| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0014534883720930232, | |
| "eval_steps": 500, | |
| "global_step": 60, | |
| "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": 10086.0, | |
| "completions/max_terminated_length": 10086.0, | |
| "completions/mean_length": 4296.546875, | |
| "completions/mean_terminated_length": 4296.546875, | |
| "completions/min_length": 1720.0, | |
| "completions/min_terminated_length": 1720.0, | |
| "epoch": 2.4224806201550387e-05, | |
| "grad_norm": 0.016954593260394005, | |
| "kl": 0.0009393692016601562, | |
| "learning_rate": 0.0, | |
| "loss": -0.0003, | |
| "num_tokens": 601834.0, | |
| "reward": 0.4602593183517456, | |
| "reward_std": 0.24803586304187775, | |
| "rewards/avg_thinking_length_func": 185.02471923828125, | |
| "rewards/correct_answer_reward_func": 0.453125, | |
| "rewards/efficient_thinking_reward_func": 0.8889554441999474, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.7176268100738525, | |
| "rewards/tool_execution_reward_func": 1.9936248064041138, | |
| "rewards/visit_tool_reward_func": 0.9308543801307678, | |
| "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": 4.8449612403100775e-05, | |
| "grad_norm": 0.016953615886545852, | |
| "kl": 0.0009393692016601562, | |
| "learning_rate": 6.25e-08, | |
| "loss": -0.0003, | |
| "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": 7.267441860465116e-05, | |
| "grad_norm": 0.016864690676516626, | |
| "kl": 0.0009565353393554688, | |
| "learning_rate": 1.25e-07, | |
| "loss": -0.0003, | |
| "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": 9.689922480620155e-05, | |
| "grad_norm": 0.016822420848305722, | |
| "kl": 0.0009622573852539062, | |
| "learning_rate": 1.875e-07, | |
| "loss": -0.0003, | |
| "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": 0.0, | |
| "completions/max_length": 9385.0, | |
| "completions/max_terminated_length": 9385.0, | |
| "completions/mean_length": 4270.703125, | |
| "completions/mean_terminated_length": 4270.703125, | |
| "completions/min_length": 1390.0, | |
| "completions/min_terminated_length": 1390.0, | |
| "epoch": 0.00012112403100775194, | |
| "grad_norm": 0.025862550499858347, | |
| "kl": 0.000957489013671875, | |
| "learning_rate": 2.5e-07, | |
| "loss": 0.0031, | |
| "num_tokens": 1199795.0, | |
| "reward": 0.566771388053894, | |
| "reward_std": 0.48137491941452026, | |
| "rewards/avg_thinking_length_func": 182.33303833007812, | |
| "rewards/correct_answer_reward_func": 0.578125, | |
| "rewards/efficient_thinking_reward_func": 0.8707049785861538, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.7195165157318115, | |
| "rewards/tool_execution_reward_func": 1.9965277910232544, | |
| "rewards/visit_tool_reward_func": 0.9274243116378784, | |
| "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": 0.00014534883720930232, | |
| "grad_norm": 0.025877236026611388, | |
| "kl": 0.0009489059448242188, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "loss": 0.0031, | |
| "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": 0.0001695736434108527, | |
| "grad_norm": 0.025817236127475232, | |
| "kl": 0.0009660720825195312, | |
| "learning_rate": 3.75e-07, | |
| "loss": 0.0031, | |
| "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": 0.0001937984496124031, | |
| "grad_norm": 0.02584169829863559, | |
| "kl": 0.0009441375732421875, | |
| "learning_rate": 4.375e-07, | |
| "loss": 0.0031, | |
| "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": 7008.0, | |
| "completions/max_terminated_length": 7008.0, | |
| "completions/mean_length": 4088.546875, | |
| "completions/mean_terminated_length": 4088.546875, | |
| "completions/min_length": 1705.0, | |
| "completions/min_terminated_length": 1705.0, | |
| "epoch": 0.00021802325581395349, | |
| "grad_norm": 0.01625597308376849, | |
| "kl": 0.0009918212890625, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0013, | |
| "num_tokens": 1783761.0, | |
| "reward": 0.3732198178768158, | |
| "reward_std": 0.2907864451408386, | |
| "rewards/avg_thinking_length_func": 177.95510864257812, | |
| "rewards/correct_answer_reward_func": 0.390625, | |
| "rewards/efficient_thinking_reward_func": 0.8993925619789238, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.6866124868392944, | |
| "rewards/tool_execution_reward_func": 1.950781226158142, | |
| "rewards/visit_tool_reward_func": 0.8574961423873901, | |
| "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": 0.00024224806201550387, | |
| "grad_norm": 0.016618535814852814, | |
| "kl": 0.0009899139404296875, | |
| "learning_rate": 5.625e-07, | |
| "loss": 0.0013, | |
| "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": 0.00026647286821705426, | |
| "grad_norm": 0.016248156263205492, | |
| "kl": 0.0009660720825195312, | |
| "learning_rate": 6.249999999999999e-07, | |
| "loss": 0.0013, | |
| "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": 0.00029069767441860465, | |
| "grad_norm": 0.016111032400620007, | |
| "kl": 0.0009870529174804688, | |
| "learning_rate": 6.875e-07, | |
| "loss": 0.0013, | |
| "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": 6572.0, | |
| "completions/max_terminated_length": 6572.0, | |
| "completions/mean_length": 4119.703125, | |
| "completions/mean_terminated_length": 4119.703125, | |
| "completions/min_length": 1356.0, | |
| "completions/min_terminated_length": 1356.0, | |
| "epoch": 0.00031492248062015503, | |
| "grad_norm": 0.019643777904198217, | |
| "kl": 0.0009822845458984375, | |
| "learning_rate": 7.5e-07, | |
| "loss": -0.0008, | |
| "num_tokens": 2367034.0, | |
| "reward": 0.6774564981460571, | |
| "reward_std": 0.3563808798789978, | |
| "rewards/avg_thinking_length_func": 176.69476318359375, | |
| "rewards/correct_answer_reward_func": 0.625, | |
| "rewards/efficient_thinking_reward_func": 0.8704519537344548, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.6578426361083984, | |
| "rewards/tool_execution_reward_func": 2.0, | |
| "rewards/visit_tool_reward_func": 0.9361900091171265, | |
| "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": 0.0003391472868217054, | |
| "grad_norm": 0.0194815826710202, | |
| "kl": 0.0010242462158203125, | |
| "learning_rate": 8.125e-07, | |
| "loss": -0.0008, | |
| "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": 0.0003633720930232558, | |
| "grad_norm": 0.019402854833833996, | |
| "kl": 0.0010585784912109375, | |
| "learning_rate": 8.75e-07, | |
| "loss": -0.0008, | |
| "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": 0.0003875968992248062, | |
| "grad_norm": 0.019438299719581362, | |
| "kl": 0.0011272430419921875, | |
| "learning_rate": 9.374999999999999e-07, | |
| "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": 0.0, | |
| "completions/max_length": 7597.0, | |
| "completions/max_terminated_length": 7597.0, | |
| "completions/mean_length": 4205.671875, | |
| "completions/mean_terminated_length": 4205.671875, | |
| "completions/min_length": 1507.0, | |
| "completions/min_terminated_length": 1507.0, | |
| "epoch": 0.0004118217054263566, | |
| "grad_norm": 0.014823687168402296, | |
| "kl": 0.0011005401611328125, | |
| "learning_rate": 1e-06, | |
| "loss": 0.0009, | |
| "num_tokens": 2985545.0, | |
| "reward": 0.3260263204574585, | |
| "reward_std": 0.2300996333360672, | |
| "rewards/avg_thinking_length_func": 177.14329528808594, | |
| "rewards/correct_answer_reward_func": 0.375, | |
| "rewards/efficient_thinking_reward_func": 0.8988714947132084, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.8095711469650269, | |
| "rewards/tool_execution_reward_func": 1.99609375, | |
| "rewards/visit_tool_reward_func": 0.852025032043457, | |
| "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": 0.00043604651162790697, | |
| "grad_norm": 0.014727006858324664, | |
| "kl": 0.0011577606201171875, | |
| "learning_rate": 1.0625e-06, | |
| "loss": 0.0009, | |
| "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": 0.00046027131782945736, | |
| "grad_norm": 0.014837711956269274, | |
| "kl": 0.0012722015380859375, | |
| "learning_rate": 1.125e-06, | |
| "loss": 0.0009, | |
| "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": 0.00048449612403100775, | |
| "grad_norm": 0.014894430575329584, | |
| "kl": 0.00146484375, | |
| "learning_rate": 1.1874999999999999e-06, | |
| "loss": 0.0009, | |
| "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": 0.0, | |
| "completions/max_length": 7476.0, | |
| "completions/max_terminated_length": 7476.0, | |
| "completions/mean_length": 4097.921875, | |
| "completions/mean_terminated_length": 4097.921875, | |
| "completions/min_length": 1514.0, | |
| "completions/min_terminated_length": 1514.0, | |
| "epoch": 0.0005087209302325581, | |
| "grad_norm": 0.0189498267274778, | |
| "kl": 0.0019931793212890625, | |
| "learning_rate": 1.2499999999999999e-06, | |
| "loss": 0.0003, | |
| "num_tokens": 3561495.0, | |
| "reward": 0.5717383623123169, | |
| "reward_std": 0.33007949590682983, | |
| "rewards/avg_thinking_length_func": 177.5142822265625, | |
| "rewards/correct_answer_reward_func": 0.53125, | |
| "rewards/efficient_thinking_reward_func": 0.8662384906971484, | |
| "rewards/format_reward_func": 0.9937499761581421, | |
| "rewards/num_xml_reward_func": 1.779766321182251, | |
| "rewards/tool_execution_reward_func": 1.979819416999817, | |
| "rewards/visit_tool_reward_func": 0.9004297256469727, | |
| "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": 0.0005329457364341085, | |
| "grad_norm": 0.019010527717988047, | |
| "kl": 0.00229644775390625, | |
| "learning_rate": 1.3125e-06, | |
| "loss": 0.0003, | |
| "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": 0.0005571705426356589, | |
| "grad_norm": 0.01910688815244073, | |
| "kl": 0.00276947021484375, | |
| "learning_rate": 1.375e-06, | |
| "loss": 0.0003, | |
| "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": 0.0005813953488372093, | |
| "grad_norm": 0.019047374161024387, | |
| "kl": 0.00327301025390625, | |
| "learning_rate": 1.4375e-06, | |
| "loss": 0.0003, | |
| "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": 7779.0, | |
| "completions/max_terminated_length": 7779.0, | |
| "completions/mean_length": 4011.9375, | |
| "completions/mean_terminated_length": 4011.9375, | |
| "completions/min_length": 1884.0, | |
| "completions/min_terminated_length": 1884.0, | |
| "epoch": 0.0006056201550387597, | |
| "grad_norm": 0.01969391991938911, | |
| "kl": 0.0029449462890625, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.0003, | |
| "num_tokens": 4148002.0, | |
| "reward": 0.4466557502746582, | |
| "reward_std": 0.2478387951850891, | |
| "rewards/avg_thinking_length_func": 174.6974639892578, | |
| "rewards/correct_answer_reward_func": 0.40625, | |
| "rewards/efficient_thinking_reward_func": 0.9054659197504085, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.806973934173584, | |
| "rewards/tool_execution_reward_func": 1.9922122955322266, | |
| "rewards/visit_tool_reward_func": 0.871803879737854, | |
| "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": 0.0006298449612403101, | |
| "grad_norm": 0.01979038843755439, | |
| "kl": 0.003414154052734375, | |
| "learning_rate": 1.5624999999999999e-06, | |
| "loss": 0.0003, | |
| "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": 0.0006540697674418605, | |
| "grad_norm": 0.019676702255338734, | |
| "kl": 0.004245758056640625, | |
| "learning_rate": 1.625e-06, | |
| "loss": 0.0003, | |
| "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": 0.0006782945736434108, | |
| "grad_norm": 0.0198896583655868, | |
| "kl": 0.00508880615234375, | |
| "learning_rate": 1.6875e-06, | |
| "loss": 0.0003, | |
| "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": 7881.0, | |
| "completions/max_terminated_length": 7881.0, | |
| "completions/mean_length": 4278.0, | |
| "completions/mean_terminated_length": 4278.0, | |
| "completions/min_length": 1269.0, | |
| "completions/min_terminated_length": 1269.0, | |
| "epoch": 0.0007025193798449612, | |
| "grad_norm": 0.02473412222614823, | |
| "kl": 0.00722503662109375, | |
| "learning_rate": 1.75e-06, | |
| "loss": 0.0005, | |
| "num_tokens": 4732732.0, | |
| "reward": 0.639769971370697, | |
| "reward_std": 0.3489268720149994, | |
| "rewards/avg_thinking_length_func": 183.79090881347656, | |
| "rewards/correct_answer_reward_func": 0.640625, | |
| "rewards/efficient_thinking_reward_func": 0.8433743364598003, | |
| "rewards/format_reward_func": 0.9991071224212646, | |
| "rewards/num_xml_reward_func": 1.686936616897583, | |
| "rewards/tool_execution_reward_func": 1.9818710088729858, | |
| "rewards/visit_tool_reward_func": 0.923589289188385, | |
| "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": 0.0007267441860465116, | |
| "grad_norm": 0.024757116664213524, | |
| "kl": 0.0076904296875, | |
| "learning_rate": 1.8125e-06, | |
| "loss": 0.0005, | |
| "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": 0.000750968992248062, | |
| "grad_norm": 0.02444644630643307, | |
| "kl": 0.0073394775390625, | |
| "learning_rate": 1.8749999999999998e-06, | |
| "loss": 0.0005, | |
| "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": 0.0007751937984496124, | |
| "grad_norm": 0.024210451469423133, | |
| "kl": 0.007171630859375, | |
| "learning_rate": 1.9375e-06, | |
| "loss": 0.0005, | |
| "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": 7912.0, | |
| "completions/max_terminated_length": 7912.0, | |
| "completions/mean_length": 4317.890625, | |
| "completions/mean_terminated_length": 4317.890625, | |
| "completions/min_length": 1736.0, | |
| "completions/min_terminated_length": 1736.0, | |
| "epoch": 0.0007994186046511628, | |
| "grad_norm": 0.020658762871057952, | |
| "kl": 0.007049560546875, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0, | |
| "num_tokens": 5347783.0, | |
| "reward": 0.33683592081069946, | |
| "reward_std": 0.32624948024749756, | |
| "rewards/avg_thinking_length_func": 177.01129150390625, | |
| "rewards/correct_answer_reward_func": 0.375, | |
| "rewards/efficient_thinking_reward_func": 0.8817601664392056, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.5408036708831787, | |
| "rewards/tool_execution_reward_func": 1.9917367696762085, | |
| "rewards/visit_tool_reward_func": 0.9276807308197021, | |
| "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": 0.0008236434108527132, | |
| "grad_norm": 0.02072632567074888, | |
| "kl": 0.0077972412109375, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0, | |
| "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": 0.0008478682170542636, | |
| "grad_norm": 0.020770020029080613, | |
| "kl": 0.0087432861328125, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0, | |
| "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": 0.0008720930232558139, | |
| "grad_norm": 0.020487067102301602, | |
| "kl": 0.0097198486328125, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0, | |
| "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": 0.0, | |
| "completions/max_length": 7378.0, | |
| "completions/max_terminated_length": 7378.0, | |
| "completions/mean_length": 4152.5, | |
| "completions/mean_terminated_length": 4152.5, | |
| "completions/min_length": 1423.0, | |
| "completions/min_terminated_length": 1423.0, | |
| "epoch": 0.0008963178294573643, | |
| "grad_norm": 0.022364107178309313, | |
| "kl": 0.0112152099609375, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0001, | |
| "num_tokens": 5921090.0, | |
| "reward": 0.6556656360626221, | |
| "reward_std": 0.5008378028869629, | |
| "rewards/avg_thinking_length_func": 170.4791259765625, | |
| "rewards/correct_answer_reward_func": 0.625, | |
| "rewards/efficient_thinking_reward_func": 0.8892575272805912, | |
| "rewards/format_reward_func": 0.987500011920929, | |
| "rewards/num_xml_reward_func": 1.5408031940460205, | |
| "rewards/tool_execution_reward_func": 1.96875, | |
| "rewards/visit_tool_reward_func": 0.9249746799468994, | |
| "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": 0.0009205426356589147, | |
| "grad_norm": 0.022597206540891295, | |
| "kl": 0.0123443603515625, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0001, | |
| "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": 0.0009447674418604651, | |
| "grad_norm": 0.02246679376217943, | |
| "kl": 0.013580322265625, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0001, | |
| "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": 0.0009689922480620155, | |
| "grad_norm": 0.022296105800735398, | |
| "kl": 0.015106201171875, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0001, | |
| "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": 7494.0, | |
| "completions/max_terminated_length": 7494.0, | |
| "completions/mean_length": 4562.296875, | |
| "completions/mean_terminated_length": 4562.296875, | |
| "completions/min_length": 2143.0, | |
| "completions/min_terminated_length": 2143.0, | |
| "epoch": 0.0009932170542635659, | |
| "grad_norm": 0.021503135345542313, | |
| "kl": 0.015594482421875, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0007, | |
| "num_tokens": 6556719.0, | |
| "reward": 0.47225743532180786, | |
| "reward_std": 0.3904932141304016, | |
| "rewards/avg_thinking_length_func": 169.57839965820312, | |
| "rewards/correct_answer_reward_func": 0.4375, | |
| "rewards/efficient_thinking_reward_func": 0.917264621947748, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.817958116531372, | |
| "rewards/tool_execution_reward_func": 1.9884111881256104, | |
| "rewards/visit_tool_reward_func": 0.9651369452476501, | |
| "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": 0.0010174418604651163, | |
| "grad_norm": 0.02149252867250571, | |
| "kl": 0.01715087890625, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0007, | |
| "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": 0.0010416666666666667, | |
| "grad_norm": 0.02173596902997293, | |
| "kl": 0.018798828125, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0007, | |
| "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": 0.001065891472868217, | |
| "grad_norm": 0.02188237517399594, | |
| "kl": 0.020751953125, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0007, | |
| "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": 0.0, | |
| "completions/max_length": 9017.0, | |
| "completions/max_terminated_length": 9017.0, | |
| "completions/mean_length": 4664.796875, | |
| "completions/mean_terminated_length": 4664.796875, | |
| "completions/min_length": 1910.0, | |
| "completions/min_terminated_length": 1910.0, | |
| "epoch": 0.0010901162790697674, | |
| "grad_norm": 0.02354857583102173, | |
| "kl": 0.020477294921875, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0014, | |
| "num_tokens": 7181732.0, | |
| "reward": 0.7991669178009033, | |
| "reward_std": 0.36247026920318604, | |
| "rewards/avg_thinking_length_func": 171.8461151123047, | |
| "rewards/correct_answer_reward_func": 0.703125, | |
| "rewards/efficient_thinking_reward_func": 0.8915984372821139, | |
| "rewards/format_reward_func": 0.9998437166213989, | |
| "rewards/num_xml_reward_func": 1.8501074314117432, | |
| "rewards/tool_execution_reward_func": 1.9971354007720947, | |
| "rewards/visit_tool_reward_func": 1.071668028831482, | |
| "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": 0.0011143410852713178, | |
| "grad_norm": 0.023994471938115103, | |
| "kl": 0.0224609375, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0014, | |
| "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": 0.0011385658914728682, | |
| "grad_norm": 0.026516939220345738, | |
| "kl": 0.02508544921875, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0014, | |
| "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": 0.0011627906976744186, | |
| "grad_norm": 0.024485287814160223, | |
| "kl": 0.0262451171875, | |
| "learning_rate": 2e-06, | |
| "loss": -0.0014, | |
| "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": 8522.0, | |
| "completions/max_terminated_length": 8522.0, | |
| "completions/mean_length": 4866.125, | |
| "completions/mean_terminated_length": 4866.125, | |
| "completions/min_length": 1959.0, | |
| "completions/min_terminated_length": 1959.0, | |
| "epoch": 0.001187015503875969, | |
| "grad_norm": 0.02407332594201, | |
| "kl": 0.032012939453125, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0014, | |
| "num_tokens": 7868034.0, | |
| "reward": 0.39128515124320984, | |
| "reward_std": 0.3533371090888977, | |
| "rewards/avg_thinking_length_func": 164.74734497070312, | |
| "rewards/correct_answer_reward_func": 0.359375, | |
| "rewards/efficient_thinking_reward_func": 0.9209367558816545, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.6406757831573486, | |
| "rewards/tool_execution_reward_func": 1.98927903175354, | |
| "rewards/visit_tool_reward_func": 1.0120830535888672, | |
| "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": 0.0012112403100775194, | |
| "grad_norm": 0.02479690454991753, | |
| "kl": 0.035888671875, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0014, | |
| "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": 0.0012354651162790698, | |
| "grad_norm": 0.027012142633289393, | |
| "kl": 0.04046630859375, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0014, | |
| "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": 0.0012596899224806201, | |
| "grad_norm": 0.026499465739179152, | |
| "kl": 0.04803466796875, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0014, | |
| "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": 7622.0, | |
| "completions/max_terminated_length": 7622.0, | |
| "completions/mean_length": 4509.75, | |
| "completions/mean_terminated_length": 4509.75, | |
| "completions/min_length": 1816.0, | |
| "completions/min_terminated_length": 1816.0, | |
| "epoch": 0.0012839147286821705, | |
| "grad_norm": 0.019741394516818018, | |
| "kl": 0.04510498046875, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0, | |
| "num_tokens": 8481102.0, | |
| "reward": 0.7655854225158691, | |
| "reward_std": 0.27847254276275635, | |
| "rewards/avg_thinking_length_func": 158.9434051513672, | |
| "rewards/correct_answer_reward_func": 0.671875, | |
| "rewards/efficient_thinking_reward_func": 0.884494477975468, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.8834664821624756, | |
| "rewards/tool_execution_reward_func": 2.0, | |
| "rewards/visit_tool_reward_func": 1.1049017906188965, | |
| "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": 0.001308139534883721, | |
| "grad_norm": 0.028517188784132036, | |
| "kl": 0.06060791015625, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0001, | |
| "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": 0.0013323643410852713, | |
| "grad_norm": 0.02643367822401968, | |
| "kl": 0.06280517578125, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0001, | |
| "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": 0.0013565891472868217, | |
| "grad_norm": 0.020594752118506976, | |
| "kl": 0.056884765625, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0001, | |
| "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": 7375.0, | |
| "completions/max_terminated_length": 7375.0, | |
| "completions/mean_length": 4285.046875, | |
| "completions/mean_terminated_length": 4285.046875, | |
| "completions/min_length": 2418.0, | |
| "completions/min_terminated_length": 2418.0, | |
| "epoch": 0.001380813953488372, | |
| "grad_norm": 0.019100627823517295, | |
| "kl": 0.06195068359375, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0005, | |
| "num_tokens": 9112297.0, | |
| "reward": 0.5274717807769775, | |
| "reward_std": 0.2380232810974121, | |
| "rewards/avg_thinking_length_func": 145.75924682617188, | |
| "rewards/correct_answer_reward_func": 0.453125, | |
| "rewards/efficient_thinking_reward_func": 0.9274070198828231, | |
| "rewards/format_reward_func": 1.0, | |
| "rewards/num_xml_reward_func": 1.7929463386535645, | |
| "rewards/tool_execution_reward_func": 1.9959805011749268, | |
| "rewards/visit_tool_reward_func": 1.0335674285888672, | |
| "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": 0.0014050387596899225, | |
| "grad_norm": 0.019834849658967178, | |
| "kl": 0.06695556640625, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0005, | |
| "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": 0.0014292635658914728, | |
| "grad_norm": 0.020359737753586633, | |
| "kl": 0.0740966796875, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0005, | |
| "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": 0.0014534883720930232, | |
| "grad_norm": 0.020904893352951728, | |
| "kl": 0.085693359375, | |
| "learning_rate": 2e-06, | |
| "loss": 0.0005, | |
| "step": 60 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 640, | |
| "num_input_tokens_seen": 9112297, | |
| "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": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |