math-llada1.5 / trainer_state.json
mousezhang's picture
Upload LoRA adapter
7a546c6 verified
Raw
History Blame Contribute Delete
76.5 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.08,
"eval_steps": 500,
"global_step": 600,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"clip_ratio": 0.0,
"completion_length": 225.26667022705078,
"correct_distill_rate": 0.4333333373069763,
"distill_active_tokens": 90.0000015258789,
"distill_lambda": 0.6075,
"distill_loss": 2.125720429420471,
"distill_pairs": 8316.93349609375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.0013333333333333333,
"grad_norm": 2.6880300045013428,
"grpo_loss": 0.02052102223969996,
"kl": 0.5130260229110718,
"learning_rate": 8.333333333333333e-07,
"loss": 1.3137,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.5854833126068115,
"reward_std": 0.43471636846661565,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.8666666626930237,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.031183331552892924,
"step": 10,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 32.0,
"teacher_step_min": 32.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 210.9000045776367,
"correct_distill_rate": 0.366666679084301,
"distill_active_tokens": 65.46667022705078,
"distill_lambda": 0.6241666666666666,
"distill_loss": 1.4288323879241944,
"distill_pairs": 4331.3333984375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.0026666666666666666,
"grad_norm": 2.736140727996826,
"grpo_loss": 0.009467763220891356,
"kl": 0.23669322729110717,
"learning_rate": 1e-06,
"loss": 0.8988,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.504316693544388,
"reward_std": 0.36546106189489364,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7333333224058152,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.029316667839884757,
"step": 20,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 219.43333740234374,
"correct_distill_rate": 0.25000000149011614,
"distill_active_tokens": 49.5166675567627,
"distill_lambda": 0.6408333333333334,
"distill_loss": 1.1963014721870422,
"distill_pairs": 3633.15009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.004,
"grad_norm": 0.14850719273090363,
"grpo_loss": 0.01645257160998881,
"kl": 0.4113136172294617,
"learning_rate": 1e-06,
"loss": 0.783,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.2393333554267882,
"reward_std": 0.24247467098757625,
"rewards/boxed_and_answer_tags_format_reward": 0.7333333373069764,
"rewards/correctness_reward_func_math": 0.5000000029802323,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.00599999837577343,
"step": 30,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 19.2,
"teacher_step_min": 19.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 200.08333587646484,
"correct_distill_rate": 0.3666666731238365,
"distill_active_tokens": 67.78333549499511,
"distill_lambda": 0.6575,
"distill_loss": 1.6008101224899292,
"distill_pairs": 4800.4833984375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.005333333333333333,
"grad_norm": 0.15767081081867218,
"grpo_loss": 0.012945569586008787,
"kl": 0.3236380159854889,
"learning_rate": 1e-06,
"loss": 1.0673,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.5304667174816131,
"reward_std": 0.5186675733886659,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7333333343267441,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.055466667469590905,
"step": 40,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 22.4,
"teacher_step_min": 22.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 186.75000381469727,
"correct_distill_rate": 0.3666666716337204,
"distill_active_tokens": 73.0333351135254,
"distill_lambda": 0.6741666666666667,
"distill_loss": 1.5580667197704314,
"distill_pairs": 6345.100244140625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.006666666666666667,
"grad_norm": 0.24159300327301025,
"grpo_loss": 0.01483447402715683,
"kl": 0.3708606868982315,
"learning_rate": 1e-06,
"loss": 1.0624,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.5424500405788422,
"reward_std": 0.3353448113426566,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.7333333313465118,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.05911666887113824,
"step": 50,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 35.2,
"teacher_step_min": 35.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 221.78333587646483,
"correct_distill_rate": 0.5333333343267441,
"distill_active_tokens": 112.0000015258789,
"distill_lambda": 0.6908333333333333,
"distill_loss": 2.3937833666801454,
"distill_pairs": 8623.46708984375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.6,
"epoch": 0.008,
"grad_norm": 0.3055548071861267,
"grpo_loss": 0.011301343655213713,
"kl": 0.2825312837958336,
"learning_rate": 1e-06,
"loss": 1.657,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.7920667588710786,
"reward_std": 0.14835824826732277,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 1.0666666686534882,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.02459999794373289,
"step": 60,
"teacher_correct_rate": 0.6,
"teacher_step_mean": 57.6,
"teacher_step_min": 57.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 212.3000045776367,
"correct_distill_rate": 0.4333333447575569,
"distill_active_tokens": 85.36666831970214,
"distill_lambda": 0.7074999999999999,
"distill_loss": 1.859810322523117,
"distill_pairs": 6283.166821289063,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.009333333333333334,
"grad_norm": 3.2357449531555176,
"grpo_loss": 0.012210034625604748,
"kl": 0.30524938702583315,
"learning_rate": 1e-06,
"loss": 1.328,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.6074167370796204,
"reward_std": 0.598262820020318,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.8666666656732559,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.0009166655130684376,
"step": 70,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 38.4,
"teacher_step_min": 38.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 230.08333587646484,
"correct_distill_rate": 0.33333334028720857,
"distill_active_tokens": 72.83333435058594,
"distill_lambda": 0.7241666666666666,
"distill_loss": 1.851210343837738,
"distill_pairs": 5134.6333984375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.010666666666666666,
"grad_norm": 4.007431507110596,
"grpo_loss": 0.013526016799733042,
"kl": 0.33814932927489283,
"learning_rate": 1e-06,
"loss": 1.3495,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.3737167090177536,
"reward_std": 0.4614322084933519,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.6666666686534881,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.03461666759103536,
"step": 80,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 19.2,
"teacher_step_min": 19.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 222.3166702270508,
"correct_distill_rate": 0.30000001192092896,
"distill_active_tokens": 57.466668701171876,
"distill_lambda": 0.7408333333333333,
"distill_loss": 1.3765698671340942,
"distill_pairs": 3938.266845703125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.012,
"grad_norm": 0.1017308309674263,
"grpo_loss": 0.016420237766578794,
"kl": 0.4105050727725029,
"learning_rate": 1e-06,
"loss": 1.034,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.3503333747386932,
"reward_std": 0.41955652721226216,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.6,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0003333304077386856,
"step": 90,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 16.0,
"teacher_step_min": 16.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 205.83333892822264,
"correct_distill_rate": 0.21666666716337205,
"distill_active_tokens": 36.58333492279053,
"distill_lambda": 0.7575000000000001,
"distill_loss": 0.7874769687652587,
"distill_pairs": 2232.75,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.013333333333333334,
"grad_norm": 0.3074188530445099,
"grpo_loss": 0.008403621288016439,
"kl": 0.21008960157632828,
"learning_rate": 1e-06,
"loss": 0.6066,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.1235833644866944,
"reward_std": 0.17550912760198117,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.4333333343267441,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.009749999921768903,
"step": 100,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 9.6,
"teacher_step_min": 9.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 234.48333587646485,
"correct_distill_rate": 0.03333333432674408,
"distill_active_tokens": 6.15,
"distill_lambda": 0.7741666666666667,
"distill_loss": 0.15709410905838012,
"distill_pairs": 289.85,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.0,
"epoch": 0.014666666666666666,
"grad_norm": 0.24623394012451172,
"grpo_loss": 0.010652612382546067,
"kl": 0.26631477773189544,
"learning_rate": 1e-06,
"loss": 0.1317,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 0.7648500204086304,
"reward_std": 0.2296853244304657,
"rewards/boxed_and_answer_tags_format_reward": 0.6916666686534881,
"rewards/correctness_reward_func_math": 0.06666666865348816,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.00651666522026062,
"step": 110,
"teacher_correct_rate": 0.0,
"teacher_step_mean": 0.0,
"teacher_step_min": 0.0,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 235.03333435058593,
"correct_distill_rate": 0.21666667014360427,
"distill_active_tokens": 44.23333473205567,
"distill_lambda": 0.7908333333333333,
"distill_loss": 0.9640660881996155,
"distill_pairs": 3211.2334716796877,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.016,
"grad_norm": 2.431365966796875,
"grpo_loss": 0.010101206135004759,
"kl": 0.25253024101257326,
"learning_rate": 1e-06,
"loss": 0.7742,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.159583330154419,
"reward_std": 0.4241820402443409,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.43333334028720855,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.02374999983003363,
"step": 120,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 19.2,
"teacher_step_min": 19.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 219.46667098999023,
"correct_distill_rate": 0.26666666716337206,
"distill_active_tokens": 56.200001907348636,
"distill_lambda": 0.8074999999999999,
"distill_loss": 1.316667914390564,
"distill_pairs": 4502.733435058593,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.017333333333333333,
"grad_norm": 2.7000391483306885,
"grpo_loss": 0.009718735562637448,
"kl": 0.24296781048178673,
"learning_rate": 1e-06,
"loss": 1.0768,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.2444000244140625,
"reward_std": 0.2644742280244827,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.5333333343267441,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.03893333524465561,
"step": 130,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 35.2,
"teacher_step_min": 35.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 217.61667175292968,
"correct_distill_rate": 0.11666666716337204,
"distill_active_tokens": 15.866667175292969,
"distill_lambda": 0.8241666666666667,
"distill_loss": 0.3696224808692932,
"distill_pairs": 1320.66669921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.018666666666666668,
"grad_norm": 0.1918932944536209,
"grpo_loss": 0.009995615761727094,
"kl": 0.24988963082432747,
"learning_rate": 1e-06,
"loss": 0.3139,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 0.8714166939258575,
"reward_std": 0.14481605102773756,
"rewards/boxed_and_answer_tags_format_reward": 0.6416666686534882,
"rewards/correctness_reward_func_math": 0.23333333432674408,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.003583334758877754,
"step": 140,
"teacher_correct_rate": 0.1,
"teacher_step_mean": 12.8,
"teacher_step_min": 12.8,
"zero_std_ratio": 0.2
},
{
"clip_ratio": 0.0,
"completion_length": 205.2166717529297,
"correct_distill_rate": 0.4666666716337204,
"distill_active_tokens": 89.13333587646484,
"distill_lambda": 0.8408333333333333,
"distill_loss": 2.006036472320557,
"distill_pairs": 7598.33349609375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.02,
"grad_norm": 3.5439958572387695,
"grpo_loss": 0.014377873996272683,
"kl": 0.3594446286559105,
"learning_rate": 1e-06,
"loss": 1.7022,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.6571334183216095,
"reward_std": 0.32457001935690644,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.9333333313465119,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.02379999744007364,
"step": 150,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 44.8,
"teacher_step_min": 44.8,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 212.55000228881835,
"correct_distill_rate": 0.28333334177732467,
"distill_active_tokens": 53.75000114440918,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.3273056626319886,
"distill_pairs": 4536.116796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.021333333333333333,
"grad_norm": 0.22994132339954376,
"grpo_loss": 0.01386714018881321,
"kl": 0.3466774269938469,
"learning_rate": 1e-06,
"loss": 1.1421,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.2933833837509154,
"reward_std": 0.5133170209592208,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.5666666716337204,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.023283334076404573,
"step": 160,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 19.2,
"teacher_step_min": 19.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 198.86667175292968,
"correct_distill_rate": 0.5166666686534882,
"distill_active_tokens": 87.9000030517578,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.9566763639450073,
"distill_pairs": 5979.3001953125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.02266666666666667,
"grad_norm": 0.25841712951660156,
"grpo_loss": 0.013187666796147824,
"kl": 0.3296900033950806,
"learning_rate": 1e-06,
"loss": 1.6764,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.83508340716362,
"reward_std": 0.5739686183631421,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 1.0333333373069764,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.05175000224262476,
"step": 170,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 32.0,
"teacher_step_min": 32.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 224.45000457763672,
"correct_distill_rate": 0.3500000089406967,
"distill_active_tokens": 78.95000381469727,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.871301567554474,
"distill_pairs": 6672.51689453125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.024,
"grad_norm": 0.12022630125284195,
"grpo_loss": 0.010507767833769322,
"kl": 0.2626928314566612,
"learning_rate": 1e-06,
"loss": 1.6011,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.3656833827495576,
"reward_std": 0.44513750318437817,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.7000000059604645,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.08431666661053897,
"step": 180,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 38.4,
"teacher_step_min": 38.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 212.5666717529297,
"correct_distill_rate": 0.30000000447034836,
"distill_active_tokens": 48.716667556762694,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.0750308156013488,
"distill_pairs": 3128.883349609375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.025333333333333333,
"grad_norm": 0.09582177549600601,
"grpo_loss": 0.010563625581562519,
"kl": 0.2640887230634689,
"learning_rate": 1e-06,
"loss": 0.9243,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.3451000690460204,
"reward_std": 0.512438530754298,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.6000000089406967,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.04510000087320805,
"step": 190,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 16.0,
"teacher_step_min": 16.0,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 215.26667175292968,
"correct_distill_rate": 0.4500000163912773,
"distill_active_tokens": 98.25000267028808,
"distill_lambda": 0.8499999999999999,
"distill_loss": 2.4063288807868957,
"distill_pairs": 8177.483666992188,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.02666666666666667,
"grad_norm": 4.973355293273926,
"grpo_loss": 0.011660819593816995,
"kl": 0.29151821732521055,
"learning_rate": 1e-06,
"loss": 2.057,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.6288000881671905,
"reward_std": 0.6533891387283802,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.8999999970197677,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.021200002171099186,
"step": 200,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 35.2,
"teacher_step_min": 35.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 233.00000457763673,
"correct_distill_rate": 0.11666667014360428,
"distill_active_tokens": 25.866667556762696,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.5513983786106109,
"distill_pairs": 1621.066748046875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.028,
"grad_norm": 2.5393974781036377,
"grpo_loss": 0.009242854733020067,
"kl": 0.23107043504714966,
"learning_rate": 1e-06,
"loss": 0.4779,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 0.9898167014122009,
"reward_std": 0.4581179592758417,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.23333334028720856,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.006483332440257073,
"step": 210,
"teacher_correct_rate": 0.1,
"teacher_step_mean": 3.2,
"teacher_step_min": 3.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 216.58333740234374,
"correct_distill_rate": 0.3500000059604645,
"distill_active_tokens": 80.6500015258789,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.7652148485183716,
"distill_pairs": 5844.68349609375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.029333333333333333,
"grad_norm": 0.0977863296866417,
"grpo_loss": 0.01464300644584,
"kl": 0.3660737805068493,
"learning_rate": 1e-06,
"loss": 1.5151,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.4528167247772217,
"reward_std": 0.21277470737695695,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.7,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0028166670352220537,
"step": 220,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 227.2500030517578,
"correct_distill_rate": 0.20000000298023224,
"distill_active_tokens": 41.083334350585936,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.9485945224761962,
"distill_pairs": 4213.85009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.030666666666666665,
"grad_norm": 0.14790078997612,
"grpo_loss": 0.008689700672402979,
"kl": 0.21724254712462426,
"learning_rate": 1e-06,
"loss": 0.815,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.0973500072956086,
"reward_std": 0.25364548747893423,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.4000000059604645,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.04431666758318897,
"step": 230,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 207.13333892822266,
"correct_distill_rate": 0.5333333358168602,
"distill_active_tokens": 100.03333625793456,
"distill_lambda": 0.8499999999999999,
"distill_loss": 2.290933895111084,
"distill_pairs": 8255.700219726563,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.032,
"grad_norm": 0.1777622252702713,
"grpo_loss": 0.010405205469578505,
"kl": 0.260128790140152,
"learning_rate": 1e-06,
"loss": 1.9577,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.7873333901166917,
"reward_std": 0.3152889137621969,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 1.0666666716337203,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.02933333513792604,
"step": 240,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 48.0,
"teacher_step_min": 48.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 199.65000610351564,
"correct_distill_rate": 0.5666666701436043,
"distill_active_tokens": 104.95000381469727,
"distill_lambda": 0.8499999999999999,
"distill_loss": 2.4449519991874693,
"distill_pairs": 7209.71689453125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.7,
"epoch": 0.03333333333333333,
"grad_norm": 0.15934187173843384,
"grpo_loss": 0.01144193485379219,
"kl": 0.2860474333167076,
"learning_rate": 1e-06,
"loss": 2.0897,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.8981500446796418,
"reward_std": 0.46845852844417096,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 1.1333333402872086,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.014816667046397924,
"step": 250,
"teacher_correct_rate": 0.7,
"teacher_step_mean": 48.0,
"teacher_step_min": 48.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 208.68333740234374,
"correct_distill_rate": 0.4833333417773247,
"distill_active_tokens": 85.80000267028808,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.9418034315109254,
"distill_pairs": 6362.20009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.034666666666666665,
"grad_norm": 0.4368448257446289,
"grpo_loss": 0.02998018339276314,
"kl": 0.7495038151741028,
"learning_rate": 1e-06,
"loss": 1.6805,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.7180000245571136,
"reward_std": 0.3200799889862537,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.9666666597127914,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0013333334354683757,
"step": 260,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 212.58333740234374,
"correct_distill_rate": 0.20000000447034835,
"distill_active_tokens": 41.16666679382324,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.8686363637447357,
"distill_pairs": 3271.90009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.036,
"grad_norm": 0.14111259579658508,
"grpo_loss": 0.011410776525735855,
"kl": 0.2852680191397667,
"learning_rate": 1e-06,
"loss": 0.7498,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.1286333858966828,
"reward_std": 0.22248505465686322,
"rewards/boxed_and_answer_tags_format_reward": 0.7333333373069764,
"rewards/correctness_reward_func_math": 0.39999999701976774,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.004700000677257776,
"step": 270,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 19.2,
"teacher_step_min": 19.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 218.63333587646486,
"correct_distill_rate": 0.25000000149011614,
"distill_active_tokens": 46.566667556762695,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.0118345975875855,
"distill_pairs": 3675.5667724609375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.037333333333333336,
"grad_norm": 0.7554723024368286,
"grpo_loss": 0.009661140898242594,
"kl": 0.24152729138731957,
"learning_rate": 1e-06,
"loss": 0.8697,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.2342000484466553,
"reward_std": 0.2651735613122582,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.5000000029802323,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.015800000354647636,
"step": 280,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 215.98333587646485,
"correct_distill_rate": 0.4666666716337204,
"distill_active_tokens": 98.05000228881836,
"distill_lambda": 0.8499999999999999,
"distill_loss": 2.3486120223999025,
"distill_pairs": 7988.883569335938,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.03866666666666667,
"grad_norm": 0.2115684151649475,
"grpo_loss": 0.0214917603880167,
"kl": 0.537292867898941,
"learning_rate": 1e-06,
"loss": 2.0178,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.6662500381469727,
"reward_std": 0.5625137768685817,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.9333333432674408,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.01708333354908973,
"step": 290,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 57.6,
"teacher_step_min": 57.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 215.0500030517578,
"correct_distill_rate": 0.2333333373069763,
"distill_active_tokens": 51.083334350585936,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.0875648260116577,
"distill_pairs": 4289.71669921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.04,
"grad_norm": 0.9998526573181152,
"grpo_loss": 0.012024427251890301,
"kl": 0.3006098799407482,
"learning_rate": 1e-06,
"loss": 0.9365,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.2354666948318482,
"reward_std": 0.2608564658090472,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.4666666746139526,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.018800000287592412,
"step": 300,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 216.86667022705078,
"correct_distill_rate": 0.21666667014360427,
"distill_active_tokens": 37.75000114440918,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.8116459727287293,
"distill_pairs": 1460.9167236328126,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.04133333333333333,
"grad_norm": 2.209340810775757,
"grpo_loss": 0.012333661876618862,
"kl": 0.30833988040685656,
"learning_rate": 1e-06,
"loss": 0.7022,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.159300059080124,
"reward_std": 0.4111998528242111,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.43333334028720855,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.015700001653749497,
"step": 310,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 9.6,
"teacher_step_min": 9.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 197.35000457763672,
"correct_distill_rate": 0.33333333730697634,
"distill_active_tokens": 48.516667175292966,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.95401531457901,
"distill_pairs": 3255.4834716796877,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.042666666666666665,
"grad_norm": 0.12615324556827545,
"grpo_loss": 0.01128495978191495,
"kl": 0.2821231111884117,
"learning_rate": 1e-06,
"loss": 0.8222,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.421933376789093,
"reward_std": 0.42693512998521327,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.6666666746139527,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.013600004091858863,
"step": 320,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 201.38333740234376,
"correct_distill_rate": 0.36666667461395264,
"distill_active_tokens": 77.36666831970214,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.7716629028320312,
"distill_pairs": 5991.433544921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.044,
"grad_norm": 2.417386770248413,
"grpo_loss": 0.010987034067511559,
"kl": 0.27467371970415116,
"learning_rate": 1e-06,
"loss": 1.5169,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.4943500816822053,
"reward_std": 0.5299370455555618,
"rewards/boxed_and_answer_tags_format_reward": 0.7250000059604644,
"rewards/correctness_reward_func_math": 0.7333333492279053,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.036016667075455186,
"step": 330,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 185.6833366394043,
"correct_distill_rate": 0.3833333358168602,
"distill_active_tokens": 58.850001907348634,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.3537674069404602,
"distill_pairs": 4531.016717529297,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.04533333333333334,
"grad_norm": 4.176764011383057,
"grpo_loss": 0.012563177198171616,
"kl": 0.314078788459301,
"learning_rate": 1e-06,
"loss": 1.1633,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.5147666931152344,
"reward_std": 0.24802324501797557,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7666666716337204,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0064333342015743256,
"step": 340,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 28.8,
"teacher_step_min": 28.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 215.51667022705078,
"correct_distill_rate": 0.41666666716337203,
"distill_active_tokens": 81.68333549499512,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.74827299118042,
"distill_pairs": 6414.316967773438,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.04666666666666667,
"grad_norm": 0.1791873425245285,
"grpo_loss": 0.010752544412389398,
"kl": 0.268813268840313,
"learning_rate": 1e-06,
"loss": 1.4968,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.5963000178337097,
"reward_std": 0.1307512115687132,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.8333333343267441,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.02129999715834856,
"step": 350,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 35.2,
"teacher_step_min": 35.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 217.1500030517578,
"correct_distill_rate": 0.23333334028720856,
"distill_active_tokens": 46.93333435058594,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.12767835855484,
"distill_pairs": 3653.3333984375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.048,
"grad_norm": 3.1705570220947266,
"grpo_loss": 0.010592608759179712,
"kl": 0.2648145452141762,
"learning_rate": 1e-06,
"loss": 0.9691,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.2010166943073273,
"reward_std": 0.5750916752964258,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.4666666805744171,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.015649998933076857,
"step": 360,
"teacher_correct_rate": 0.1,
"teacher_step_mean": 12.8,
"teacher_step_min": 12.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 204.76667175292968,
"correct_distill_rate": 0.28333334177732467,
"distill_active_tokens": 51.11666755676269,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.2496458530426025,
"distill_pairs": 3885.41669921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.04933333333333333,
"grad_norm": 0.10218092799186707,
"grpo_loss": 0.01081814575009048,
"kl": 0.27045253217220305,
"learning_rate": 1e-06,
"loss": 1.073,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.347400051355362,
"reward_std": 0.30492510497570036,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.5666666597127914,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0390666700899601,
"step": 370,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 9.6,
"teacher_step_min": 9.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 211.00000381469727,
"correct_distill_rate": 0.30000000447034836,
"distill_active_tokens": 41.066667556762695,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.9792587637901307,
"distill_pairs": 3302.40009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.050666666666666665,
"grad_norm": 3.3643829822540283,
"grpo_loss": 0.013536219531670213,
"kl": 0.33840410187840464,
"learning_rate": 1e-06,
"loss": 0.8459,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.3615167319774628,
"reward_std": 0.20768519304692745,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.5999999970197678,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.011516667832620443,
"step": 380,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 16.0,
"teacher_step_min": 16.0,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 201.63333740234376,
"correct_distill_rate": 0.4333333373069763,
"distill_active_tokens": 83.46666793823242,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.9385169744491577,
"distill_pairs": 7416.80009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.052,
"grad_norm": 3.7896780967712402,
"grpo_loss": 0.012287950376048684,
"kl": 0.3071973204612732,
"learning_rate": 1e-06,
"loss": 1.66,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.6277500629425048,
"reward_std": 0.2536654853262007,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.8666666626930237,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.011083332449197769,
"step": 390,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 35.2,
"teacher_step_min": 35.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 204.01667022705078,
"correct_distill_rate": 0.23333333432674408,
"distill_active_tokens": 42.650000762939456,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.9760748863220214,
"distill_pairs": 4001.0834716796876,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.05333333333333334,
"grad_norm": 0.22813820838928223,
"grpo_loss": 0.010810398310422898,
"kl": 0.27025932520627977,
"learning_rate": 1e-06,
"loss": 0.8405,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.205583357810974,
"reward_std": 0.16969854801427572,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.46666666865348816,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.011083334311842918,
"step": 400,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 198.4333381652832,
"correct_distill_rate": 0.2500000029802322,
"distill_active_tokens": 45.050001525878905,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.9781241893768311,
"distill_pairs": 4078.683544921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.05466666666666667,
"grad_norm": 0.31846821308135986,
"grpo_loss": 0.010574485035613179,
"kl": 0.26436221301555635,
"learning_rate": 1e-06,
"loss": 0.842,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.2471666753292083,
"reward_std": 0.3832199780270457,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.5000000059604645,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0054999998770654205,
"step": 410,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 9.6,
"teacher_step_min": 9.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 207.53333892822266,
"correct_distill_rate": 0.36666667759418486,
"distill_active_tokens": 63.50000152587891,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.3927794098854065,
"distill_pairs": 4910.366796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.056,
"grad_norm": 0.08432240039110184,
"grpo_loss": 0.01054160986095667,
"kl": 0.2635391980409622,
"learning_rate": 1e-06,
"loss": 1.1944,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.475433385372162,
"reward_std": 0.479615673981607,
"rewards/boxed_and_answer_tags_format_reward": 0.7333333373069764,
"rewards/correctness_reward_func_math": 0.7333333313465118,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.008766665140865371,
"step": 420,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 22.4,
"teacher_step_min": 22.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 182.6500030517578,
"correct_distill_rate": 0.46666667610406876,
"distill_active_tokens": 78.71666831970215,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.849145781993866,
"distill_pairs": 5981.550268554687,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.6,
"epoch": 0.05733333333333333,
"grad_norm": 3.753330707550049,
"grpo_loss": 0.01213481854647398,
"kl": 0.30336867570877074,
"learning_rate": 1e-06,
"loss": 1.5839,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.7227833986282348,
"reward_std": 0.5145857689902187,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.9333333402872086,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.03945000128151151,
"step": 430,
"teacher_correct_rate": 0.6,
"teacher_step_mean": 41.6,
"teacher_step_min": 41.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 206.2833381652832,
"correct_distill_rate": 0.41666666865348817,
"distill_active_tokens": 78.0833351135254,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.5649609327316285,
"distill_pairs": 5959.783544921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.058666666666666666,
"grad_norm": 0.1965167224407196,
"grpo_loss": 0.010877907881513238,
"kl": 0.2719471111893654,
"learning_rate": 1e-06,
"loss": 1.3411,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.6078333735466004,
"reward_std": 0.28344318851595746,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.8333333373069763,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.03283333298750222,
"step": 440,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 28.8,
"teacher_step_min": 28.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 197.83334045410157,
"correct_distill_rate": 0.38333334028720856,
"distill_active_tokens": 74.43333587646484,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.7809359312057496,
"distill_pairs": 5425.833544921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.06,
"grad_norm": 0.132995143532753,
"grpo_loss": 0.010555988736450672,
"kl": 0.263897967338562,
"learning_rate": 1e-06,
"loss": 1.5244,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.5677167236804963,
"reward_std": 0.5472017996013164,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7666666686534882,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0593833327293396,
"step": 450,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 19.2,
"teacher_step_min": 19.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 228.5666702270508,
"correct_distill_rate": 0.333333346247673,
"distill_active_tokens": 78.20000228881835,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.9166529178619385,
"distill_pairs": 6457.2669677734375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.06133333333333333,
"grad_norm": 4.037680149078369,
"grpo_loss": 0.009952851478010416,
"kl": 0.24881922453641891,
"learning_rate": 1e-06,
"loss": 1.6391,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.3918334186077117,
"reward_std": 0.5024258546531201,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.6666666567325592,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.024833334423601628,
"step": 460,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 12.8,
"teacher_step_min": 12.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 209.85000305175782,
"correct_distill_rate": 0.4666666746139526,
"distill_active_tokens": 88.58333587646484,
"distill_lambda": 0.8499999999999999,
"distill_loss": 2.019017732143402,
"distill_pairs": 7173.283544921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.06266666666666666,
"grad_norm": 4.533484935760498,
"grpo_loss": 0.01233915607444942,
"kl": 0.30847918838262556,
"learning_rate": 1e-06,
"loss": 1.7285,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.6745166659355164,
"reward_std": 0.6021390356123447,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.9333333373069763,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.0004833339131437242,
"step": 470,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 16.0,
"teacher_step_min": 16.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 207.28333587646483,
"correct_distill_rate": 0.25000000149011614,
"distill_active_tokens": 47.20000038146973,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.0004250586032868,
"distill_pairs": 3533.6000732421876,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.064,
"grad_norm": 1.3196847438812256,
"grpo_loss": 0.010060150502249598,
"kl": 0.25150332078337667,
"learning_rate": 1e-06,
"loss": 0.8604,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.292116677761078,
"reward_std": 0.32734045423567293,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.5000000029802323,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.05044999858364463,
"step": 480,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 22.4,
"teacher_step_min": 22.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 196.68333892822267,
"correct_distill_rate": 0.3166666716337204,
"distill_active_tokens": 61.55000247955322,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.551763552427292,
"distill_pairs": 5761.383480834961,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.06533333333333333,
"grad_norm": 0.5217611193656921,
"grpo_loss": 0.011460760375484824,
"kl": 0.28651883602142336,
"learning_rate": 1e-06,
"loss": 1.3305,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.4346166670322418,
"reward_std": 0.6206270655617118,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.6333333432674408,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.05961666863877326,
"step": 490,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 195.9333381652832,
"correct_distill_rate": 0.6000000089406967,
"distill_active_tokens": 103.50000381469727,
"distill_lambda": 0.8499999999999999,
"distill_loss": 2.4938096404075623,
"distill_pairs": 9465.5669921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.6,
"epoch": 0.06666666666666667,
"grad_norm": 0.6471823453903198,
"grpo_loss": 0.014869017386808991,
"kl": 0.3717253111302853,
"learning_rate": 1e-06,
"loss": 2.1346,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 2.0183333337306975,
"reward_std": 0.2991330319084227,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 1.1999999940395356,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.06833333394606597,
"step": 500,
"teacher_correct_rate": 0.6,
"teacher_step_mean": 60.8,
"teacher_step_min": 60.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 226.51666870117188,
"correct_distill_rate": 0.3500000134110451,
"distill_active_tokens": 80.60000190734863,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.5574386894702912,
"distill_pairs": 7413.80009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.068,
"grad_norm": 0.23387746512889862,
"grpo_loss": 0.008425060054287314,
"kl": 0.21062467321753503,
"learning_rate": 1e-06,
"loss": 1.3322,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.4449167370796203,
"reward_std": 0.5978949490934611,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7000000029802322,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0032500016503036024,
"step": 510,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 200.63333892822266,
"correct_distill_rate": 0.4000000059604645,
"distill_active_tokens": 77.68333625793457,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.7711429297924042,
"distill_pairs": 6168.183642578125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.06933333333333333,
"grad_norm": 3.8891050815582275,
"grpo_loss": 0.010330989630892873,
"kl": 0.2582731544971466,
"learning_rate": 1e-06,
"loss": 1.5158,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.5358167350292207,
"reward_std": 0.46995821124874054,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.800000011920929,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.03581666275858879,
"step": 520,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 38.4,
"teacher_step_min": 38.4,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 220.36667175292968,
"correct_distill_rate": 0.3833333432674408,
"distill_active_tokens": 87.31666946411133,
"distill_lambda": 0.8499999999999999,
"distill_loss": 2.019676947593689,
"distill_pairs": 6000.68349609375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.07066666666666667,
"grad_norm": 3.0355794429779053,
"grpo_loss": 0.010777838341891766,
"kl": 0.2694434531033039,
"learning_rate": 1e-06,
"loss": 1.7275,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.4891834259033203,
"reward_std": 0.43126023281365633,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.7666666746139527,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.02748333397321403,
"step": 530,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 22.4,
"teacher_step_min": 22.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 209.1500030517578,
"correct_distill_rate": 0.26666666865348815,
"distill_active_tokens": 49.31666717529297,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.0885800957679748,
"distill_pairs": 3861.0834716796876,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.072,
"grad_norm": 0.21334092319011688,
"grpo_loss": 0.026782693155109883,
"kl": 0.6695667035877705,
"learning_rate": 1e-06,
"loss": 0.9521,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.2078833639621736,
"reward_std": 0.4981345862150192,
"rewards/boxed_and_answer_tags_format_reward": 0.6416666686534882,
"rewards/correctness_reward_func_math": 0.5333333373069763,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.03288333183154464,
"step": 540,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 22.4,
"teacher_step_min": 22.4,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 197.73333663940429,
"correct_distill_rate": 0.41666667312383654,
"distill_active_tokens": 67.6666690826416,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.5838679194450378,
"distill_pairs": 5490.2003173828125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.07333333333333333,
"grad_norm": 2.2654335498809814,
"grpo_loss": 0.012652527447789908,
"kl": 0.31631288304924965,
"learning_rate": 1e-06,
"loss": 1.3589,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.591100013256073,
"reward_std": 0.41575390547513963,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.8333333343267441,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.057766666635870934,
"step": 550,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 22.4,
"teacher_step_min": 22.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 211.71667098999023,
"correct_distill_rate": 0.2166666716337204,
"distill_active_tokens": 40.150000762939456,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.8401986479759216,
"distill_pairs": 3125.716796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.07466666666666667,
"grad_norm": 0.1575131118297577,
"grpo_loss": 0.0103579958435148,
"kl": 0.25894920378923414,
"learning_rate": 1e-06,
"loss": 0.7245,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.1959666967391969,
"reward_std": 0.2713510602712631,
"rewards/boxed_and_answer_tags_format_reward": 0.7333333373069764,
"rewards/correctness_reward_func_math": 0.43333333134651186,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.029300001449882984,
"step": 560,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 19.2,
"teacher_step_min": 19.2,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 217.26666946411132,
"correct_distill_rate": 0.21666666716337205,
"distill_active_tokens": 37.23333396911621,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.8638959288597107,
"distill_pairs": 2244.3667236328124,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.076,
"grad_norm": 0.08301323652267456,
"grpo_loss": 0.007887230976484716,
"kl": 0.19717951416969298,
"learning_rate": 1e-06,
"loss": 0.7422,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.173600047826767,
"reward_std": 0.35183555632829666,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.4333333343267441,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.0014000001829117537,
"step": 570,
"teacher_correct_rate": 0.1,
"teacher_step_mean": 6.4,
"teacher_step_min": 6.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 191.45000228881835,
"correct_distill_rate": 0.1,
"distill_active_tokens": 14.566667175292968,
"distill_lambda": 0.8499999999999999,
"distill_loss": 0.35042343139648435,
"distill_pairs": 923.56669921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.07733333333333334,
"grad_norm": 0.4160461127758026,
"grpo_loss": 0.012368399603292345,
"kl": 0.30920849591493604,
"learning_rate": 1e-06,
"loss": 0.3102,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.0015000641345977,
"reward_std": 0.08022082569077611,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.2,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.05150000043213367,
"step": 580,
"teacher_correct_rate": 0.1,
"teacher_step_mean": 6.4,
"teacher_step_min": 6.4,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 222.38333740234376,
"correct_distill_rate": 0.21666667461395264,
"distill_active_tokens": 48.116667938232425,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.1550864934921266,
"distill_pairs": 3228.15009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.07866666666666666,
"grad_norm": 0.36123189330101013,
"grpo_loss": 0.009179295971989632,
"kl": 0.22948118895292283,
"learning_rate": 1e-06,
"loss": 0.991,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.12813338637352,
"reward_std": 0.3610767122358084,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.43333332538604735,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.05519999749958515,
"step": 590,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 12.8,
"teacher_step_min": 12.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 205.35000305175782,
"correct_distill_rate": 0.21666667014360427,
"distill_active_tokens": 40.083334350585936,
"distill_lambda": 0.8499999999999999,
"distill_loss": 1.0030710339546203,
"distill_pairs": 3116.716796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.08,
"grad_norm": 2.4419198036193848,
"grpo_loss": 0.010236841207370163,
"kl": 0.255920447409153,
"learning_rate": 1e-06,
"loss": 0.8628,
"productive_jump_progress_mean": 0.0,
"productive_jump_score_mean": 0.0,
"productive_jump_selected_rate": 0.0,
"productive_jump_span_mean": 0.0,
"productive_jump_tail_entropy_mean": 0.0,
"reward": 1.1566666930913925,
"reward_std": 0.6550612782128156,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.43333334028720855,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.02666666265577078,
"step": 600,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
}
],
"logging_steps": 10,
"max_steps": 600,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}