Files
DualMind-TKD-Agentic-1.7B/training_artifacts/DualMind100_train_log.jsonl

60 lines
34 KiB
Plaintext
Raw Normal View History

{"epoch": 1, "batch_idx": 0, "global_step": 1, "opt_step": 0, "combined": 2.19950724, "ce": 2.19950724, "kd": 0.0, "gap_energy": 0.0, "kd_alpha": 0.0, "proof_weight": 1.99846939, "ce_num": 60.49145889, "ce_den": 27.50227737, "kd_num": 0.0, "kd_den": 1.0, "gap_num": 0.0, "gap_den": 1.0, "kd_tokens": 0.0, "teacher_topk_mass_num": 0.0, "student_topk_mass_num": 0.0, "mass_den": 1.0, "lr": 4.2857142857142857e-07, "difficulty": 1.7609130144119263, "memory": "alloc=14.0GiB reserved=18.9GiB peak=16.4GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 9, "global_step": 10, "opt_step": 2, "combined": 1.21293616, "ce": 1.21293616, "kd": 0.0, "gap_energy": 0.0, "kd_alpha": 0.0, "proof_weight": 1.98469388, "ce_num": 586.51037598, "ce_den": 483.54595947, "kd_num": 0.0, "kd_den": 1.0, "gap_num": 0.0, "gap_den": 1.0, "kd_tokens": 0.0, "teacher_topk_mass_num": 0.0, "student_topk_mass_num": 0.0, "mass_den": 1.0, "lr": 1.2857142857142856e-06, "difficulty": 6.203751564025879, "memory": "alloc=26.8GiB reserved=38.6GiB peak=35.6GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 19, "global_step": 20, "opt_step": 5, "combined": 1.07730448, "ce": 1.07730448, "kd": 0.0, "gap_energy": 0.0, "kd_alpha": 0.0, "proof_weight": 1.96938776, "ce_num": 1827.9753418, "ce_den": 1696.80480957, "kd_num": 0.0, "kd_den": 1.0, "gap_num": 0.0, "gap_den": 1.0, "kd_tokens": 0.0, "teacher_topk_mass_num": 0.0, "student_topk_mass_num": 0.0, "mass_den": 1.0, "lr": 2.571428571428571e-06, "difficulty": 7.312134742736816, "memory": "alloc=20.4GiB reserved=38.6GiB peak=35.6GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 29, "global_step": 30, "opt_step": 7, "combined": 0.97815299, "ce": 0.97815299, "kd": 0.0, "gap_energy": 0.0, "kd_alpha": 0.0, "proof_weight": 1.95408163, "ce_num": 2458.76733398, "ce_den": 2513.68383789, "kd_num": 0.0, "kd_den": 1.0, "gap_num": 0.0, "gap_den": 1.0, "kd_tokens": 0.0, "teacher_topk_mass_num": 0.0, "student_topk_mass_num": 0.0, "mass_den": 1.0, "lr": 2.999660117172384e-06, "difficulty": 8.063447952270508, "memory": "alloc=26.8GiB reserved=38.6GiB peak=35.6GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 39, "global_step": 40, "opt_step": 10, "combined": 0.80651736, "ce": 0.80651736, "kd": 0.0, "gap_energy": 0.0, "kd_alpha": 0.0, "proof_weight": 1.93877551, "ce_num": 3162.8203125, "ce_den": 3921.57739258, "kd_num": 0.0, "kd_den": 1.0, "gap_num": 0.0, "gap_den": 1.0, "kd_tokens": 0.0, "teacher_topk_mass_num": 0.0, "student_topk_mass_num": 0.0, "mass_den": 1.0, "lr": 2.9945652968935727e-06, "difficulty": 8.520319938659668, "memory": "alloc=20.4GiB reserved=38.6GiB peak=35.6GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 49, "global_step": 50, "opt_step": 12, "combined": 0.60383594, "ce": 0.60383594, "kd": 0.0, "gap_energy": 0.0, "kd_alpha": 0.0, "proof_weight": 1.92346939, "ce_num": 2507.10180664, "ce_den": 4151.95849609, "kd_num": 0.0, "kd_den": 1.0, "gap_num": 0.0, "gap_den": 1.0, "kd_tokens": 0.0, "teacher_topk_mass_num": 0.0, "student_topk_mass_num": 0.0, "mass_den": 1.0, "lr": 2.987782178386712e-06, "difficulty": 8.635387420654297, "memory": "alloc=26.8GiB reserved=38.6GiB peak=35.6GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 59, "global_step": 60, "opt_step": 15, "combined": 0.48840919, "ce": 0.48823461, "kd": 0.53238404, "gap_energy": 0.13432807, "kd_alpha": 0.0039548, "proof_weight": 1.90816327, "ce_num": 2901.15576172, "ce_den": 5942.13476562, "kd_num": 3163.49755859, "kd_den": 5942.13476562, "gap_num": 797.59191895, "gap_den": 5937.64160156, "kd_tokens": 2650.0, "teacher_topk_mass_num": 2431.18603516, "student_topk_mass_num": 2339.93945312, "mass_den": 2650.0, "lr": 2.9725617878713683e-06, "difficulty": 9.01607894897461, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 69, "global_step": 70, "opt_step": 17, "combined": 0.64849538, "ce": 0.64515531, "kd": 0.78591424, "gap_energy": 0.12753046, "kd_alpha": 0.02372881, "proof_weight": 1.89285714, "ce_num": 4418.65429688, "ce_den": 6848.97753906, "kd_num": 5382.70898438, "kd_den": 6848.97753906, "gap_num": 872.86529541, "gap_den": 6844.3671875, "kd_tokens": 3039.0, "teacher_topk_mass_num": 2719.81274414, "student_topk_mass_num": 2384.62744141, "mass_den": 3039.0, "lr": 2.9590808525085274e-06, "difficulty": 9.196342468261719, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 79, "global_step": 80, "opt_step": 20, "combined": 0.45821229, "ce": 0.44078094, "kd": 0.84147513, "gap_energy": 0.19842488, "kd_alpha": 0.04350282, "proof_weight": 1.87755102, "ce_num": 3115.63183594, "ce_den": 7068.43603516, "kd_num": 5947.91308594, "kd_den": 7068.43603516, "gap_num": 1401.50024414, "gap_den": 7063.12792969, "kd_tokens": 2963.0, "teacher_topk_mass_num": 2739.51367188, "student_topk_mass_num": 2400.4050293, "mass_den": 2963.0, "lr": 2.9339262969984575e-06, "difficulty": 9.049243927001953, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 89, "global_step": 90, "opt_step": 22, "combined": 0.29784098, "ce": 0.2527481, "kd": 0.96537668, "gap_energy": 0.30821687, "kd_alpha": 0.06327684, "proof_weight": 1.8622449, "ce_num": 1589.84631348, "ce_den": 6290.24023438, "kd_num": 6072.45117188, "kd_den": 6290.24023438, "gap_num": 1937.32543945, "gap_den": 6285.59179688, "kd_tokens": 2697.0, "teacher_topk_mass_num": 2572.86108398, "student_topk_mass_num": 2246.55224609, "mass_den": 2697.0, "lr": 2.913917075363645e-06, "difficulty": 9.136136054992676, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 99, "global_step": 100, "opt_step": 25, "combined": 0.65063834, "ce": 0.62219143, "kd": 0.96471536, "gap_energy": 0.18682474, "kd_alpha": 0.08305085, "proof_weight": 1.84693878, "ce_num": 3035.79345703, "ce_den": 4879.1953125, "kd_num": 4707.03466797, "kd_den": 4879.1953125, "gap_num": 910.70910645, "gap_den": 4874.67089844, "kd_tokens": 2399.0, "teacher_topk_mass_num": 1882.3659668, "student_topk_mass_num": 1627.14233398, "mass_den": 2399.0, "lr": 2.8791446879894015e-06, "difficulty": 8.839614868164062, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 109, "global_step": 110, "opt_step": 27, "combined": 0.51954746, "ce": 0.49451286, "kd": 0.73798132, "gap_energy": 0.18867157, "kd_alpha": 0.10282486, "proof_weight": 1.83163265, "ce_num": 3955.71142578, "ce_den": 7999.20849609, "kd_num": 5903.26660156, "kd_den": 7999.20849609, "gap_num": 1508.29418945, "gap_den": 7994.28417969, "kd_tokens": 3590.0, "teacher_topk_mass_num": 3250.93408203, "student_topk_mass_num": 2921.56396484, "mass_den": 3590.0, "lr": 2.8528588076542965e-06, "difficulty": 9.443330764770508, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 119, "global_step": 120, "opt_step": 30, "combined": 0.35144728, "ce": 0.3200354, "kd": 0.57625228, "gap_energy": 0.0713951, "kd_alpha": 0.12259887, "proof_weight": 1.81632653, "ce_num": 2391.49194336, "ce_den": 7472.5859375, "kd_num": 4306.09472656, "kd_den": 7472.5859375, "gap_num": 533.13110352, "gap_den": 7467.33447266, "kd_tokens": 3282.0, "teacher_topk_mass_num": 2973.54248047, "student_topk_mass_num": 2701.47192383, "mass_den": 3282.0, "lr": 2.8089058713625192e-06, "difficulty": 9.33741569519043, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 129, "global_step": 130, "opt_step": 32, "combined": 0.61775106, "ce": 0.61014116, "kd": 0.6635918, "gap_energy": 0.10788401, "kd_alpha": 0.14237288, "proof_weight": 1.80102041, "ce_num": 5045.71289062, "ce_den": 8269.74707031, "kd_num": 5487.73632812, "kd_den": 8269.74707031, "gap_num": 891.60717773, "gap_den": 8264.49804688, "kd_tokens": 3530.0, "teacher_topk_mass_num": 2828.89599609, "student_topk_mass_num": 2750.65283203, "mass_den": 3530.0, "lr": 2.7766738925237584e-06, "difficulty": 9.378902435302734, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 139, "global_step": 140, "opt_step": 35, "combined": 0.46766806, "ce": 0.40705362, "kd": 0.78087783, "gap_energy": 0.33781546, "kd_alpha": 0.16214689, "proof_weight": 1.78571429, "ce_num": 3316.54199219, "ce_den": 8147.67871094, "kd_num": 6362.34179688, "kd_den": 8147.67871094, "gap_num": 2750.86181641, "gap_den": 8143.09033203, "kd_tokens": 3491.0, "teacher_topk_mass_num": 3209.97265625, "student_topk_mass_num": 3108.03955078, "mass_den": 3491.0, "lr": 2.7240931409904047e-06, "difficulty": 9.392586708068848, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 149, "global_step": 150, "opt_step": 37, "combined": 0.53738558, "ce": 0.45108336, "kd": 0.92547762, "gap_energy": 0.65703124, "kd_alpha": 0.1819209, "proof_weight": 1.77040816, "ce_num": 3765.02099609, "ce_den": 8346.61914062, "kd_num": 7724.609375, "kd_den": 8346.61914062, "gap_num": 5481.05664062, "gap_den": 8342.15527344, "kd_tokens": 3726.0, "teacher_topk_mass_num": 3358.96166992, "student_topk_mass_num": 3260.4050293, "mass_den": 3726.0, "lr": 2.6863203994833153e-06, "difficulty": 9.397011756896973, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 159, "global_step": 160, "opt_step": 40, "combined": 0.34333396, "ce": 0.26012513, "kd": 0.67267311, "gap_energy": 0.32599863, "kd_alpha": 0.20169492, "proof_weight": 1.75510204, "ce_num": 2292.73486328, "ce_den": 8813.96875, "kd_num": 5928.91992188, "kd_den": 8813.96875, "gap_num": 2871.65820312, "gap_den": 8808.8046875, "kd_tokens": 3818.0, "teacher_topk_mass_num": 3620.66650391, "student_topk_mass_num": 3510.72509766, "mass_den": 3818.0, "lr": 2.625773066166979e-06, "difficulty": 9.601572036743164, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 169, "global_step": 170, "opt_step": 42, "combined": 0.28035346, "ce": 0.27071178, "kd": 0.31424689, "gap_energy": 0.05172772, "kd_alpha": 0.22146893, "proof_weight": 1.73979592, "ce_num": 2396.01489258, "ce_den": 8850.796875, "kd_num": 2781.33544922, "kd_den": 8850.796875, "gap_num": 457.55905151, "gap_den": 8845.52832031, "kd_tokens": 3818.0, "teacher_topk_mass_num": 3658.04345703, "student_topk_mass_num": 3594.16357422, "mass_den": 3818.0, "lr": 2.5829345761322675e-06, "difficulty": 9.6195068359375, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 179, "global_step": 180, "opt_step": 45, "combined": 0.25831455, "ce": 0.2564809, "kd": 0.26408181, "gap_energy": 0.02563814, "kd_alpha": 0.24124294, "proof_weight": 1.7244898, "ce_num": 2221.45703125, "ce_den": 8661.29589844, "kd_num": 2287.29077148, "kd_den": 8661.29589844, "gap_num": 221.94287109, "gap_den": 8656.74804688, "kd_tokens": 3748.0, "teacher_topk_mass_num": 3553.6015625, "student_topk_mass_num": 3502.31494141, "mass_den": 3748.0, "lr": 2.5151820788797823e-06, "difficulty": 9.612282752990723, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 1, "batch_idx": 189, "global_step": 190, "opt_step": 47, "combined": 0.4030064, "ce": 0.38660949, "kd": 0.44942892, "gap_energy": 0.06206001, "kd_alpha": 0.26101695, "proof_weight": 1.70918367, "ce_num": 3204.91943359, "ce_den": 8289.81054688, "kd_num": 3725.68066406, "kd_den": 8289.81054688, "gap_num": 514.19836426, "gap_den": 8285.50292969, "kd_tokens": 3865.0, "teacher_topk_mass_num": 3518.7331543, "student_topk_mass_num": 3433.62817383, "mass_den": 3865.0, "lr": 2.4678165591851067e-06, "difficulty": 9.535543441772461, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 3, "global_step": 200, "opt_step": 50, "combined": 1.24225593, "ce": 1.39534032, "kd": 0.85015059, "gap_energy": 0.269521, "kd_alpha": 0.28079096, "proof_weight": 1.69387755, "ce_num": 446.31286621, "ce_den": 319.85949707, "kd_num": 271.92874146, "kd_den": 319.85949707, "gap_num": 85.47875977, "gap_den": 317.15066528, "kd_tokens": 246.0, "teacher_topk_mass_num": 142.12185669, "student_topk_mass_num": 135.27420044, "mass_den": 246.0, "lr": 2.3937109249603387e-06, "difficulty": 5.464296340942383, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 13, "global_step": 210, "opt_step": 52, "combined": 0.66685486, "ce": 0.63092232, "kd": 0.75047231, "gap_energy": 0.27310866, "kd_alpha": 0.30056497, "proof_weight": 1.67857143, "ce_num": 396.93060303, "ce_den": 629.12756348, "kd_num": 472.14282227, "kd_den": 629.12756348, "gap_num": 171.09176636, "gap_den": 626.46044922, "kd_tokens": 479.0, "teacher_topk_mass_num": 378.57348633, "student_topk_mass_num": 363.54272461, "mass_den": 479.0, "lr": 2.3424140244979735e-06, "difficulty": 6.788113594055176, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 23, "global_step": 220, "opt_step": 55, "combined": 0.59748399, "ce": 0.58014715, "kd": 0.63426751, "gap_energy": 0.12397655, "kd_alpha": 0.32033898, "proof_weight": 1.66326531, "ce_num": 1008.77160645, "ce_den": 1738.8203125, "kd_num": 1102.87719727, "kd_den": 1738.8203125, "gap_num": 215.08178711, "gap_den": 1734.85864258, "kd_tokens": 1021.0, "teacher_topk_mass_num": 801.39776611, "student_topk_mass_num": 810.95361328, "mass_den": 1021.0, "lr": 2.2628871746483883e-06, "difficulty": 7.726119041442871, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 33, "global_step": 230, "opt_step": 57, "combined": 0.35832763, "ce": 0.3339847, "kd": 0.40555772, "gap_energy": 0.0712208, "kd_alpha": 0.34011299, "proof_weight": 1.64795918, "ce_num": 980.17529297, "ce_den": 2934.79101562, "kd_num": 1190.22717285, "kd_den": 2934.79101562, "gap_num": 208.73866272, "gap_den": 2930.86669922, "kd_tokens": 1484.0, "teacher_topk_mass_num": 1332.2479248, "student_topk_mass_num": 1324.51086426, "mass_den": 1484.0, "lr": 2.208303981704803e-06, "difficulty": 8.34626579284668, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 43, "global_step": 240, "opt_step": 60, "combined": 0.35486567, "ce": 0.33603364, "kd": 0.38983947, "gap_energy": 0.05347834, "kd_alpha": 0.35, "proof_weight": 1.63265306, "ce_num": 1235.95581055, "ce_den": 3678.07177734, "kd_num": 1433.85754395, "kd_den": 3678.07177734, "gap_num": 196.49060059, "gap_den": 3674.20898438, "kd_tokens": 1778.0, "teacher_topk_mass_num": 1631.93139648, "student_topk_mass_num": 1639.41564941, "mass_den": 1778.0, "lr": 2.124356012509813e-06, "difficulty": 8.55252456665039, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 53, "global_step": 250, "opt_step": 62, "combined": 0.43773556, "ce": 0.46106946, "kd": 0.39440125, "gap_energy": 0.04402284, "kd_alpha": 0.35, "proof_weight": 1.61734694, "ce_num": 2207.42407227, "ce_den": 4787.6171875, "kd_num": 1888.2421875, "kd_den": 4787.6171875, "gap_num": 210.58125305, "gap_den": 4783.45410156, "kd_tokens": 2192.0, "teacher_topk_mass_num": 1961.63867188, "student_topk_mass_num": 1959.3458252, "mass_den": 2192.0, "lr": 2.067172942406179e-06, "difficulty": 8.677350997924805, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 63, "global_step": 260, "opt_step": 65, "combined": 0.41169688, "ce": 0.39724419, "kd": 0.4385376, "gap_energy": 0.05188184, "kd_alpha": 0.35, "proof_weight": 1.60204082, "ce_num": 2263.84179688, "ce_den": 5698.8671875, "kd_num": 2499.16748047, "kd_den": 5698.8671875, "gap_num": 295.4510498, "gap_den": 5694.69091797, "kd_tokens": 2943.0, "teacher_topk_mass_num": 2676.89599609, "student_topk_mass_num": 2602.33740234, "mass_den": 2943.0, "lr": 1.979859548286157e-06, "difficulty": 9.147865295410156, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 73, "global_step": 270, "opt_step": 67, "combined": 0.31871417, "ce": 0.29050359, "kd": 0.37110531, "gap_energy": 0.03902761, "kd_alpha": 0.35, "proof_weight": 1.58673469, "ce_num": 1429.89501953, "ce_den": 4922.125, "kd_num": 1826.62670898, "kd_den": 4922.125, "gap_num": 191.9515686, "gap_den": 4918.35351562, "kd_tokens": 2449.0, "teacher_topk_mass_num": 2294.19091797, "student_topk_mass_num": 2247.55615234, "mass_den": 2449.0, "lr": 1.9207957113074184e-06, "difficulty": 9.025297164916992, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 83, "global_step": 280, "opt_step": 70, "combined": 0.31959426, "ce": 0.30900902, "kd": 0.33925256, "gap_energy": 0.03257714, "kd_alpha": 0.35, "proof_weight": 1.57142857, "ce_num": 1426.18310547, "ce_den": 4615.34472656, "kd_num": 1565.76757812, "kd_den": 4615.34472656, "gap_num": 150.23193359, "gap_den": 4611.57519531, "kd_tokens": 2419.0, "teacher_topk_mass_num": 2247.19628906, "student_topk_mass_num": 2205.96362305, "mass_den": 2419.0, "lr": 1.831214908853835e-06, "difficulty": 8.914824485778809, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 93, "global_step": 290, "opt_step": 72, "combined": 0.40945679, "ce": 0.40299642, "kd": 0.4214547, "gap_energy": 0.05730927, "kd_alpha": 0.35, "proof_weight": 1.55612245, "ce_num": 2152.30273438, "ce_den": 5340.74902344, "kd_num": 2250.88378906, "kd_den": 5340.74902344, "gap_num": 305.85800171, "gap_den": 5336.97265625, "kd_tokens": 2809.0, "teacher_topk_mass_num": 2531.46459961, "student_topk_mass_num": 2505.27807617, "mass_den": 2809.0, "lr": 1.7710130670196354e-06, "difficulty": 9.15400505065918, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 103, "global_step": 300, "opt_step": 75, "combined": 0.39351124, "ce": 0.38491118, "kd": 0.40948281, "gap_energy": 0.05109351, "kd_alpha": 0.35, "proof_weight": 1.54081633, "ce_num": 1915.125, "ce_den": 4975.49853516, "kd_num": 2037.38110352, "kd_den": 4975.49853516, "gap_num": 254.01828003, "gap_den": 4971.63525391, "kd_tokens": 2583.0, "teacher_topk_mass_num": 2381.25341797, "student_topk_mass_num": 2361.17797852, "mass_den": 2583.0, "lr": 1.6802913867993369e-06, "difficulty": 9.028308868408203, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 113, "global_step": 310, "opt_step": 77, "combined": 0.36131692, "ce": 0.30587906, "kd": 0.46427298, "gap_energy": 0.0437492, "kd_alpha": 0.35, "proof_weight": 1.5255102, "ce_num": 2284.91210938, "ce_den": 7469.98535156, "kd_num": 3468.11230469, "kd_den": 7469.98535156, "gap_num": 326.60665894, "gap_den": 7465.43115234, "kd_tokens": 3472.0, "teacher_topk_mass_num": 3023.33691406, "student_topk_mass_num": 3130.13110352, "mass_den": 3472.0, "lr": 1.6197086132006635e-06, "difficulty": 9.511463165283203, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 123, "global_step": 320, "opt_step": 80, "combined": 0.31258738, "ce": 0.30897042, "kd": 0.31930465, "gap_energy": 0.03223233, "kd_alpha": 0.35, "proof_weight": 1.51020408, "ce_num": 2031.16308594, "ce_den": 6573.97265625, "kd_num": 2099.10009766, "kd_den": 6573.97265625, "gap_num": 211.77005005, "gap_den": 6570.11425781, "kd_tokens": 3547.0, "teacher_topk_mass_num": 3221.05810547, "student_topk_mass_num": 3193.78173828, "mass_den": 3547.0, "lr": 1.5289869329803647e-06, "difficulty": 9.425006866455078, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 133, "global_step": 330, "opt_step": 82, "combined": 0.37523079, "ce": 0.36893314, "kd": 0.38692647, "gap_energy": 0.04796501, "kd_alpha": 0.35, "proof_weight": 1.49489796, "ce_num": 2777.39501953, "ce_den": 7528.1796875, "kd_num": 2912.85205078, "kd_den": 7528.1796875, "gap_num": 360.87277222, "gap_den": 7523.66699219, "kd_tokens": 3835.0, "teacher_topk_mass_num": 3558.80957031, "student_topk_mass_num": 3521.45751953, "mass_den": 3835.0, "lr": 1.4687850911461655e-06, "difficulty": 9.515018463134766, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 143, "global_step": 340, "opt_step": 85, "combined": 0.3591302, "ce": 0.36028352, "kd": 0.3569884, "gap_energy": 0.03716599, "kd_alpha": 0.35, "proof_weight": 1.47959184, "ce_num": 2544.3293457, "ce_den": 7062.01953125, "kd_num": 2521.05908203, "kd_den": 7062.01953125, "gap_num": 262.29891968, "gap_den": 7057.49853516, "kd_tokens": 3668.0, "teacher_topk_mass_num": 3338.10302734, "student_topk_mass_num": 3297.52075195, "mass_den": 3668.0, "lr": 1.3792042886925817e-06, "difficulty": 9.386213302612305, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 153, "global_step": 350, "opt_step": 87, "combined": 0.31356943, "ce": 0.28112417, "kd": 0.37382498, "gap_energy": 0.03802348, "kd_alpha": 0.35, "proof_weight": 1.46428571, "ce_num": 2047.1159668, "ce_den": 7281.89257812, "kd_num": 2722.15332031, "kd_den": 7281.89257812, "gap_num": 276.73950195, "gap_den": 7278.12158203, "kd_tokens": 3810.0, "teacher_topk_mass_num": 3609.49633789, "student_topk_mass_num": 3521.68920898, "mass_den": 3810.0, "lr": 1.3201404517138427e-06, "difficulty": 9.56966781616211, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 163, "global_step": 360, "opt_step": 90, "combined": 0.3101933, "ce": 0.32049137, "kd": 0.29106832, "gap_energy": 0.02370189, "kd_alpha": 0.35, "proof_weight": 1.44897959, "ce_num": 2269.39501953, "ce_den": 7080.98632812, "kd_num": 2061.05078125, "kd_den": 7080.98632812, "gap_num": 167.73965454, "gap_den": 7077.05957031, "kd_tokens": 3767.0, "teacher_topk_mass_num": 3466.20703125, "student_topk_mass_num": 3470.0456543, "mass_den": 3767.0, "lr": 1.2328270575938212e-06, "difficulty": 9.567719459533691, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 173, "global_step": 370, "opt_step": 92, "combined": 0.25587186, "ce": 0.24827541, "kd": 0.2699796, "gap_energy": 0.03437245, "kd_alpha": 0.35, "proof_weight": 1.43367347, "ce_num": 1768.06774902, "ce_den": 7121.39697266, "kd_num": 1922.63183594, "kd_den": 7121.39697266, "gap_num": 244.64147949, "gap_den": 7117.37207031, "kd_tokens": 3774.0, "teacher_topk_mass_num": 3600.57617188, "student_topk_mass_num": 3582.96337891, "mass_den": 3774.0, "lr": 1.1756439874901873e-06, "difficulty": 9.657493591308594, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 183, "global_step": 380, "opt_step": 95, "combined": 0.2495057, "ce": 0.23697412, "kd": 0.27277863, "gap_energy": 0.03516343, "kd_alpha": 0.35, "proof_weight": 1.41836735, "ce_num": 1761.13122559, "ce_den": 7431.74511719, "kd_num": 2027.22119141, "kd_den": 7431.74511719, "gap_num": 261.18389893, "gap_den": 7427.71386719, "kd_tokens": 3983.0, "teacher_topk_mass_num": 3788.17041016, "student_topk_mass_num": 3781.07714844, "mass_den": 3983.0, "lr": 1.0916960182951974e-06, "difficulty": 9.683958053588867, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 2, "batch_idx": 193, "global_step": 390, "opt_step": 97, "combined": 0.30295968, "ce": 0.31477621, "kd": 0.28101465, "gap_energy": 0.02040295, "kd_alpha": 0.35, "proof_weight": 1.40306122, "ce_num": 2344.95214844, "ce_den": 7449.58496094, "kd_num": 2093.44262695, "kd_den": 7449.58496094, "gap_num": 151.9080658, "gap_den": 7445.39550781, "kd_tokens": 3860.0, "teacher_topk_mass_num": 3460.72460938, "student_topk_mass_num": 3475.14306641, "mass_den": 3860.0, "lr": 1.0371128253516119e-06, "difficulty": 9.59878158569336, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 7, "global_step": 400, "opt_step": 100, "combined": 0.92327392, "ce": 0.99677551, "kd": 0.78677112, "gap_energy": 0.13165115, "kd_alpha": 0.35, "proof_weight": 1.3877551, "ce_num": 421.11520386, "ce_den": 422.47747803, "kd_num": 332.39306641, "kd_den": 422.47747803, "gap_num": 55.2678833, "gap_den": 419.80554199, "kd_tokens": 324.0, "teacher_topk_mass_num": 214.39503479, "student_topk_mass_num": 201.57331848, "mass_den": 324.0, "lr": 9.57585975502027e-07, "difficulty": 5.915072441101074, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 17, "global_step": 410, "opt_step": 102, "combined": 0.50501037, "ce": 0.44381145, "kd": 0.61866558, "gap_energy": 0.0960589, "kd_alpha": 0.35, "proof_weight": 1.37244898, "ce_num": 392.18963623, "ce_den": 883.6852417, "kd_num": 546.70562744, "kd_den": 883.6852417, "gap_num": 84.62545776, "gap_den": 880.97467041, "kd_tokens": 674.0, "teacher_topk_mass_num": 562.82067871, "student_topk_mass_num": 528.24304199, "mass_den": 674.0, "lr": 9.062890750396618e-07, "difficulty": 7.160126686096191, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 27, "global_step": 420, "opt_step": 105, "combined": 0.56388533, "ce": 0.51320404, "kd": 0.65800786, "gap_energy": 0.12683682, "kd_alpha": 0.35, "proof_weight": 1.35714286, "ce_num": 1160.43701172, "ce_den": 2261.16113281, "kd_num": 1487.86181641, "kd_den": 2261.16113281, "gap_num": 286.32626343, "gap_den": 2257.43798828, "kd_tokens": 1385.0, "teacher_topk_mass_num": 1162.08874512, "student_topk_mass_num": 1125.13244629, "mass_den": 1385.0, "lr": 8.321834408148934e-07, "difficulty": 8.027276992797852, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 37, "global_step": 430, "opt_step": 107, "combined": 0.34577429, "ce": 0.33484906, "kd": 0.36606401, "gap_energy": 0.08746573, "kd_alpha": 0.35, "proof_weight": 1.34183673, "ce_num": 1006.72387695, "ce_den": 3006.50048828, "kd_num": 1100.57165527, "kd_den": 3006.50048828, "gap_num": 262.65948486, "gap_den": 3002.99902344, "kd_tokens": 1761.0, "teacher_topk_mass_num": 1567.68347168, "student_topk_mass_num": 1570.17810059, "mass_den": 1761.0, "lr": 7.848179211202175e-07, "difficulty": 8.473159790039062, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 47, "global_step": 440, "opt_step": 110, "combined": 0.39065716, "ce": 0.39188409, "kd": 0.38837859, "gap_energy": 0.03742781, "kd_alpha": 0.35, "proof_weight": 1.32653061, "ce_num": 1219.81176758, "ce_den": 3112.68505859, "kd_num": 1208.90026855, "kd_den": 3112.68505859, "gap_num": 116.3790741, "gap_den": 3109.42822266, "kd_tokens": 1870.0, "teacher_topk_mass_num": 1658.35217285, "student_topk_mass_num": 1659.3380127, "mass_den": 1870.0, "lr": 7.170654238677327e-07, "difficulty": 8.60601806640625, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 57, "global_step": 450, "opt_step": 112, "combined": 0.16394296, "ce": 0.14618537, "kd": 0.19692138, "gap_energy": 0.02102127, "kd_alpha": 0.35, "proof_weight": 1.31122449, "ce_num": 476.72827148, "ce_den": 3261.12158203, "kd_num": 642.18457031, "kd_den": 3261.12158203, "gap_num": 68.48384857, "gap_den": 3257.83666992, "kd_tokens": 1909.0, "teacher_topk_mass_num": 1841.36376953, "student_topk_mass_num": 1828.80688477, "mass_den": 1909.0, "lr": 6.742269338330216e-07, "difficulty": 8.738836288452148, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 67, "global_step": 460, "opt_step": 115, "combined": 0.34792715, "ce": 0.37267017, "kd": 0.30197588, "gap_energy": 0.01995143, "kd_alpha": 0.35, "proof_weight": 1.29591837, "ce_num": 1611.00402832, "ce_den": 4322.86816406, "kd_num": 1305.40185547, "kd_den": 4322.86816406, "gap_num": 86.17908478, "gap_den": 4319.44433594, "kd_tokens": 2567.0, "teacher_topk_mass_num": 2339.13134766, "student_topk_mass_num": 2327.57617188, "mass_den": 2567.0, "lr": 6.13679600516685e-07, "difficulty": 9.008158683776855, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 77, "global_step": 470, "opt_step": 117, "combined": 0.21805955, "ce": 0.20472163, "kd": 0.24283001, "gap_energy": 0.02222797, "kd_alpha": 0.35, "proof_weight": 1.28061224, "ce_num": 1057.07836914, "ce_den": 5163.49121094, "kd_num": 1253.85058594, "kd_den": 5163.49121094, "gap_num": 114.66514587, "gap_den": 5158.59716797, "kd_tokens": 3050.0, "teacher_topk_mass_num": 2907.29199219, "student_topk_mass_num": 2896.83374023, "mass_den": 3050.0, "lr": 5.759068590095958e-07, "difficulty": 8.891244888305664, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 87, "global_step": 480, "opt_step": 120, "combined": 0.4542684, "ce": 0.48165402, "kd": 0.40340948, "gap_energy": 0.06115909, "kd_alpha": 0.35, "proof_weight": 1.26530612, "ce_num": 2448.62890625, "ce_den": 5083.79199219, "kd_num": 2050.84985352, "kd_den": 5083.79199219, "gap_num": 310.69360352, "gap_den": 5080.08886719, "kd_tokens": 3270.0, "teacher_topk_mass_num": 2836.52124023, "student_topk_mass_num": 2793.31738281, "mass_den": 3270.0, "lr": 5.233261074762414e-07, "difficulty": 9.195501327514648, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 97, "global_step": 490, "opt_step": 122, "combined": 0.35276407, "ce": 0.34306455, "kd": 0.37077755, "gap_energy": 0.03944525, "kd_alpha": 0.35, "proof_weight": 1.25, "ce_num": 1354.73242188, "ce_den": 3948.9140625, "kd_num": 1464.16870117, "kd_den": 3948.9140625, "gap_num": 155.63720703, "gap_den": 3945.65087891, "kd_tokens": 2435.0, "teacher_topk_mass_num": 2298.7253418, "student_topk_mass_num": 2252.71826172, "mass_den": 2435.0, "lr": 4.910941286374807e-07, "difficulty": 8.969837188720703, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 107, "global_step": 500, "opt_step": 125, "combined": 0.30729848, "ce": 0.30684832, "kd": 0.30813453, "gap_energy": 0.02581502, "kd_alpha": 0.35, "proof_weight": 1.23469388, "ce_num": 1901.15356445, "ce_den": 6195.74365234, "kd_num": 1909.12255859, "kd_den": 6195.74365234, "gap_num": 159.84867859, "gap_den": 6192.08105469, "kd_tokens": 3523.0, "teacher_topk_mass_num": 3198.31567383, "student_topk_mass_num": 3213.59204102, "mass_den": 3523.0, "lr": 4.471411923457035e-07, "difficulty": 9.413188934326172, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 117, "global_step": 510, "opt_step": 127, "combined": 0.31296164, "ce": 0.30628365, "kd": 0.32536358, "gap_energy": 0.04044421, "kd_alpha": 0.35, "proof_weight": 1.21938776, "ce_num": 1758.57641602, "ce_den": 5741.65917969, "kd_num": 1868.12670898, "kd_den": 5741.65917969, "gap_num": 232.08267212, "gap_den": 5738.34179688, "kd_tokens": 3478.0, "teacher_topk_mass_num": 3133.03076172, "student_topk_mass_num": 3117.89526367, "mass_den": 3478.0, "lr": 4.208553120105986e-07, "difficulty": 9.388130187988281, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 127, "global_step": 520, "opt_step": 130, "combined": 0.54961079, "ce": 0.57953328, "kd": 0.49404049, "gap_energy": 0.02451204, "kd_alpha": 0.35, "proof_weight": 1.20408163, "ce_num": 3456.52905273, "ce_den": 5964.33203125, "kd_num": 2946.62158203, "kd_den": 5964.33203125, "gap_num": 146.1060791, "gap_den": 5960.58496094, "kd_tokens": 3780.0, "teacher_topk_mass_num": 2950.45654297, "student_topk_mass_num": 3000.10839844, "mass_den": 3780.0, "lr": 3.860829246363548e-07, "difficulty": 9.353657722473145, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 137, "global_step": 530, "opt_step": 132, "combined": 0.25616327, "ce": 0.2474604, "kd": 0.27232572, "gap_energy": 0.03138196, "kd_alpha": 0.35, "proof_weight": 1.18877551, "ce_num": 1285.36254883, "ce_den": 5194.21533203, "kd_num": 1414.51843262, "kd_den": 5194.21533203, "gap_num": 162.90638733, "gap_den": 5191.08300781, "kd_tokens": 3460.0, "teacher_topk_mass_num": 3235.41064453, "student_topk_mass_num": 3186.61279297, "mass_den": 3460.0, "lr": 3.6607370300154275e-07, "difficulty": 9.462833404541016, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 147, "global_step": 540, "opt_step": 135, "combined": 0.2797212, "ce": 0.27460113, "kd": 0.28922996, "gap_energy": 0.02170261, "kd_alpha": 0.35, "proof_weight": 1.17346939, "ce_num": 1391.22668457, "ce_den": 5066.35449219, "kd_num": 1465.34155273, "kd_den": 5066.35449219, "gap_num": 109.88298035, "gap_den": 5063.12304688, "kd_tokens": 3281.0, "teacher_topk_mass_num": 3077.32519531, "student_topk_mass_num": 3067.69018555, "mass_den": 3281.0, "lr": 3.409191474914727e-07, "difficulty": 9.389362335205078, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 157, "global_step": 550, "opt_step": 137, "combined": 0.191145, "ce": 0.17197637, "kd": 0.22674394, "gap_energy": 0.02278897, "kd_alpha": 0.35, "proof_weight": 1.15816327, "ce_num": 1032.36450195, "ce_den": 6002.94384766, "kd_num": 1361.13110352, "kd_den": 6002.94384766, "gap_num": 136.73205566, "gap_den": 5999.92285156, "kd_tokens": 3806.0, "teacher_topk_mass_num": 3634.18554688, "student_topk_mass_num": 3607.42773438, "mass_den": 3806.0, "lr": 3.274382121286321e-07, "difficulty": 9.663820266723633, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 167, "global_step": 560, "opt_step": 140, "combined": 0.27715239, "ce": 0.26765925, "kd": 0.29478255, "gap_energy": 0.02763841, "kd_alpha": 0.35, "proof_weight": 1.14285714, "ce_num": 1546.05029297, "ce_den": 5776.18847656, "kd_num": 1702.71960449, "kd_den": 5776.18847656, "gap_num": 159.55589294, "gap_den": 5772.97558594, "kd_tokens": 3862.0, "teacher_topk_mass_num": 3626.77905273, "student_topk_mass_num": 3650.0546875, "mass_den": 3862.0, "lr": 3.1221782161328813e-07, "difficulty": 9.599081039428711, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 177, "global_step": 570, "opt_step": 142, "combined": 0.36908805, "ce": 0.41059637, "kd": 0.29200116, "gap_energy": 0.01557151, "kd_alpha": 0.35, "proof_weight": 1.12755102, "ce_num": 2398.20751953, "ce_den": 5840.79101562, "kd_num": 1705.51782227, "kd_den": 5840.79101562, "gap_num": 90.89710236, "gap_den": 5837.39697266, "kd_tokens": 3945.0, "teacher_topk_mass_num": 3590.37695312, "student_topk_mass_num": 3634.23754883, "mass_den": 3945.0, "lr": 3.054347031064272e-07, "difficulty": 9.548123359680176, "memory": "alloc=26.8GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}
{"epoch": 3, "batch_idx": 187, "global_step": 580, "opt_step": 145, "combined": 0.24322987, "ce": 0.23830642, "kd": 0.25237346, "gap_energy": 0.01909368, "kd_alpha": 0.35, "proof_weight": 1.1122449, "ce_num": 1339.30322266, "ce_den": 5620.08886719, "kd_num": 1418.36120605, "kd_den": 5620.08886719, "gap_num": 107.24766541, "gap_den": 5616.91796875, "kd_tokens": 3840.0, "teacher_topk_mass_num": 3661.22802734, "student_topk_mass_num": 3657.67651367, "mass_den": 3840.0, "lr": 3.0033988282761614e-07, "difficulty": 9.662643432617188, "memory": "alloc=20.4GiB reserved=39.8GiB peak=37.0GiB total=79.3GiB"}