math-llada8b / trainer_state.json
mousezhang's picture
Upload LoRA adapter
700b899 verified
Raw
History Blame Contribute Delete
76.4 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": 218.36667098999024,
"correct_distill_rate": 0.3500000089406967,
"distill_active_tokens": 65.38333587646484,
"distill_lambda": 0.4135000000000001,
"distill_loss": 1.7464235067367553,
"distill_pairs": 5472.750073242188,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.0013333333333333333,
"grad_norm": 0.9360572695732117,
"grpo_loss": 0.012857666192576288,
"kl": 0.3214402884244919,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.7397,
"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": 1.4063667178153991,
"reward_std": 0.4767286445945501,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7000000059604645,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.03529999847523868,
"step": 10,
"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": 194.8166717529297,
"correct_distill_rate": 0.33333334177732465,
"distill_active_tokens": 55.75000267028808,
"distill_lambda": 0.44350000000000006,
"distill_loss": 1.2291141271591186,
"distill_pairs": 4074.3834716796873,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.0026666666666666666,
"grad_norm": 0.1271679848432541,
"grpo_loss": 0.013159083062782883,
"kl": 0.3289754629135132,
"learning_rate": 2e-06,
"loss": 0.5555,
"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": 1.4066500663757324,
"reward_std": 0.4101110230665654,
"rewards/boxed_and_answer_tags_format_reward": 0.6916666686534881,
"rewards/correctness_reward_func_math": 0.6666666716337204,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.04831666480749845,
"step": 20,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 222.28333587646483,
"correct_distill_rate": 0.3833333447575569,
"distill_active_tokens": 82.33333549499511,
"distill_lambda": 0.47350000000000003,
"distill_loss": 1.970672792196274,
"distill_pairs": 6543.266772460937,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.004,
"grad_norm": 0.21332769095897675,
"grpo_loss": 0.015663789957761763,
"kl": 0.3915931716561317,
"learning_rate": 2e-06,
"loss": 0.938,
"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": 1.4463333904743194,
"reward_std": 0.5057359144091607,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.7666666656732559,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.07033333256840706,
"step": 30,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 64.0,
"teacher_step_min": 64.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 231.61666946411134,
"correct_distill_rate": 0.4000000059604645,
"distill_active_tokens": 91.38333587646484,
"distill_lambda": 0.5035000000000001,
"distill_loss": 2.172255039215088,
"distill_pairs": 7337.9500732421875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.005333333333333333,
"grad_norm": 2.4386839866638184,
"grpo_loss": 0.014839121932163834,
"kl": 0.3709773555397987,
"learning_rate": 2e-06,
"loss": 1.114,
"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": 1.4891833543777466,
"reward_std": 0.5412817992269993,
"rewards/boxed_and_answer_tags_format_reward": 0.725,
"rewards/correctness_reward_func_math": 0.8,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.03581666788086295,
"step": 40,
"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": 220.4166687011719,
"correct_distill_rate": 0.3500000014901161,
"distill_active_tokens": 79.86666793823242,
"distill_lambda": 0.5335,
"distill_loss": 1.8883683562278748,
"distill_pairs": 7331.33359375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.006666666666666667,
"grad_norm": 0.5964341163635254,
"grpo_loss": 0.017776214703917503,
"kl": 0.4444050818681717,
"learning_rate": 2e-06,
"loss": 1.0271,
"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": 1.391450011730194,
"reward_std": 0.256594761647284,
"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": -0.05021666563116014,
"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": 230.8166702270508,
"correct_distill_rate": 0.46666666865348816,
"distill_active_tokens": 96.85000076293946,
"distill_lambda": 0.5635,
"distill_loss": 2.332971441745758,
"distill_pairs": 5431.15009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.008,
"grad_norm": 0.29712194204330444,
"grpo_loss": 0.013440664997324348,
"kl": 0.3360165491700172,
"learning_rate": 2e-06,
"loss": 1.3267,
"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": 1.5680333515629172,
"reward_std": 0.47415485233068466,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.9333333373069763,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.11529999426566065,
"step": 60,
"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": 238.8166702270508,
"correct_distill_rate": 0.0833333358168602,
"distill_active_tokens": 19.73333396911621,
"distill_lambda": 0.5934999999999999,
"distill_loss": 0.41496833562850954,
"distill_pairs": 2062.4,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.009333333333333334,
"grad_norm": 0.29933038353919983,
"grpo_loss": 0.02071264535188675,
"kl": 0.5178156644105911,
"learning_rate": 2e-06,
"loss": 0.2652,
"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": 0.8293500125408173,
"reward_std": 0.2434039993211627,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.1666666716337204,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.07898333184421062,
"step": 70,
"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": 207.8166702270508,
"correct_distill_rate": 0.28333333879709244,
"distill_active_tokens": 49.56666717529297,
"distill_lambda": 0.6235000000000002,
"distill_loss": 1.2688277602195739,
"distill_pairs": 4842.166845703125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.010666666666666666,
"grad_norm": 0.26456767320632935,
"grpo_loss": 0.020783916860818864,
"kl": 0.5195970058441162,
"learning_rate": 2e-06,
"loss": 0.8093,
"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": 1.290000033378601,
"reward_std": 0.4725196108222008,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.566666665673256,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.026666670572012663,
"step": 80,
"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": 223.70000457763672,
"correct_distill_rate": 0.26666666865348815,
"distill_active_tokens": 65.01666717529297,
"distill_lambda": 0.6534999999999999,
"distill_loss": 1.5083132982254028,
"distill_pairs": 5972.316796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.012,
"grad_norm": 3.222008466720581,
"grpo_loss": 0.01341043901629746,
"kl": 0.3352597802877426,
"learning_rate": 2e-06,
"loss": 1.0032,
"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": 1.1958167135715485,
"reward_std": 0.4874852511798963,
"rewards/boxed_and_answer_tags_format_reward": 0.675,
"rewards/correctness_reward_func_math": 0.5333333373069763,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.012516667135059834,
"step": 90,
"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": 225.61666870117188,
"correct_distill_rate": 0.15000000596046448,
"distill_active_tokens": 34.21666793823242,
"distill_lambda": 0.6835,
"distill_loss": 0.7823053479194642,
"distill_pairs": 3343.916748046875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.013333333333333334,
"grad_norm": 0.6916502714157104,
"grpo_loss": 0.01803543260321021,
"kl": 0.450884810090065,
"learning_rate": 2e-06,
"loss": 0.556,
"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": 0.9944167017936707,
"reward_std": 0.4423261934891343,
"rewards/boxed_and_answer_tags_format_reward": 0.725,
"rewards/correctness_reward_func_math": 0.3,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.030583335598930717,
"step": 100,
"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": 238.1500030517578,
"correct_distill_rate": 0.45000000596046447,
"distill_active_tokens": 100.41666870117187,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.1814135193824766,
"distill_pairs": 7460.650366210937,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.014666666666666666,
"grad_norm": 0.3627510666847229,
"grpo_loss": 0.018571638874709605,
"kl": 0.4642893671989441,
"learning_rate": 2e-06,
"loss": 1.5456,
"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": 1.5331500589847564,
"reward_std": 0.37911299923434855,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.9,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.06684999838471413,
"step": 110,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 51.2,
"teacher_step_min": 51.2,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 210.9166702270508,
"correct_distill_rate": 0.4500000014901161,
"distill_active_tokens": 86.23333625793457,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.066215443611145,
"distill_pairs": 6689.233447265625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.016,
"grad_norm": 0.5176424384117126,
"grpo_loss": 0.014304549526423216,
"kl": 0.35761232376098634,
"learning_rate": 2e-06,
"loss": 1.4607,
"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": 1.6255833804607391,
"reward_std": 0.27584579698741435,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.9000000029802322,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.016083330987021326,
"step": 120,
"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": 215.26666870117188,
"correct_distill_rate": 0.25000000149011614,
"distill_active_tokens": 59.716667556762694,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.3513760089874267,
"distill_pairs": 3717.883337402344,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.017333333333333333,
"grad_norm": 0.8967616558074951,
"grpo_loss": 0.01419508489780128,
"kl": 0.3548759162425995,
"learning_rate": 2e-06,
"loss": 0.9602,
"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": 1.2300833761692047,
"reward_std": 0.25678171738982203,
"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": -0.01991666592657566,
"step": 130,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 16.0,
"teacher_step_min": 16.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 228.95000305175782,
"correct_distill_rate": 0.46666666865348816,
"distill_active_tokens": 105.43333587646484,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.3538686275482177,
"distill_pairs": 8601.23359375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.018666666666666668,
"grad_norm": 0.384183794260025,
"grpo_loss": 0.013619191199541091,
"kl": 0.3404792442917824,
"learning_rate": 2e-06,
"loss": 1.6613,
"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": 1.6679500222206116,
"reward_std": 0.37572124116122724,
"rewards/boxed_and_answer_tags_format_reward": 0.7333333373069764,
"rewards/correctness_reward_func_math": 0.9333333373069763,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.001283331774175167,
"step": 140,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 28.8,
"teacher_step_min": 28.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 240.01667022705078,
"correct_distill_rate": 0.25000000447034837,
"distill_active_tokens": 62.86666831970215,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.3873838663101197,
"distill_pairs": 5325.93349609375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.02,
"grad_norm": 2.4878554344177246,
"grpo_loss": 0.01858982089906931,
"kl": 0.4647453084588051,
"learning_rate": 2e-06,
"loss": 0.9898,
"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": 1.1181333482265472,
"reward_std": 0.3490286782383919,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.5000000089406967,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.08186667133122683,
"step": 150,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 12.8,
"teacher_step_min": 12.8,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 215.6500045776367,
"correct_distill_rate": 0.40000001043081285,
"distill_active_tokens": 86.63333625793457,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.0749982595443726,
"distill_pairs": 6986.433569335937,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.021333333333333333,
"grad_norm": 1.990696668624878,
"grpo_loss": 0.012278367765247822,
"kl": 0.3069575861096382,
"learning_rate": 2e-06,
"loss": 1.4648,
"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": 1.5073167324066161,
"reward_std": 0.6222962126135826,
"rewards/boxed_and_answer_tags_format_reward": 0.7333333373069764,
"rewards/correctness_reward_func_math": 0.7999999970197678,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.026016666553914547,
"step": 160,
"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": 228.46667022705077,
"correct_distill_rate": 0.28333333879709244,
"distill_active_tokens": 59.10000114440918,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.5006764113903046,
"distill_pairs": 3995.8335693359377,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.02266666666666667,
"grad_norm": 0.49599167704582214,
"grpo_loss": 0.017388735432177782,
"kl": 0.434717458486557,
"learning_rate": 2e-06,
"loss": 1.0679,
"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": 1.227750027179718,
"reward_std": 0.4114114996045828,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.566666665673256,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.08891666708514094,
"step": 170,
"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": 233.68333587646484,
"correct_distill_rate": 0.4000000014901161,
"distill_active_tokens": 90.3166675567627,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.0341420412063598,
"distill_pairs": 7434.483471679688,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.024,
"grad_norm": 0.2026839554309845,
"grpo_loss": 0.012693088967353106,
"kl": 0.3173263043165207,
"learning_rate": 2e-06,
"loss": 1.4366,
"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": 1.4828666865825653,
"reward_std": 0.32966375835239886,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.8000000029802322,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.06713333297520876,
"step": 180,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 44.8,
"teacher_step_min": 44.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 218.7833335876465,
"correct_distill_rate": 0.45000001341104506,
"distill_active_tokens": 95.80000381469726,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.4878697872161863,
"distill_pairs": 7501.0001953125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.025333333333333333,
"grad_norm": 3.5202090740203857,
"grpo_loss": 0.011401048023253679,
"kl": 0.2850246116518974,
"learning_rate": 2e-06,
"loss": 1.7529,
"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": 1.5406333923339843,
"reward_std": 0.5798586037009954,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.9000000029802322,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.059366670157760384,
"step": 190,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 44.8,
"teacher_step_min": 44.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 180.80000534057618,
"correct_distill_rate": 0.6166666731238365,
"distill_active_tokens": 102.3166690826416,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.5358146905899046,
"distill_pairs": 6516.883514404297,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.02666666666666667,
"grad_norm": 2.7591538429260254,
"grpo_loss": 0.012810665648430586,
"kl": 0.32026459723711015,
"learning_rate": 2e-06,
"loss": 1.7879,
"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": 2.041133427619934,
"reward_std": 0.4327510800678283,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 1.233333334326744,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.05779999829828739,
"step": 200,
"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": 218.3000045776367,
"correct_distill_rate": 0.1500000014901161,
"distill_active_tokens": 30.883333587646483,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.7339631259441376,
"distill_pairs": 2758.4500732421875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.028,
"grad_norm": 0.7159352898597717,
"grpo_loss": 0.023589813243597747,
"kl": 0.5897454500198365,
"learning_rate": 2e-06,
"loss": 0.5374,
"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": 1.0282666563987732,
"reward_std": 0.44314303873106836,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.3000000029802322,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.01339999851770699,
"step": 210,
"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": 230.63333435058593,
"correct_distill_rate": 0.11666667014360428,
"distill_active_tokens": 25.650000762939452,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.6510390162467956,
"distill_pairs": 2181.81669921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.029333333333333333,
"grad_norm": 1.5977497100830078,
"grpo_loss": 0.012654812633991241,
"kl": 0.3163697272539139,
"learning_rate": 2e-06,
"loss": 0.4684,
"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": 0.9319500207901001,
"reward_std": 0.33822232224047183,
"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": -0.05138333383947611,
"step": 220,
"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": 221.8166702270508,
"correct_distill_rate": 0.3666666686534882,
"distill_active_tokens": 70.33333358764648,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.607384157180786,
"distill_pairs": 5024.066796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.030666666666666665,
"grad_norm": 0.18692904710769653,
"grpo_loss": 0.01416043252684176,
"kl": 0.3540095329284668,
"learning_rate": 2e-06,
"loss": 1.1393,
"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": 1.4528000414371491,
"reward_std": 0.4694087415933609,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.7333333373069764,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.030533331260085107,
"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": 212.55000381469728,
"correct_distill_rate": 0.5166666701436042,
"distill_active_tokens": 93.28333549499511,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.204019916057587,
"distill_pairs": 6631.516772460937,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.6,
"epoch": 0.032,
"grad_norm": 2.6596450805664062,
"grpo_loss": 0.013668722473084927,
"kl": 0.34171686619520186,
"learning_rate": 2e-06,
"loss": 1.5565,
"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": 1.7840333938598634,
"reward_std": 0.3311736247502267,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 1.0333333402872085,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.0007000008132308722,
"step": 240,
"teacher_correct_rate": 0.6,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 210.60000381469726,
"correct_distill_rate": 0.4000000059604645,
"distill_active_tokens": 92.05000152587891,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.190346908569336,
"distill_pairs": 6130.3501953125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.03333333333333333,
"grad_norm": 2.9248268604278564,
"grpo_loss": 0.016169977188110352,
"kl": 0.404248882830143,
"learning_rate": 2e-06,
"loss": 1.5494,
"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": 1.4574000239372253,
"reward_std": 0.3499408717267215,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.8,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.04260000009089708,
"step": 250,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 12.8,
"teacher_step_min": 12.8,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 229.01667175292968,
"correct_distill_rate": 0.333333346247673,
"distill_active_tokens": 79.05000152587891,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.9633725166320801,
"distill_pairs": 7036.683544921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.034666666666666665,
"grad_norm": 3.324446678161621,
"grpo_loss": 0.023108621640130878,
"kl": 0.5777149647474289,
"learning_rate": 2e-06,
"loss": 1.3975,
"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": 1.208383357524872,
"reward_std": 0.45492203570902345,
"rewards/boxed_and_answer_tags_format_reward": 0.6166666656732559,
"rewards/correctness_reward_func_math": 0.6666666567325592,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.07494999822229147,
"step": 260,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.2
},
{
"clip_ratio": 0.0,
"completion_length": 222.1500030517578,
"correct_distill_rate": 0.3833333447575569,
"distill_active_tokens": 87.11666908264161,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.8918882012367249,
"distill_pairs": 7470.216943359375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.036,
"grad_norm": 3.3958823680877686,
"grpo_loss": 0.015124699426814913,
"kl": 0.37811670303344724,
"learning_rate": 2e-06,
"loss": 1.3394,
"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": 1.5167166948318482,
"reward_std": 0.4619201775640249,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7666666656732559,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.00838333424180746,
"step": 270,
"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": 227.60000457763672,
"correct_distill_rate": 0.25000000596046446,
"distill_active_tokens": 54.78333587646485,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.2441058158874512,
"distill_pairs": 3980.416796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.037333333333333336,
"grad_norm": 5.159823894500732,
"grpo_loss": 0.012744125351309777,
"kl": 0.31860269606113434,
"learning_rate": 2e-06,
"loss": 0.8836,
"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": 1.1811166822910308,
"reward_std": 0.25040809661149976,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.5,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.06054999846965074,
"step": 280,
"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": 200.88333587646486,
"correct_distill_rate": 0.10000000298023223,
"distill_active_tokens": 18.56666717529297,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.47246934175491334,
"distill_pairs": 1459.8333740234375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.0,
"epoch": 0.03866666666666667,
"grad_norm": 0.7432988882064819,
"grpo_loss": 0.009340547863394022,
"kl": 0.23351338356733323,
"learning_rate": 2e-06,
"loss": 0.3401,
"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": 0.881083345413208,
"reward_std": 0.36734714061021806,
"rewards/boxed_and_answer_tags_format_reward": 0.6916666686534881,
"rewards/correctness_reward_func_math": 0.20000000596046447,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.010583332390524447,
"step": 290,
"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": 218.1666687011719,
"correct_distill_rate": 0.2833333402872086,
"distill_active_tokens": 62.45,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.5834805727005006,
"distill_pairs": 6097.5501953125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.04,
"grad_norm": 0.13191722333431244,
"grpo_loss": 0.010767972096800804,
"kl": 0.2691974639892578,
"learning_rate": 2e-06,
"loss": 1.1192,
"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": 1.263183408975601,
"reward_std": 0.35227395556867125,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.5666666686534881,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.053483329620212315,
"step": 300,
"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": 203.3000030517578,
"correct_distill_rate": 0.3333333358168602,
"distill_active_tokens": 57.91666831970215,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.4135151922702789,
"distill_pairs": 4799.150146484375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.04133333333333333,
"grad_norm": 2.529168128967285,
"grpo_loss": 0.016041170246899127,
"kl": 0.40102770626544954,
"learning_rate": 2e-06,
"loss": 1.0055,
"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": 1.3594500601291657,
"reward_std": 0.40493726804852487,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.6666666716337204,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.05721666254103184,
"step": 310,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 28.8,
"teacher_step_min": 28.8,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 209.95000305175782,
"correct_distill_rate": 0.25000000149011614,
"distill_active_tokens": 39.033334732055664,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.9276914298534393,
"distill_pairs": 2685.76669921875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.042666666666666665,
"grad_norm": 0.46919429302215576,
"grpo_loss": 0.013499792478978635,
"kl": 0.33749391585588456,
"learning_rate": 2e-06,
"loss": 0.6629,
"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": 1.1758000373840332,
"reward_std": 0.2618331216275692,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.5000000029802323,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.02419999921694398,
"step": 320,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 28.8,
"teacher_step_min": 28.8,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 210.9166702270508,
"correct_distill_rate": 0.3333333387970924,
"distill_active_tokens": 60.68333435058594,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.3985709071159362,
"distill_pairs": 5096.116748046875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.044,
"grad_norm": 4.798004150390625,
"grpo_loss": 0.015208367817103862,
"kl": 0.3802085354924202,
"learning_rate": 2e-06,
"loss": 0.9942,
"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": 1.4078166902065277,
"reward_std": 0.5185918163508176,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.6666666656732559,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.000516666192561388,
"step": 330,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 32.0,
"teacher_step_min": 32.0,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 244.45000305175782,
"correct_distill_rate": 0.08333333432674409,
"distill_active_tokens": 20.21666717529297,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.40702306032180785,
"distill_pairs": 1509.383349609375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.04533333333333334,
"grad_norm": 0.13194796442985535,
"grpo_loss": 0.01111379754729569,
"kl": 0.2778447732329369,
"learning_rate": 2e-06,
"loss": 0.296,
"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": 0.832316666841507,
"reward_std": 0.35398513115942476,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.16666666865348817,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.03434999953024089,
"step": 340,
"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": 200.6500015258789,
"correct_distill_rate": 0.3500000059604645,
"distill_active_tokens": 42.16666717529297,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.0164209842681884,
"distill_pairs": 3142.90009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.04666666666666667,
"grad_norm": 0.17235389351844788,
"grpo_loss": 0.013477419409900904,
"kl": 0.33693354576826096,
"learning_rate": 2e-06,
"loss": 0.725,
"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": 1.4807834088802339,
"reward_std": 0.22941935565322638,
"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": 0.030783333256840704,
"step": 350,
"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": 194.60000381469726,
"correct_distill_rate": 0.2833333358168602,
"distill_active_tokens": 60.95000057220459,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.4580639809370042,
"distill_pairs": 3961.4500366210937,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.048,
"grad_norm": 2.2508678436279297,
"grpo_loss": 0.011749111372046173,
"kl": 0.29372718334198,
"learning_rate": 2e-06,
"loss": 1.0324,
"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": 1.276500016450882,
"reward_std": 0.25505427923053503,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.5666666716337204,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.009833332989364863,
"step": 360,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 9.6,
"teacher_step_min": 9.6,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 192.3000045776367,
"correct_distill_rate": 0.31666667461395265,
"distill_active_tokens": 54.20000228881836,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.2620195150375366,
"distill_pairs": 3882.866796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.04933333333333333,
"grad_norm": 4.451849937438965,
"grpo_loss": 0.014936284627765417,
"kl": 0.37340673953294756,
"learning_rate": 2e-06,
"loss": 0.8983,
"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": 1.4263333618640899,
"reward_std": 0.38016765341162684,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.6333333253860474,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.04300000071525574,
"step": 370,
"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": 206.48333587646485,
"correct_distill_rate": 0.3666666686534882,
"distill_active_tokens": 74.63333435058594,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.7010697364807128,
"distill_pairs": 5580.30009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.050666666666666665,
"grad_norm": 1.5853809118270874,
"grpo_loss": 0.027918000333011152,
"kl": 0.6979479655623436,
"learning_rate": 2e-06,
"loss": 1.2187,
"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": 1.4619667470455169,
"reward_std": 0.18888297900557519,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7333333373069764,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.013033336028456688,
"step": 380,
"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": 190.36667175292968,
"correct_distill_rate": 0.16666666865348817,
"distill_active_tokens": 32.16666793823242,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.7179160356521607,
"distill_pairs": 2329.966748046875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.052,
"grad_norm": 0.4185982346534729,
"grpo_loss": 0.01336874281987548,
"kl": 0.3342174395918846,
"learning_rate": 2e-06,
"loss": 0.5159,
"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": 1.1135000467300415,
"reward_std": 0.1832519941031933,
"rewards/boxed_and_answer_tags_format_reward": 0.7333333373069764,
"rewards/correctness_reward_func_math": 0.33333333730697634,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.04683333467692137,
"step": 390,
"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": 212.3166717529297,
"correct_distill_rate": 0.31666666716337205,
"distill_active_tokens": 51.816668319702146,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.2154293179512023,
"distill_pairs": 4320.98359375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.05333333333333334,
"grad_norm": 0.3377378284931183,
"grpo_loss": 0.010715312231332064,
"kl": 0.2678819209337234,
"learning_rate": 2e-06,
"loss": 0.8615,
"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": 1.3341833770275116,
"reward_std": 0.15651564560830594,
"rewards/boxed_and_answer_tags_format_reward": 0.7333333313465118,
"rewards/correctness_reward_func_math": 0.6333333343267441,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.032483333721756935,
"step": 400,
"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": 179.51667022705078,
"correct_distill_rate": 0.6000000089406967,
"distill_active_tokens": 93.48333511352538,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.9986286878585815,
"distill_pairs": 8239.85029296875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.6,
"epoch": 0.05466666666666667,
"grad_norm": 0.5718596577644348,
"grpo_loss": 0.013780045555904508,
"kl": 0.344500358402729,
"learning_rate": 2e-06,
"loss": 1.4128,
"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": 2.000366687774658,
"reward_std": 0.3066445928707253,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 1.1999999940395356,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.058700001891702415,
"step": 410,
"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": 192.3166717529297,
"correct_distill_rate": 0.11666667014360428,
"distill_active_tokens": 22.8,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.589948958158493,
"distill_pairs": 1178.8,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.056,
"grad_norm": 0.45858168601989746,
"grpo_loss": 0.011721086828038097,
"kl": 0.29302656948566436,
"learning_rate": 2e-06,
"loss": 0.4247,
"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": 0.9488000333309173,
"reward_std": 0.33747670613229275,
"rewards/boxed_and_answer_tags_format_reward": 0.6916666686534881,
"rewards/correctness_reward_func_math": 0.23333334028720856,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.023799999861512332,
"step": 420,
"teacher_correct_rate": 0.1,
"teacher_step_mean": 6.4,
"teacher_step_min": 6.4,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 198.90000762939454,
"correct_distill_rate": 0.38333333730697633,
"distill_active_tokens": 67.88333511352539,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.431118631362915,
"distill_pairs": 5665.45009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.05733333333333333,
"grad_norm": 3.6496245861053467,
"grpo_loss": 0.017165888380259276,
"kl": 0.42914492785930636,
"learning_rate": 2e-06,
"loss": 1.0189,
"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": 1.5596167623996735,
"reward_std": 0.34454694148153064,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.7666666626930236,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.042950001487042755,
"step": 430,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 41.6,
"teacher_step_min": 41.6,
"zero_std_ratio": 0.0
},
{
"clip_ratio": 0.0,
"completion_length": 197.6500015258789,
"correct_distill_rate": 0.23333333432674408,
"distill_active_tokens": 46.51666793823242,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.0730075359344482,
"distill_pairs": 3469.75009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.058666666666666666,
"grad_norm": 0.2143731564283371,
"grpo_loss": 0.008997786138206721,
"kl": 0.22494387105107308,
"learning_rate": 2e-06,
"loss": 0.7601,
"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": 1.1525666952133178,
"reward_std": 0.16749067343771457,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.46666666865348816,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.014099998027086258,
"step": 440,
"teacher_correct_rate": 0.2,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 224.5500061035156,
"correct_distill_rate": 0.0833333358168602,
"distill_active_tokens": 20.800000381469726,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.5073355913162232,
"distill_pairs": 2232.000048828125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.1,
"epoch": 0.06,
"grad_norm": 0.13713976740837097,
"grpo_loss": 0.007414947217330337,
"kl": 0.18537293151021003,
"learning_rate": 2e-06,
"loss": 0.3625,
"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": 0.8447333693504333,
"reward_std": 0.32054444095119833,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.1666666716337204,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.021933332830667496,
"step": 450,
"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": 204.68333587646484,
"correct_distill_rate": 0.3833333343267441,
"distill_active_tokens": 63.150001525878906,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.4486887335777283,
"distill_pairs": 4682.45009765625,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.06133333333333333,
"grad_norm": 0.13258688151836395,
"grpo_loss": 0.011527160182595253,
"kl": 0.2881775885820389,
"learning_rate": 2e-06,
"loss": 1.0256,
"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": 1.501816737651825,
"reward_std": 0.2814405869692564,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.7666666686534882,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.035149999614804986,
"step": 460,
"teacher_correct_rate": 0.5,
"teacher_step_mean": 28.8,
"teacher_step_min": 28.8,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 222.86666870117188,
"correct_distill_rate": 0.3666666731238365,
"distill_active_tokens": 79.66666755676269,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.5911810755729676,
"distill_pairs": 5118.200122070312,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.06266666666666666,
"grad_norm": 3.8933823108673096,
"grpo_loss": 0.010026454040780664,
"kl": 0.25066058039665223,
"learning_rate": 2e-06,
"loss": 1.1239,
"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": 1.4889000415802003,
"reward_std": 0.3256177700124681,
"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": 0.013899997001863085,
"step": 470,
"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": 231.0666702270508,
"correct_distill_rate": 0.13333333432674407,
"distill_active_tokens": 34.13333358764648,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.6651593446731567,
"distill_pairs": 2833.066796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.064,
"grad_norm": 0.6032183766365051,
"grpo_loss": 0.007583877909928561,
"kl": 0.18959567099809646,
"learning_rate": 2e-06,
"loss": 0.4732,
"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": 0.9517500519752502,
"reward_std": 0.23708552001044153,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.26666666865348815,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.05658333429601044,
"step": 480,
"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": 213.98333740234375,
"correct_distill_rate": 0.3666666761040688,
"distill_active_tokens": 75.13333587646484,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.8703262329101562,
"distill_pairs": 6137.13359375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.6,
"epoch": 0.06533333333333333,
"grad_norm": 3.2628140449523926,
"grpo_loss": 0.01136380508542061,
"kl": 0.2840937629342079,
"learning_rate": 2e-06,
"loss": 1.3206,
"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": 1.3915000557899475,
"reward_std": 0.48741065794602034,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.7333333402872085,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.04183333655819297,
"step": 490,
"teacher_correct_rate": 0.6,
"teacher_step_mean": 44.8,
"teacher_step_min": 44.8,
"zero_std_ratio": 0.1
},
{
"clip_ratio": 0.0,
"completion_length": 221.8166732788086,
"correct_distill_rate": 0.20000000894069672,
"distill_active_tokens": 39.65000228881836,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.0484423875808715,
"distill_pairs": 2248.350048828125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.06666666666666667,
"grad_norm": 0.2131481021642685,
"grpo_loss": 0.01141567649319768,
"kl": 0.28539085686206817,
"learning_rate": 2e-06,
"loss": 0.7453,
"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": 1.1578667104244231,
"reward_std": 0.3293949570506811,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.3999999940395355,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.007866668631322682,
"step": 500,
"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": 212.03333587646483,
"correct_distill_rate": 0.4000000074505806,
"distill_active_tokens": 62.81666717529297,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.41912682056427,
"distill_pairs": 5868.3833984375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.5,
"epoch": 0.068,
"grad_norm": 3.680525541305542,
"grpo_loss": 0.012760422751307487,
"kl": 0.31900972276926043,
"learning_rate": 2e-06,
"loss": 1.0061,
"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": 1.5433500170707704,
"reward_std": 0.3890270760282874,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.8000000029802322,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.006650000438094139,
"step": 510,
"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": 223.73333587646485,
"correct_distill_rate": 0.16666666865348817,
"distill_active_tokens": 31.166667175292968,
"distill_lambda": 0.7000000000000001,
"distill_loss": 0.7643749833106994,
"distill_pairs": 2527.766796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.06933333333333333,
"grad_norm": 1.63319730758667,
"grpo_loss": 0.00907530221156776,
"kl": 0.22688169702887534,
"learning_rate": 2e-06,
"loss": 0.5441,
"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": 1.0498000383377075,
"reward_std": 0.5366952225565911,
"rewards/boxed_and_answer_tags_format_reward": 0.75,
"rewards/correctness_reward_func_math": 0.33333333730697634,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.03353333636187017,
"step": 520,
"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": 209.01666870117188,
"correct_distill_rate": 0.4000000089406967,
"distill_active_tokens": 74.86666946411133,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.605245339870453,
"distill_pairs": 6138.466845703125,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.07066666666666667,
"grad_norm": 3.3146023750305176,
"grpo_loss": 0.009697011020034552,
"kl": 0.24242507368326188,
"learning_rate": 2e-06,
"loss": 1.1334,
"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": 1.4763500094413757,
"reward_std": 0.32808963749557735,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.7999999940395355,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.02365000396966934,
"step": 530,
"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": 212.73333663940429,
"correct_distill_rate": 0.31666667461395265,
"distill_active_tokens": 63.466668701171876,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.3980355978012085,
"distill_pairs": 4859.200146484375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.072,
"grad_norm": 3.232771158218384,
"grpo_loss": 0.008403513114899397,
"kl": 0.2100867137312889,
"learning_rate": 2e-06,
"loss": 0.987,
"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": 1.3306000292301179,
"reward_std": 0.44829795956611634,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.6333333373069763,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.002733335830271244,
"step": 540,
"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": 184.10000305175782,
"correct_distill_rate": 0.2666666701436043,
"distill_active_tokens": 52.58333396911621,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.251877450942993,
"distill_pairs": 4427.416674804687,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.07333333333333333,
"grad_norm": 2.370800495147705,
"grpo_loss": 0.013602001033723355,
"kl": 0.3400494694709778,
"learning_rate": 2e-06,
"loss": 0.8899,
"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": 1.3255000174045564,
"reward_std": 0.4342447152361274,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.5333333402872086,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": 0.05050000064074993,
"step": 550,
"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": 217.76667022705078,
"correct_distill_rate": 0.3666666761040688,
"distill_active_tokens": 80.05000228881836,
"distill_lambda": 0.7000000000000001,
"distill_loss": 2.0042829751968383,
"distill_pairs": 5829.816943359375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.07466666666666667,
"grad_norm": 3.586153745651245,
"grpo_loss": 0.011807580757886172,
"kl": 0.29518680945038794,
"learning_rate": 2e-06,
"loss": 1.4148,
"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": 1.4508001029491424,
"reward_std": 0.7100536167621613,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.7333333283662796,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.02419999975245446,
"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": 213.73333740234375,
"correct_distill_rate": 0.2333333373069763,
"distill_active_tokens": 45.800001525878905,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.1419212341308593,
"distill_pairs": 3603.266748046875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.076,
"grad_norm": 0.092085100710392,
"grpo_loss": 0.014549511345103383,
"kl": 0.3637384161353111,
"learning_rate": 2e-06,
"loss": 0.8139,
"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": 1.1262499690055847,
"reward_std": 0.28337307972833514,
"rewards/boxed_and_answer_tags_format_reward": 0.7,
"rewards/correctness_reward_func_math": 0.4666666626930237,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.040416667610406874,
"step": 570,
"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": 187.73333587646485,
"correct_distill_rate": 0.3833333417773247,
"distill_active_tokens": 72.25000038146973,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.6827910363674163,
"distill_pairs": 5868.016748046875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.3,
"epoch": 0.07733333333333334,
"grad_norm": 3.060189723968506,
"grpo_loss": 0.029745447728782893,
"kl": 0.7436354175209999,
"learning_rate": 2e-06,
"loss": 1.2077,
"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": 1.4130167007446288,
"reward_std": 0.48727381117641927,
"rewards/boxed_and_answer_tags_format_reward": 0.65,
"rewards/correctness_reward_func_math": 0.7666666716337204,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.003650001063942909,
"step": 580,
"teacher_correct_rate": 0.3,
"teacher_step_mean": 25.6,
"teacher_step_min": 25.6,
"zero_std_ratio": 0.2
},
{
"clip_ratio": 0.0,
"completion_length": 225.63333740234376,
"correct_distill_rate": 0.3000000029802322,
"distill_active_tokens": 53.90000228881836,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.3403135061264038,
"distill_pairs": 4990.366796875,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.2,
"epoch": 0.07866666666666666,
"grad_norm": 0.14780326187610626,
"grpo_loss": 0.01263717208057642,
"kl": 0.3159283846616745,
"learning_rate": 2e-06,
"loss": 0.9509,
"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": 1.3194500505924225,
"reward_std": 0.25968616977334025,
"rewards/boxed_and_answer_tags_format_reward": 0.7416666686534882,
"rewards/correctness_reward_func_math": 0.6000000059604644,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.022216665698215364,
"step": 590,
"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": 225.3166702270508,
"correct_distill_rate": 0.33333334475755694,
"distill_active_tokens": 80.15000267028809,
"distill_lambda": 0.7000000000000001,
"distill_loss": 1.9326006054878235,
"distill_pairs": 6198.250146484375,
"distill_skipped_no_correct_rate": 0.0,
"eligible_block_rate": 0.4,
"epoch": 0.08,
"grad_norm": 0.32919612526893616,
"grpo_loss": 0.007417573081329465,
"kl": 0.18543810546398162,
"learning_rate": 2e-06,
"loss": 1.3602,
"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": 1.3179833829402923,
"reward_std": 0.6959338076412678,
"rewards/boxed_and_answer_tags_format_reward": 0.6833333313465119,
"rewards/correctness_reward_func_math": 0.6666666775941849,
"rewards/soft_format_reward_func": 0.0,
"rewards/xmlcount_reward_func": -0.032016669400036335,
"step": 600,
"teacher_correct_rate": 0.4,
"teacher_step_mean": 28.8,
"teacher_step_min": 28.8,
"zero_std_ratio": 0.0
}
],
"logging_steps": 10,
"max_steps": 600,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"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
}