
| "grad_norm": 0.3371767290491545, | |
| "kl": 0.734375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0111, | |
| "step": 55 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00033898305084745765, | |
| "grad_norm": 0.35169159593818344, | |
| "kl": 0.69140625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.011, | |
| "step": 56 | |
| }, | |
| { | |
| "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": 2908.0, | |
| "completions/max_terminated_length": 2908.0, | |
| "completions/mean_length": 1546.0625, | |
| "completions/mean_terminated_length": 1546.0625, | |
| "completions/min_length": 1002.0, | |
| "completions/min_terminated_length": 1002.0, | |
| "epoch": 0.0003450363196125908, | |
| "grad_norm": 157.35744089074626, | |
| "kl": 16.322265625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0504, | |
| "num_tokens": 460731.0, | |
| "reward": 0.19058775901794434, | |
| "reward_std": 0.35528117418289185, | |
| "rewards/correct_answer_reward_func": 0.5, | |
| "rewards/efficient_thinking_reward_func": 5.443548202514648, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.2291666269302368, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 57 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000351089588377724, | |
| "grad_norm": 190.19152750617562, | |
| "kl": 7.6640625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0339, | |
| "step": 58 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00035714285714285714, | |
| "grad_norm": 6.638740120478302, | |
| "kl": 1.9609375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0123, | |
| "step": 59 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0003631961259079903, | |
| "grad_norm": 0.9919110476899615, | |
| "kl": 1.177734375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0097, | |
| "step": 60 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 3020.0, | |
| "completions/max_terminated_length": 3020.0, | |
| "completions/mean_length": 1400.375, | |
| "completions/mean_terminated_length": 1723.5384615384614, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1168.0, | |
| "epoch": 0.0003692493946731235, | |
| "grad_norm": 13723.1434300017, | |
| "kl": 499.875, | |
| "learning_rate": 3e-06, | |
| "loss": 1.6878, | |
| "num_tokens": 503045.0, | |
| "reward": -0.17267484962940216, | |
| "reward_std": 0.3091583847999573, | |
| "rewards/correct_answer_reward_func": 0.25, | |
| "rewards/efficient_thinking_reward_func": 2.4193549156188965, | |
| "rewards/format_reward_func": 0.9662500023841858, | |
| "rewards/num_xml_reward_func": 0.5625, | |
| "rewards/tool_execution_reward_func": 1.125, | |
| "step": 61 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00037530266343825664, | |
| "grad_norm": 583.6315212975338, | |
| "kl": 59.34375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.1944, | |
| "step": 62 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00038135593220338984, | |
| "grad_norm": 6.259462206668412, | |
| "kl": 1.6533203125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0194, | |
| "step": 63 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000387409200968523, | |
| "grad_norm": 1.0697409446207862, | |
| "kl": 0.7666015625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0167, | |
| "step": 64 | |
| }, | |
| { | |
| "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": 0.0625, | |
| "completions/max_length": 2703.0, | |
| "completions/max_terminated_length": 2703.0, | |
| "completions/mean_length": 1332.75, | |
| "completions/mean_terminated_length": 1421.6, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 664.0, | |
| "epoch": 0.0003934624697336562, | |
| "grad_norm": 0.8884534962640673, | |
| "kl": 1.0087890625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0275, | |
| "num_tokens": 536045.0, | |
| "reward": 0.12867216765880585, | |
| "reward_std": 0.4248371720314026, | |
| "rewards/correct_answer_reward_func": 0.5, | |
| "rewards/efficient_thinking_reward_func": 5.460975170135498, | |
| "rewards/format_reward_func": 0.9925000071525574, | |
| "rewards/num_xml_reward_func": 1.2291666269302368, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 65 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00039951573849878933, | |
| "grad_norm": 0.8129440217914238, | |
| "kl": 0.91796875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0278, | |
| "step": 66 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00040556900726392253, | |
| "grad_norm": 2074.331071701581, | |
| "kl": 95.181640625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.275, | |
| "step": 67 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0004116222760290557, | |
| "grad_norm": 25.42750541497437, | |
| "kl": 2.0732421875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0241, | |
| "step": 68 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 1718.0, | |
| "completions/max_terminated_length": 1718.0, | |
| "completions/mean_length": 1171.125, | |
| "completions/mean_terminated_length": 1441.3846153846155, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 800.0, | |
| "epoch": 0.0004176755447941889, | |
| "grad_norm": 3132.1136439947386, | |
| "kl": 258.822265625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.993, | |
| "num_tokens": 574691.0, | |
| "reward": -0.17273107171058655, | |
| "reward_std": 0.3091248571872711, | |
| "rewards/correct_answer_reward_func": 0.25, | |
| "rewards/efficient_thinking_reward_func": 1.2329180240631104, | |
| "rewards/format_reward_func": 0.90625, | |
| "rewards/num_xml_reward_func": 0.28125, | |
| "rewards/tool_execution_reward_func": 0.9375, | |
| "step": 69 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000423728813559322, | |
| "grad_norm": 83.706286199725, | |
| "kl": 14.453125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0731, | |
| "step": 70 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0004297820823244552, | |
| "grad_norm": 13.647323217615291, | |
| "kl": 3.970703125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0323, | |
| "step": 71 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00043583535108958837, | |
| "grad_norm": 1.7482065120313235, | |
| "kl": 1.4990234375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0241, | |
| "step": 72 | |
| }, | |
| { | |
| "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": 0.3125, | |
| "completions/max_length": 2964.0, | |
| "completions/max_terminated_length": 2964.0, | |
| "completions/mean_length": 1056.0, | |
| "completions/mean_terminated_length": 1536.0, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1002.0, | |
| "epoch": 0.00044188861985472157, | |
| "grad_norm": 0.8589483061045617, | |
| "kl": 1.400390625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0032, | |
| "num_tokens": 619647.0, | |
| "reward": 0.12374991923570633, | |
| "reward_std": 0.3325650691986084, | |
| "rewards/correct_answer_reward_func": 0.375, | |
| "rewards/efficient_thinking_reward_func": 3.02419376373291, | |
| "rewards/format_reward_func": 0.875, | |
| "rewards/num_xml_reward_func": 0.6666666269302368, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 73 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0004479418886198547, | |
| "grad_norm": 0.6265176251259559, | |
| "kl": 1.064453125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.004, | |
| "step": 74 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0004539951573849879, | |
| "grad_norm": 0.6501523287860633, | |
| "kl": 1.0078125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0042, | |
| "step": 75 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00046004842615012106, | |
| "grad_norm": 0.6557107326779114, | |
| "kl": 1.041015625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0041, | |
| "step": 76 | |
| }, | |
| { | |
| "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": 0.125, | |
| "completions/max_length": 3914.0, | |
| "completions/max_terminated_length": 3914.0, | |
| "completions/mean_length": 1885.25, | |
| "completions/mean_terminated_length": 2154.5714285714284, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1259.0, | |
| "epoch": 0.00046610169491525426, | |
| "grad_norm": 1426.697250240864, | |
| "kl": 71.76171875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.3336, | |
| "num_tokens": 665623.0, | |
| "reward": -0.08629994839429855, | |
| "reward_std": 0.23610900342464447, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 2.0161290168762207, | |
| "rewards/format_reward_func": 0.9462500214576721, | |
| "rewards/num_xml_reward_func": 0.46875, | |
| "rewards/tool_execution_reward_func": 1.0625, | |
| "step": 77 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0004721549636803874, | |
| "grad_norm": 12.49919870676222, | |
| "kl": 1.5146484375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0013, | |
| "step": 78 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00047820823244552055, | |
| "grad_norm": 1.8191523325467, | |
| "kl": 0.966796875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0021, | |
| "step": 79 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00048426150121065375, | |
| "grad_norm": 0.8614783265852023, | |
| "kl": 0.779296875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.003, | |
| "step": 80 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 3240.0, | |
| "completions/max_terminated_length": 3240.0, | |
| "completions/mean_length": 1146.0625, | |
| "completions/mean_terminated_length": 1410.5384615384614, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 954.0, | |
| "epoch": 0.000490314769975787, | |
| "grad_norm": 1343141904879.566, | |
| "kl": 132607115273.9043, | |
| "learning_rate": 3e-06, | |
| "loss": 187695104.0, | |
| "num_tokens": 703828.0, | |
| "reward": -0.19236184656620026, | |
| "reward_std": 0.43547719717025757, | |
| "rewards/correct_answer_reward_func": 0.5, | |
| "rewards/efficient_thinking_reward_func": 2.1309096813201904, | |
| "rewards/format_reward_func": 0.9219642877578735, | |
| "rewards/num_xml_reward_func": 0.46875, | |
| "rewards/tool_execution_reward_func": 0.9375, | |
| "step": 81 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00049636803874092, | |
| "grad_norm": 7965471.976798021, | |
| "kl": 1048591.578125, | |
| "learning_rate": 3e-06, | |
| "loss": 1488.7067, | |
| "step": 82 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005024213075060532, | |
| "grad_norm": 16601.817266727197, | |
| "kl": 1132.98046875, | |
| "learning_rate": 3e-06, | |
| "loss": 2.5826, | |
| "step": 83 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005084745762711864, | |
| "grad_norm": 3.849730508254259, | |
| "kl": 1.3916015625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0188, | |
| "step": 84 | |
| }, | |
| { | |
| "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": 0.125, | |
| "completions/max_length": 2706.0, | |
| "completions/max_terminated_length": 2706.0, | |
| "completions/mean_length": 1372.625, | |
| "completions/mean_terminated_length": 1568.7142857142858, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 13.0, | |
| "epoch": 0.0005145278450363196, | |
| "grad_norm": 75.5463129972844, | |
| "kl": 41.2880859375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0314, | |
| "num_tokens": 741602.0, | |
| "reward": -0.09613406658172607, | |
| "reward_std": 0.323569655418396, | |
| "rewards/correct_answer_reward_func": 0.25, | |
| "rewards/efficient_thinking_reward_func": 2.0161290168762207, | |
| "rewards/format_reward_func": 0.9337499737739563, | |
| "rewards/num_xml_reward_func": 0.46875, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 85 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005205811138014527, | |
| "grad_norm": 8.361466445717037, | |
| "kl": 11.2041015625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0104, | |
| "step": 86 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005266343825665859, | |
| "grad_norm": 0.4858997056210773, | |
| "kl": 3.3115234375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0073, | |
| "step": 87 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005326876513317191, | |
| "grad_norm": 0.48251870840305755, | |
| "kl": 1.1220703125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0071, | |
| "step": 88 | |
| }, | |
| { | |
| "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": 2879.0, | |
| "completions/max_terminated_length": 2879.0, | |
| "completions/mean_length": 1475.9375, | |
| "completions/mean_terminated_length": 1475.9375, | |
| "completions/min_length": 949.0, | |
| "completions/min_terminated_length": 949.0, | |
| "epoch": 0.0005387409200968523, | |
| "grad_norm": 817.4364276672566, | |
| "kl": 11.91796875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0879, | |
| "num_tokens": 772797.0, | |
| "reward": -0.029312893748283386, | |
| "reward_std": 0.3894536793231964, | |
| "rewards/correct_answer_reward_func": 0.375, | |
| "rewards/efficient_thinking_reward_func": 4.233870983123779, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 0.9479166269302368, | |
| "rewards/tool_execution_reward_func": 1.125, | |
| "step": 89 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005447941888619854, | |
| "grad_norm": 3.3298619813321695, | |
| "kl": 1.474609375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0212, | |
| "step": 90 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005508474576271186, | |
| "grad_norm": 1.1274332258815973, | |
| "kl": 1.1123046875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0196, | |
| "step": 91 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005569007263922518, | |
| "grad_norm": 0.8850067409872743, | |
| "kl": 1.123046875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0196, | |
| "step": 92 | |
| }, | |
| { | |
| "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": 0.125, | |
| "completions/max_length": 2178.0, | |
| "completions/max_terminated_length": 2178.0, | |
| "completions/mean_length": 1189.5625, | |
| "completions/mean_terminated_length": 1359.5, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 13.0, | |
| "epoch": 0.000562953995157385, | |
| "grad_norm": 1.3395550909317473, | |
| "kl": 1.9501953125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0077, | |
| "num_tokens": 807642.0, | |
| "reward": -0.009627980180084705, | |
| "reward_std": 0.2253102958202362, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 4.435483932495117, | |
| "rewards/format_reward_func": 0.8824999928474426, | |
| "rewards/num_xml_reward_func": 1.03125, | |
| "rewards/tool_execution_reward_func": 0.9375, | |
| "step": 93 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005690072639225181, | |
| "grad_norm": 0.5682146184631082, | |
| "kl": 1.765625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0072, | |
| "step": 94 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005750605326876513, | |
| "grad_norm": 0.28055748126607594, | |
| "kl": 1.4423828125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0067, | |
| "step": 95 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005811138014527845, | |
| "grad_norm": 0.2679104384917441, | |
| "kl": 1.2880859375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0067, | |
| "step": 96 | |
| }, | |
| { | |
| "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": 0.125, | |
| "completions/max_length": 2370.0, | |
| "completions/max_terminated_length": 2370.0, | |
| "completions/mean_length": 1182.8125, | |
| "completions/mean_terminated_length": 1351.7857142857142, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 942.0, | |
| "epoch": 0.0005871670702179177, | |
| "grad_norm": 74.74923285644503, | |
| "kl": 5.83984375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0258, | |
| "num_tokens": 842339.0, | |
| "reward": 0.15201856195926666, | |
| "reward_std": 0.4216999411582947, | |
| "rewards/correct_answer_reward_func": 0.5625, | |
| "rewards/efficient_thinking_reward_func": 5.040322303771973, | |
| "rewards/format_reward_func": 0.9916666746139526, | |
| "rewards/num_xml_reward_func": 1.1145833730697632, | |
| "rewards/tool_execution_reward_func": 1.1875, | |
| "step": 97 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0005932203389830508, | |
| "grad_norm": 6.21548581436623, | |
| "kl": 2.208984375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0066, | |
| "step": 98 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000599273607748184, | |
| "grad_norm": 0.9670438333546869, | |
| "kl": 1.333984375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0034, | |
| "step": 99 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006053268765133172, | |
| "grad_norm": 0.9705895262826928, | |
| "kl": 1.177734375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0029, | |
| "step": 100 | |
| }, | |
| { | |
| "clip_ratio/high_max": NaN, | |
| "clip_ratio/high_mean": NaN, | |
| "clip_ratio/low_mean": NaN, | |
| "clip_ratio/low_min": NaN, | |
| "clip_ratio/region_mean": NaN, | |
| "completions/clipped_ratio": 0.3125, | |
| "completions/max_length": 2850.0, | |
| "completions/max_terminated_length": 2850.0, | |
| "completions/mean_length": 1026.1875, | |
| "completions/mean_terminated_length": 1492.6363636363637, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1078.0, | |
| "epoch": 0.0006113801452784503, | |
| "grad_norm": 0.34914402483189466, | |
| "kl": NaN, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0038, | |
| "num_tokens": 886858.0, | |
| "reward": -0.009644638746976852, | |
| "reward_std": 0.22530952095985413, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 5.01161003112793, | |
| "rewards/format_reward_func": 0.949999988079071, | |
| "rewards/num_xml_reward_func": 0.9479166269302368, | |
| "rewards/tool_execution_reward_func": 1.125, | |
| "step": 101 | |
| }, | |
| { | |
| "clip_ratio/high_max": NaN, | |
| "clip_ratio/high_mean": NaN, | |
| "clip_ratio/low_mean": NaN, | |
| "clip_ratio/low_min": NaN, | |
| "clip_ratio/region_mean": NaN, | |
| "epoch": 0.0006174334140435835, | |
| "grad_norm": 0.409729728396783, | |
| "kl": NaN, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0039, | |
| "step": 102 | |
| }, | |
| { | |
| "clip_ratio/high_max": NaN, | |
| "clip_ratio/high_mean": NaN, | |
| "clip_ratio/low_mean": NaN, | |
| "clip_ratio/low_min": NaN, | |
| "clip_ratio/region_mean": NaN, | |
| "epoch": 0.0006234866828087167, | |
| "grad_norm": 34.71662901286587, | |
| "kl": NaN, | |
| "learning_rate": 3e-06, | |
| "loss": 0.016, | |
| "step": 103 | |
| }, | |
| { | |
| "clip_ratio/high_max": NaN, | |
| "clip_ratio/high_mean": NaN, | |
| "clip_ratio/low_mean": NaN, | |
| "clip_ratio/low_min": NaN, | |
| "clip_ratio/region_mean": NaN, | |
| "epoch": 0.0006295399515738499, | |
| "grad_norm": 178.12705241502368, | |
| "kl": NaN, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0662, | |
| "step": 104 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 2669.0, | |
| "completions/max_terminated_length": 2669.0, | |
| "completions/mean_length": 1197.0625, | |
| "completions/mean_terminated_length": 1473.3076923076924, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1011.0, | |
| "epoch": 0.000635593220338983, | |
| "grad_norm": 113656977492.3838, | |
| "kl": 4060086272.1398926, | |
| "learning_rate": 3e-06, | |
| "loss": 26552720.0, | |
| "num_tokens": 925879.0, | |
| "reward": 0.12848791480064392, | |
| "reward_std": 0.3391018807888031, | |
| "rewards/correct_answer_reward_func": 0.375, | |
| "rewards/efficient_thinking_reward_func": 6.138603687286377, | |
| "rewards/format_reward_func": 0.9183332920074463, | |
| "rewards/num_xml_reward_func": 1.1041667461395264, | |
| "rewards/tool_execution_reward_func": 1.0625, | |
| "step": 105 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006416464891041162, | |
| "grad_norm": 862090008.796202, | |
| "kl": 29097984.143554688, | |
| "learning_rate": 3e-06, | |
| "loss": 190154.0312, | |
| "step": 106 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006476997578692494, | |
| "grad_norm": 5971.671602768365, | |
| "kl": 225.146484375, | |
| "learning_rate": 3e-06, | |
| "loss": 1.4726, | |
| "step": 107 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006537530266343826, | |
| "grad_norm": 197.81456021441926, | |
| "kl": 10.714111328125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.071, | |
| "step": 108 | |
| }, | |
| { | |
| "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": 0.3125, | |
| "completions/max_length": 2507.0, | |
| "completions/max_terminated_length": 2507.0, | |
| "completions/mean_length": 1018.3125, | |
| "completions/mean_terminated_length": 1481.1818181818182, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 920.0, | |
| "epoch": 0.0006598062953995157, | |
| "grad_norm": 211.32333218858383, | |
| "kl": 29.7578125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0762, | |
| "num_tokens": 970272.0, | |
| "reward": -0.05291781574487686, | |
| "reward_std": 0.282351553440094, | |
| "rewards/correct_answer_reward_func": 0.1875, | |
| "rewards/efficient_thinking_reward_func": 4.623542785644531, | |
| "rewards/format_reward_func": 0.9093055725097656, | |
| "rewards/num_xml_reward_func": 0.5659722089767456, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 109 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006658595641646489, | |
| "grad_norm": 47.690518846315975, | |
| "kl": 5.3515625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0217, | |
| "step": 110 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006719128329297821, | |
| "grad_norm": 3.1184302707951286, | |
| "kl": 1.0947265625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0071, | |
| "step": 111 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006779661016949153, | |
| "grad_norm": 0.5838887307984975, | |
| "kl": 0.3798828125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0055, | |
| "step": 112 | |
| }, | |
| { | |
| "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": 0.25, | |
| "completions/max_length": 2388.0, | |
| "completions/max_terminated_length": 2388.0, | |
| "completions/mean_length": 976.4375, | |
| "completions/mean_terminated_length": 1301.9166666666667, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 522.0, | |
| "epoch": 0.0006840193704600484, | |
| "grad_norm": 2.279673569786722, | |
| "kl": 0.669921875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0027, | |
| "num_tokens": 1009859.0, | |
| "reward": -0.08624996989965439, | |
| "reward_std": 0.2361285239458084, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 5.102198123931885, | |
| "rewards/format_reward_func": 0.9054166674613953, | |
| "rewards/num_xml_reward_func": 0.71875, | |
| "rewards/tool_execution_reward_func": 0.90625, | |
| "step": 113 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006900726392251816, | |
| "grad_norm": 0.29363313071510083, | |
| "kl": 0.310546875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0017, | |
| "step": 114 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0006961259079903148, | |
| "grad_norm": 0.25416521021234684, | |
| "kl": 0.2568359375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0015, | |
| "step": 115 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000702179176755448, | |
| "grad_norm": 0.2532709696883375, | |
| "kl": 0.25048828125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0015, | |
| "step": 116 | |
| }, | |
| { | |
| "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": 0.375, | |
| "completions/max_length": 2938.0, | |
| "completions/max_terminated_length": 2938.0, | |
| "completions/mean_length": 825.1875, | |
| "completions/mean_terminated_length": 1320.3, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 265.0, | |
| "epoch": 0.0007082324455205811, | |
| "grad_norm": 0.06499596334850685, | |
| "kl": 0.195556640625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0004, | |
| "num_tokens": 1055258.0, | |
| "reward": 0.00015617192548234016, | |
| "reward_std": 0.000146350241266191, | |
| "rewards/correct_answer_reward_func": 0.0, | |
| "rewards/efficient_thinking_reward_func": 5.059846878051758, | |
| "rewards/format_reward_func": 0.8541666865348816, | |
| "rewards/num_xml_reward_func": 0.78125, | |
| "rewards/tool_execution_reward_func": 1.0625, | |
| "step": 117 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007142857142857143, | |
| "grad_norm": 0.12561554187034915, | |
| "kl": 0.206298828125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0004, | |
| "step": 118 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007203389830508475, | |
| "grad_norm": 0.03335194268511418, | |
| "kl": 0.185546875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0004, | |
| "step": 119 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007263922518159806, | |
| "grad_norm": 0.016335498310956653, | |
| "kl": 0.1824951171875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0004, | |
| "step": 120 | |
| }, | |
| { | |
| "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": 0.3125, | |
| "completions/max_length": 3041.0, | |
| "completions/max_terminated_length": 3041.0, | |
| "completions/mean_length": 1208.6875, | |
| "completions/mean_terminated_length": 1758.090909090909, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 237.0, | |
| "epoch": 0.0007324455205811138, | |
| "grad_norm": 3074.215567652929, | |
| "kl": 87.255126953125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.4728, | |
| "num_tokens": 1102657.0, | |
| "reward": 0.09965705871582031, | |
| "reward_std": 0.3296266794204712, | |
| "rewards/correct_answer_reward_func": 0.3125, | |
| "rewards/efficient_thinking_reward_func": 3.1649727821350098, | |
| "rewards/format_reward_func": 0.9187500476837158, | |
| "rewards/num_xml_reward_func": 0.5104166865348816, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 121 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000738498789346247, | |
| "grad_norm": 179.50807324392667, | |
| "kl": 30.089599609375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.1486, | |
| "step": 122 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007445520581113802, | |
| "grad_norm": 20.649024610652848, | |
| "kl": 3.32763671875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0155, | |
| "step": 123 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007506053268765133, | |
| "grad_norm": 2.0418855369195485, | |
| "kl": 0.590087890625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0023, | |
| "step": 124 | |
| }, | |
| { | |
| "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": 0.3125, | |
| "completions/max_length": 1461.0, | |
| "completions/max_terminated_length": 1461.0, | |
| "completions/mean_length": 844.125, | |
| "completions/mean_terminated_length": 1227.8181818181818, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 919.0, | |
| "epoch": 0.0007566585956416465, | |
| "grad_norm": 1.2107206967986066, | |
| "kl": 0.489501953125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0052, | |
| "num_tokens": 1144263.0, | |
| "reward": -0.043053146451711655, | |
| "reward_std": 0.1728256642818451, | |
| "rewards/correct_answer_reward_func": 0.0625, | |
| "rewards/efficient_thinking_reward_func": 3.937290668487549, | |
| "rewards/format_reward_func": 0.877500057220459, | |
| "rewards/num_xml_reward_func": 0.71875, | |
| "rewards/tool_execution_reward_func": 0.9166666865348816, | |
| "step": 125 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007627118644067797, | |
| "grad_norm": 0.3153399141642928, | |
| "kl": 0.267333984375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0046, | |
| "step": 126 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007687651331719129, | |
| "grad_norm": 0.30940853761801884, | |
| "kl": 0.25048828125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0046, | |
| "step": 127 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000774818401937046, | |
| "grad_norm": 0.41896360385565806, | |
| "kl": 0.291259765625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0048, | |
| "step": 128 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 2040.0, | |
| "completions/max_terminated_length": 2040.0, | |
| "completions/mean_length": 1019.8125, | |
| "completions/mean_terminated_length": 1255.1538461538462, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 915.0, | |
| "epoch": 0.0007808716707021792, | |
| "grad_norm": 3.046991653902829, | |
| "kl": 1.09619140625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0034, | |
| "num_tokens": 1180448.0, | |
| "reward": 0.023746376857161522, | |
| "reward_std": 0.263210654258728, | |
| "rewards/correct_answer_reward_func": 0.1875, | |
| "rewards/efficient_thinking_reward_func": 5.699418067932129, | |
| "rewards/format_reward_func": 0.9104167222976685, | |
| "rewards/num_xml_reward_func": 1.089583396911621, | |
| "rewards/tool_execution_reward_func": 0.8541666865348816, | |
| "step": 129 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007869249394673124, | |
| "grad_norm": 0.6371118649367347, | |
| "kl": 0.56591796875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0022, | |
| "step": 130 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007929782082324456, | |
| "grad_norm": 0.7977652642661696, | |
| "kl": 0.475341796875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.002, | |
| "step": 131 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0007990314769975787, | |
| "grad_norm": 0.3954173262150925, | |
| "kl": 0.418212890625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0019, | |
| "step": 132 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 2829.0, | |
| "completions/max_terminated_length": 2829.0, | |
| "completions/mean_length": 1284.6875, | |
| "completions/mean_terminated_length": 1581.1538461538462, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 946.0, | |
| "epoch": 0.0008050847457627119, | |
| "grad_norm": 0.16569380056333147, | |
| "kl": 0.203369140625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0001, | |
| "num_tokens": 1220839.0, | |
| "reward": 0.06701274961233139, | |
| "reward_std": 0.18233336508274078, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 8.0274658203125, | |
| "rewards/format_reward_func": 0.96833336353302, | |
| "rewards/num_xml_reward_func": 1.4375, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 133 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008111380145278451, | |
| "grad_norm": 0.16664121027946213, | |
| "kl": 0.20068359375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0001, | |
| "step": 134 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008171912832929783, | |
| "grad_norm": 0.16072020293930506, | |
| "kl": 0.194091796875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0001, | |
| "step": 135 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008232445520581114, | |
| "grad_norm": 41.76166861124815, | |
| "kl": 1.373046875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0071, | |
| "step": 136 | |
| }, | |
| { | |
| "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": 3684.0, | |
| "completions/max_terminated_length": 3684.0, | |
| "completions/mean_length": 1599.125, | |
| "completions/mean_terminated_length": 1599.125, | |
| "completions/min_length": 589.0, | |
| "completions/min_terminated_length": 589.0, | |
| "epoch": 0.0008292978208232446, | |
| "grad_norm": 646644.6699956892, | |
| "kl": 49842.197265625, | |
| "learning_rate": 3e-06, | |
| "loss": 229.4684, | |
| "num_tokens": 1254025.0, | |
| "reward": 0.013887053355574608, | |
| "reward_std": 0.34718748927116394, | |
| "rewards/correct_answer_reward_func": 0.3125, | |
| "rewards/efficient_thinking_reward_func": 5.241935729980469, | |
| "rewards/format_reward_func": 0.9479166269302368, | |
| "rewards/num_xml_reward_func": 0.9635417461395264, | |
| "rewards/tool_execution_reward_func": 0.875, | |
| "step": 137 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008353510895883778, | |
| "grad_norm": 27876.102822783414, | |
| "kl": 2136.07666015625, | |
| "learning_rate": 3e-06, | |
| "loss": 12.8243, | |
| "step": 138 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008414043583535108, | |
| "grad_norm": 790.2979277928656, | |
| "kl": 118.16162109375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.4985, | |
| "step": 139 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000847457627118644, | |
| "grad_norm": 47.030265324687, | |
| "kl": 7.62548828125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0346, | |
| "step": 140 | |
| }, | |
| { | |
| "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": 0.125, | |
| "completions/max_length": 3041.0, | |
| "completions/max_terminated_length": 3041.0, | |
| "completions/mean_length": 1098.375, | |
| "completions/mean_terminated_length": 1255.2857142857142, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 325.0, | |
| "epoch": 0.0008535108958837772, | |
| "grad_norm": 27163.40903493807, | |
| "kl": 1403.2744140625, | |
| "learning_rate": 3e-06, | |
| "loss": 7.8279, | |
| "num_tokens": 1287339.0, | |
| "reward": -0.00962173379957676, | |
| "reward_std": 0.2253105789422989, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 4.435483932495117, | |
| "rewards/format_reward_func": 0.9901785850524902, | |
| "rewards/num_xml_reward_func": 1.0625, | |
| "rewards/tool_execution_reward_func": 0.875, | |
| "step": 141 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008595641646489104, | |
| "grad_norm": 104.2213070575419, | |
| "kl": 5.6044921875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0309, | |
| "step": 142 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008656174334140435, | |
| "grad_norm": 5.517418949203676, | |
| "kl": 1.123046875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0054, | |
| "step": 143 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008716707021791767, | |
| "grad_norm": 0.5745631831412419, | |
| "kl": 0.42724609375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0033, | |
| "step": 144 | |
| }, | |
| { | |
| "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": 0.0625, | |
| "completions/max_length": 2668.0, | |
| "completions/max_terminated_length": 2668.0, | |
| "completions/mean_length": 1254.25, | |
| "completions/mean_terminated_length": 1337.8666666666666, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 377.0, | |
| "epoch": 0.0008777239709443099, | |
| "grad_norm": 11.985045430350514, | |
| "kl": 1.40576171875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0054, | |
| "num_tokens": 1319103.0, | |
| "reward": 0.18167813122272491, | |
| "reward_std": 0.28324925899505615, | |
| "rewards/correct_answer_reward_func": 0.3125, | |
| "rewards/efficient_thinking_reward_func": 5.79470682144165, | |
| "rewards/format_reward_func": 0.9524999856948853, | |
| "rewards/num_xml_reward_func": 1.1041666269302368, | |
| "rewards/tool_execution_reward_func": 0.8125, | |
| "step": 145 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008837772397094431, | |
| "grad_norm": 0.7668036294329545, | |
| "kl": 0.444580078125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0087, | |
| "step": 146 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008898305084745762, | |
| "grad_norm": 0.3955664525228031, | |
| "kl": 0.291259765625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0093, | |
| "step": 147 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0008958837772397094, | |
| "grad_norm": 0.3609851056156834, | |
| "kl": 0.26123046875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0094, | |
| "step": 148 | |
| }, | |
| { | |
| "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": 0.4375, | |
| "completions/max_length": 2821.0, | |
| "completions/max_terminated_length": 2821.0, | |
| "completions/mean_length": 717.125, | |
| "completions/mean_terminated_length": 1274.888888888889, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 228.0, | |
| "epoch": 0.0009019370460048426, | |
| "grad_norm": 0.10668119884135797, | |
| "kl": 0.25, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0054, | |
| "num_tokens": 1366797.0, | |
| "reward": 0.03348555043339729, | |
| "reward_std": 0.13349804282188416, | |
| "rewards/correct_answer_reward_func": 0.0625, | |
| "rewards/efficient_thinking_reward_func": 4.775701522827148, | |
| "rewards/format_reward_func": 0.8970834016799927, | |
| "rewards/num_xml_reward_func": 0.6145833730697632, | |
| "rewards/tool_execution_reward_func": 0.800000011920929, | |
| "step": 149 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009079903147699758, | |
| "grad_norm": 0.07581171591865946, | |
| "kl": 0.21337890625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0055, | |
| "step": 150 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009140435835351089, | |
| "grad_norm": 0.07368344797048941, | |
| "kl": 0.2066650390625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0055, | |
| "step": 151 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009200968523002421, | |
| "grad_norm": 0.06383190574866472, | |
| "kl": 0.2005615234375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0055, | |
| "step": 152 | |
| }, | |
| { | |
| "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": 0.125, | |
| "completions/max_length": 2775.0, | |
| "completions/max_terminated_length": 2775.0, | |
| "completions/mean_length": 1244.6875, | |
| "completions/mean_terminated_length": 1422.5, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 381.0, | |
| "epoch": 0.0009261501210653753, | |
| "grad_norm": 0.4120015190644857, | |
| "kl": 0.250732421875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0169, | |
| "num_tokens": 1402504.0, | |
| "reward": 0.07177050411701202, | |
| "reward_std": 0.3221079111099243, | |
| "rewards/correct_answer_reward_func": 0.25, | |
| "rewards/efficient_thinking_reward_func": 5.607255458831787, | |
| "rewards/format_reward_func": 0.9775000214576721, | |
| "rewards/num_xml_reward_func": 1.1770832538604736, | |
| "rewards/tool_execution_reward_func": 0.875, | |
| "step": 153 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009322033898305085, | |
| "grad_norm": 0.4391974629991036, | |
| "kl": 0.245361328125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0169, | |
| "step": 154 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009382566585956416, | |
| "grad_norm": 0.3734307496972298, | |
| "kl": 0.372314453125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0166, | |
| "step": 155 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009443099273607748, | |
| "grad_norm": 5.409050230753026, | |
| "kl": 1.83984375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0136, | |
| "step": 156 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 2702.0, | |
| "completions/max_terminated_length": 2702.0, | |
| "completions/mean_length": 975.8125, | |
| "completions/mean_terminated_length": 1201.0, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 349.0, | |
| "epoch": 0.000950363196125908, | |
| "grad_norm": 0.7067985121705038, | |
| "kl": 0.344482421875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0048, | |
| "num_tokens": 1437953.0, | |
| "reward": 0.06698047369718552, | |
| "reward_std": 0.18234598636627197, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 5.846774578094482, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.2760416269302368, | |
| "rewards/tool_execution_reward_func": 0.9375, | |
| "step": 157 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009564164648910411, | |
| "grad_norm": 0.11769454310604478, | |
| "kl": 0.2841796875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.005, | |
| "step": 158 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009624697336561743, | |
| "grad_norm": 0.10329568036332133, | |
| "kl": 0.286865234375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.005, | |
| "step": 159 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009685230024213075, | |
| "grad_norm": 4.24103180505241, | |
| "kl": 0.876953125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0042, | |
| "step": 160 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 2661.0, | |
| "completions/max_terminated_length": 2661.0, | |
| "completions/mean_length": 1132.625, | |
| "completions/mean_terminated_length": 1394.0, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 481.0, | |
| "epoch": 0.0009745762711864407, | |
| "grad_norm": 30399.081494084556, | |
| "kl": 3248.51220703125, | |
| "learning_rate": 3e-06, | |
| "loss": 11.4413, | |
| "num_tokens": 1475963.0, | |
| "reward": 0.0669461190700531, | |
| "reward_std": 0.18235941231250763, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 5.576177597045898, | |
| "rewards/format_reward_func": 0.9516667127609253, | |
| "rewards/num_xml_reward_func": 1.1041666269302368, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 161 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.000980629539951574, | |
| "grad_norm": 301.320864829815, | |
| "kl": 42.82861328125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.1518, | |
| "step": 162 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0009866828087167071, | |
| "grad_norm": 7.516991078245809, | |
| "kl": 2.22998046875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0084, | |
| "step": 163 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00099273607748184, | |
| "grad_norm": 0.9933703066829184, | |
| "kl": 0.6552734375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0029, | |
| "step": 164 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 2659.0, | |
| "completions/max_terminated_length": 2659.0, | |
| "completions/mean_length": 1132.6875, | |
| "completions/mean_terminated_length": 1394.076923076923, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1039.0, | |
| "epoch": 0.0009987893462469733, | |
| "grad_norm": 216588358.58769044, | |
| "kl": 14798848.96875, | |
| "learning_rate": 3e-06, | |
| "loss": 39560.0, | |
| "num_tokens": 1513922.0, | |
| "reward": 0.06703699380159378, | |
| "reward_std": 0.1823238879442215, | |
| "rewards/correct_answer_reward_func": 0.125, | |
| "rewards/efficient_thinking_reward_func": 8.870967864990234, | |
| "rewards/format_reward_func": 0.9958333373069763, | |
| "rewards/num_xml_reward_func": 1.5587797164916992, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 165 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010048426150121065, | |
| "grad_norm": 20.66026628810014, | |
| "kl": 2.27978515625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0026, | |
| "step": 166 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010108958837772397, | |
| "grad_norm": 0.5852549123131363, | |
| "kl": 0.49462890625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0021, | |
| "step": 167 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.001016949152542373, | |
| "grad_norm": 10.985658110563065, | |
| "kl": 1.380859375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0001, | |
| "step": 168 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 2760.0, | |
| "completions/max_terminated_length": 2760.0, | |
| "completions/mean_length": 1299.5625, | |
| "completions/mean_terminated_length": 1599.4615384615386, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 995.0, | |
| "epoch": 0.001023002421307506, | |
| "grad_norm": 3.1833824255736856, | |
| "kl": 0.49853515625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0108, | |
| "num_tokens": 1554603.0, | |
| "reward": 0.1483658403158188, | |
| "reward_std": 0.27007395029067993, | |
| "rewards/correct_answer_reward_func": 0.25, | |
| "rewards/efficient_thinking_reward_func": 8.19877815246582, | |
| "rewards/format_reward_func": 0.9624999761581421, | |
| "rewards/num_xml_reward_func": 1.3562500476837158, | |
| "rewards/tool_execution_reward_func": 1.0625, | |
| "step": 169 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010290556900726393, | |
| "grad_norm": 0.43961018475161184, | |
| "kl": 0.41162109375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0113, | |
| "step": 170 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010351089588377725, | |
| "grad_norm": 1.4869647567424702, | |
| "kl": 0.330078125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0114, | |
| "step": 171 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010411622276029055, | |
| "grad_norm": 1.5593896451086025, | |
| "kl": 0.34423828125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0114, | |
| "step": 172 | |
| }, | |
| { | |
| "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": 0.25, | |
| "completions/max_length": 2658.0, | |
| "completions/max_terminated_length": 2658.0, | |
| "completions/mean_length": 1120.4375, | |
| "completions/mean_terminated_length": 1493.9166666666667, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1001.0, | |
| "epoch": 0.0010472154963680387, | |
| "grad_norm": 9.153186896120884, | |
| "kl": 1.18212890625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0004, | |
| "num_tokens": 1596462.0, | |
| "reward": 0.10031714290380478, | |
| "reward_std": 0.21521808207035065, | |
| "rewards/correct_answer_reward_func": 0.1875, | |
| "rewards/efficient_thinking_reward_func": 5.926279067993164, | |
| "rewards/format_reward_func": 0.9650000333786011, | |
| "rewards/num_xml_reward_func": 1.1458332538604736, | |
| "rewards/tool_execution_reward_func": 1.0625, | |
| "step": 173 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010532687651331719, | |
| "grad_norm": 0.526258103973936, | |
| "kl": 0.560546875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0026, | |
| "step": 174 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.001059322033898305, | |
| "grad_norm": 0.14211436005078248, | |
| "kl": 0.45556640625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0029, | |
| "step": 175 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010653753026634383, | |
| "grad_norm": 0.14305060184518628, | |
| "kl": 0.46484375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0028, | |
| "step": 176 | |
| }, | |
| { | |
| "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": 3333.0, | |
| "completions/max_terminated_length": 3333.0, | |
| "completions/mean_length": 1780.0, | |
| "completions/mean_terminated_length": 1780.0, | |
| "completions/min_length": 1066.0, | |
| "completions/min_terminated_length": 1066.0, | |
| "epoch": 0.0010714285714285715, | |
| "grad_norm": 0.7121759134327387, | |
| "kl": 0.4970703125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0185, | |
| "num_tokens": 1632542.0, | |
| "reward": 0.05717029049992561, | |
| "reward_std": 0.29225456714630127, | |
| "rewards/correct_answer_reward_func": 0.25, | |
| "rewards/efficient_thinking_reward_func": 7.056451797485352, | |
| "rewards/format_reward_func": 0.9625000357627869, | |
| "rewards/num_xml_reward_func": 1.3958333730697632, | |
| "rewards/tool_execution_reward_func": 1.0625, | |
| "step": 177 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010774818401937047, | |
| "grad_norm": 0.511035937325882, | |
| "kl": 0.3818359375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0179, | |
| "step": 178 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010835351089588377, | |
| "grad_norm": 118.18849489599667, | |
| "kl": 6.0556640625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0429, | |
| "step": 179 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0010895883777239709, | |
| "grad_norm": 123.64288191458726, | |
| "kl": 3.88916015625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0334, | |
| "step": 180 | |
| }, | |
| { | |
| "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": 0.25, | |
| "completions/max_length": 2638.0, | |
| "completions/max_terminated_length": 2638.0, | |
| "completions/mean_length": 1067.25, | |
| "completions/mean_terminated_length": 1423.0, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1119.0, | |
| "epoch": 0.001095641646489104, | |
| "grad_norm": 0.44662061313508766, | |
| "kl": 0.8115234375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0038, | |
| "num_tokens": 1673550.0, | |
| "reward": 0.10039953887462616, | |
| "reward_std": 0.21517714858055115, | |
| "rewards/correct_answer_reward_func": 0.1875, | |
| "rewards/efficient_thinking_reward_func": 9.27419376373291, | |
| "rewards/format_reward_func": 0.9892857074737549, | |
| "rewards/num_xml_reward_func": 1.5580357313156128, | |
| "rewards/tool_execution_reward_func": 0.9791666865348816, | |
| "step": 181 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011016949152542373, | |
| "grad_norm": 0.19064814713811837, | |
| "kl": 0.673828125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0041, | |
| "step": 182 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011077481840193705, | |
| "grad_norm": 0.12530575938029517, | |
| "kl": 0.578125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0044, | |
| "step": 183 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011138014527845037, | |
| "grad_norm": 0.14053997598107024, | |
| "kl": 0.53515625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0045, | |
| "step": 184 | |
| }, | |
| { | |
| "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": 0.0625, | |
| "completions/max_length": 2780.0, | |
| "completions/max_terminated_length": 2780.0, | |
| "completions/mean_length": 1584.5625, | |
| "completions/mean_terminated_length": 1690.2, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1083.0, | |
| "epoch": 0.0011198547215496369, | |
| "grad_norm": 0.36922212655466874, | |
| "kl": 0.328125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0082, | |
| "num_tokens": 1710599.0, | |
| "reward": 0.26312950253486633, | |
| "reward_std": 0.31617864966392517, | |
| "rewards/correct_answer_reward_func": 0.4375, | |
| "rewards/efficient_thinking_reward_func": 8.467741966247559, | |
| "rewards/format_reward_func": 0.9862499833106995, | |
| "rewards/num_xml_reward_func": 1.5145833492279053, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 185 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.00112590799031477, | |
| "grad_norm": 0.35037508840939274, | |
| "kl": 0.3310546875, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0082, | |
| "step": 186 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.001131961259079903, | |
| "grad_norm": 0.33082477958916895, | |
| "kl": 0.35009765625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0081, | |
| "step": 187 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011380145278450363, | |
| "grad_norm": 0.345834479974622, | |
| "kl": 0.3896484375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0079, | |
| "step": 188 | |
| }, | |
| { | |
| "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": 0.25, | |
| "completions/max_length": 1811.0, | |
| "completions/max_terminated_length": 1811.0, | |
| "completions/mean_length": 1041.8125, | |
| "completions/mean_terminated_length": 1389.0833333333333, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1031.0, | |
| "epoch": 0.0011440677966101695, | |
| "grad_norm": 35.33227872117956, | |
| "kl": 6.01953125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0108, | |
| "num_tokens": 1751200.0, | |
| "reward": 0.10040043294429779, | |
| "reward_std": 0.21517671644687653, | |
| "rewards/correct_answer_reward_func": 0.1875, | |
| "rewards/efficient_thinking_reward_func": 8.064516067504883, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.5625, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 189 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011501210653753027, | |
| "grad_norm": 0.835012686734699, | |
| "kl": 0.9775390625, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0043, | |
| "step": 190 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011561743341404359, | |
| "grad_norm": 0.2586421230474391, | |
| "kl": 0.8486328125, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0047, | |
| "step": 191 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.001162227602905569, | |
| "grad_norm": 0.1651807846836529, | |
| "kl": 0.7734375, | |
| "learning_rate": 3e-06, | |
| "loss": -0.0049, | |
| "step": 192 | |
| }, | |
| { | |
| "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": 0.0625, | |
| "completions/max_length": 4084.0, | |
| "completions/max_terminated_length": 4084.0, | |
| "completions/mean_length": 1537.375, | |
| "completions/mean_terminated_length": 1639.8666666666666, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1045.0, | |
| "epoch": 0.0011682808716707023, | |
| "grad_norm": 243.49244406063949, | |
| "kl": 5.927734375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0565, | |
| "num_tokens": 1787494.0, | |
| "reward": 0.23383872210979462, | |
| "reward_std": 0.2734927237033844, | |
| "rewards/correct_answer_reward_func": 0.4375, | |
| "rewards/efficient_thinking_reward_func": 6.451612949371338, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.5, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 193 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011743341404358355, | |
| "grad_norm": 0.6933472630429098, | |
| "kl": 0.685546875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0218, | |
| "step": 194 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011803874092009684, | |
| "grad_norm": 0.38422473996938145, | |
| "kl": 0.61328125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0216, | |
| "step": 195 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0011864406779661016, | |
| "grad_norm": 0.33054590530199923, | |
| "kl": 0.6220703125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0216, | |
| "step": 196 | |
| }, | |
| { | |
| "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": 0.1875, | |
| "completions/max_length": 3909.0, | |
| "completions/max_terminated_length": 3909.0, | |
| "completions/mean_length": 1388.375, | |
| "completions/mean_terminated_length": 1708.7692307692307, | |
| "completions/min_length": 0.0, | |
| "completions/min_terminated_length": 1148.0, | |
| "epoch": 0.0011924939467312348, | |
| "grad_norm": 188.57473992181284, | |
| "kl": 14.59765625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0927, | |
| "num_tokens": 1829544.0, | |
| "reward": 0.10040343552827835, | |
| "reward_std": 0.21517521142959595, | |
| "rewards/correct_answer_reward_func": 0.1875, | |
| "rewards/efficient_thinking_reward_func": 8.870967864990234, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.5775296688079834, | |
| "rewards/tool_execution_reward_func": 1.0, | |
| "step": 197 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.001198547215496368, | |
| "grad_norm": 1.773591133545115, | |
| "kl": 1.154296875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0063, | |
| "step": 198 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0012046004842615012, | |
| "grad_norm": 0.360837942568022, | |
| "kl": 0.9296875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0049, | |
| "step": 199 | |
| }, | |
| { | |
| "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, | |
| "epoch": 0.0012106537530266344, | |
| "grad_norm": 0.21716087780595142, | |
| "kl": 0.8203125, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0044, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 1829544, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |