Files
explore-tis-minp-40-8B/training_logs/20260612_111844_metrics_explore-tis-minp_736732.csv
ModelHub XC cab2418769 初始化项目,由ModelHub XC社区提供模型
Model: laion/explore-tis-minp-40-8B
Source: Original Platform
2026-08-09 12:53:53 +08:00

22 KiB

1async/discard_rateasync/discarded_countasync/effective_batch_groupsasync/effective_batch_samplesasync/staleness_maxasync/staleness_meanasync/staleness_minasync/staleness_ratiogenerate/avg_num_tokensgenerate/avg_tokens_non_zero_rewardsgenerate/avg_tokens_zero_rewardsgenerate/max_num_tokensgenerate/min_num_tokensgenerate/std_num_tokensgenerate/tis/aligned_tokensgenerate/tis/alignment_fail_countgenerate/tis/exact_match_fractiongenerate/tis/lcs_fallback_fractiongenerate/tis/lcs_fallback_messagesgenerate/tis/unaligned_fractionloss/avg_final_rewardsloss/avg_raw_advantagesloss/avg_raw_advantages_abspolicy/final_losspolicy/log_ratio_abs_maxpolicy/log_ratio_abs_meanpolicy/log_ratio_abs_p99policy/log_ratio_abs_pos00policy/log_ratio_abs_pos10policy/log_ratio_abs_pos20policy/log_ratio_abs_pos30policy/log_ratio_abs_pos40policy/log_ratio_abs_pos50policy/log_ratio_abs_pos60policy/log_ratio_abs_pos70policy/log_ratio_abs_pos80policy/log_ratio_abs_pos90policy/n_tokens_dp_gt_10pctpolicy/n_tokens_dp_gt_1pctpolicy/n_tokens_dp_gt_50pctpolicy/policy_entropypolicy/policy_losspolicy/policy_lrpolicy/policy_update_stepspolicy/ppo_clip_ratiopolicy/raw_grad_normpolicy/rollout_train_prob_diff_meanpolicy/rollout_train_prob_diff_stdpolicy/tis/imp_ratio_capped_fractionpolicy/tis/imp_ratio_meanpolicy/tis/log_ratio_abs_meanreward/avg_pass_at_8reward/avg_raw_rewardsystem/process_rss_gbsystem/process_vms_gbsystem/ram_available_gbsystem/ram_percentsystem/ram_total_gbsystem/ram_used_gbtiming/compute_advantages_and_returnstiming/convert_to_training_inputtiming/fwd_logprobs_values_rewardtiming/policy_traintiming/run_trainingtiming/save_hf_modeltiming/steptiming/sync_weightstiming/train_critic_and_policytiming/wait_for_generation_buffertis/batch_skipped_no_logprobstis/skipped_fractiontrainer/epochtrainer/global_stepbatch_errors/total_batchesbatch_errors/total_instancesbatch_errors/total_successfulbatch_errors/total_failedbatch_errors/total_maskedbatch_errors/avg_ContextLengthExceededErrorbatch_errors/total_ContextLengthExceededErrorbatch_errors/avg_VerifierTimeoutErrorbatch_errors/total_VerifierTimeoutErrorbatch_errors/avg_AgentTimeoutErrorbatch_errors/total_AgentTimeoutErrortiming/cleanup_old_checkpointstiming/save_checkpointsbatch_errors/avg_InvalidChatHistorybatch_errors/total_InvalidChatHistorybatch_errors/avg_DaytonaErrorbatch_errors/total_DaytonaError
20.006451250.687500.257482.76176126.1668747.21132974515486.98552643273.0117.00.93560.00.00.06440.48240.00060.1012-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2542-0.00.01.00.00.019510074.020225333248.00.00.96190.0250.53570.482413.165350.8276615.711928.2857.9676242.25570.06546.346452.9064281.2738334.46687.03533502.055129.9246281.49423131.31690.00.00156451247516310.46875300.093756.00.0156251.0
30.006451210.859400.85948037.4596026.07239743.873168816875.86732839150.0106.00.94430.00.00.05570.459-0.00250.08480.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.24760.00.01.00.00.06890716672.011557151965184.00.00.92040.0220.54690.45914.655851.5332609.451929.0857.9676248.51570.06566.820860.2403356.8522417.35481342.481729.7467357.0482888.54210.00.00165342435132721.320754716981132700.0188679245283018861.00.0188679245283018861.027.272234.76420.2264150943396226512.0
40.006451221.781211.07939.38286742.76568338.25523180117604.69662955546.0129.00.88720.00.00.11280.25-0.00020.10960.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.23860.00.01.00.00.0432779648.0557215318016.00.00.82470.02840.39060.2514.857151.735606.172429.3857.9676251.79520.06526.335855.1267364.1323419.5598520.467426.5983364.366567.97040.00.00176451245525570.765625490.1258.0
50.006451232.359400.96888275.96488661.07398022.96443173217287.46543039004.0112.00.90830.00.00.09170.39650.00510.1418-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2387-0.00.01.00.00.0159518656.0202329915392.00.00.88010.02360.54690.396515.093251.9595604.103529.6857.9676253.86410.07186.798355.3895366.0651421.71981408.031431.3425366.2575948.16690.00.00186451247616360.53125340.007412.03190.031252.0
60.006451242.265600.93758417.94147542.29489260.03833175116792.80443043225.0125.00.93350.00.00.06650.4902-0.00840.13040.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.26020.00.01.00.00.019273420.017220765696.00.00.92060.02510.64060.490215.321252.2415601.641429.9857.9676256.32620.06736.530754.582337.7653392.61551149.45629.1714337.9658721.13440.00.00196350444023610.6666666666666666420.0476190476190476163.00.079365079365079365.00.222222222222222214.0
70.006451252.37500.96887685.55866459.82548402.78333173617151.7242873751.0123.00.89240.00.00.10760.3691-0.00390.11470.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2470.00.01.00.00.039120392.039058935808.00.00.86480.02740.50.369115.494952.4728599.902530.1857.9676258.06510.06596.264554.7339348.5479403.57976.15281059.24227.7102348.7795621.68360.00.00206350445918430.5714285714285714360.0317460317460317442.00.095238095238095236.00.007511.45170.079365079365079365.0
80.006451252.37500.90628150.86726569.38259030.5383173717031.40162999351.098.00.92620.00.00.07380.3574-0.00970.11890.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.26030.00.01.00.00.017383612.027595026432.00.00.8980.01940.50.357415.714652.6397598.838430.2857.9676259.12920.06646.315553.524335.2281388.99961096.30228.6169335.4087672.36550.00.00216148843522530.8032786885245902490.065573770491803284.0
90.006451241.968800.90627439.48446406.67348096.12783175916791.12222675755.092.00.9460.00.00.0540.3887-0.00260.13330.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.24010.00.01.00.00.01093299.375670519488.00.00.88940.02080.50.388717.25153.1439595.518930.6857.9676262.44870.06587.115756.5107339.5945396.36521386.593330.3841339.7883952.72350.00.00226451245723530.671875430.031252.00.0156251.00.00837.77970.1406259.0
100.006451242.203100.92197386.9186300.66258345.37873164016445.45432650438.0117.00.94340.00.00.05660.4688-0.00170.1370.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.24180.00.01.00.00.011963810.012564688896.00.00.88550.02140.60940.468817.664253.5193594.392630.7857.9676263.5750.06596.946552.1156314.3296366.72421009.276527.1353314.5424608.46560.00.00236350444922510.5396825396825397340.063492063492063494.00.0158730158730158721.00.2539682539682539516.0
110.006451242.015600.85947514.78916947.75697935.24153170416678.21382678046.0107.00.92420.00.00.07580.42580.01190.1435-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2377-0.00.01.00.00.0114651521024.0187598668562432.00.00.88360.02530.57810.425817.535253.3669593.678730.8857.9676264.28890.06646.37853.164306.7048360.14211382.795134.4674306.9113981.80270.00.00245745641018380.7368421052631579420.140350877192982458.00.070175438596491224.00.008312.30960.105263157894736846.0
120.006451241.890600.93756617.39265544.16327556.95973175716481.08182365783.0114.00.91560.00.00.08440.46680.00230.0655-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2137-0.00.01.00.00.047685716.068093718528.00.00.8760.020.53120.466817.791453.6491591.809831.0857.9676266.15780.06456.194151.2192296.2787347.7976.124801.121722.6086296.5129424.51720.00.00256249643421410.5645161290322581350.338709677419354821.00.032258064516129032.00.032258064516129032.0
130.006451262.500.98446969.00986421.3787475.46623070016352.31322520635.0124.00.92390.00.00.07610.48050.00670.0896-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2231-0.00.01.00.00.02873880.54222229248.00.00.87470.0220.57810.480517.745153.6056590.623431.2857.9676267.34420.06386.045447.7615302.4312350.4695773.891822.3592302.6432395.0130.00.00266249644617500.8064516129032258500.0078.61340.032258064516129032.0
140.006451252.453100.98447021.61526067.08248061.86123173716352.07512440050.0100.00.95480.00.00.04520.52150.00180.0982-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2307-0.00.01.00.00.08815552.011449403392.00.00.89330.01740.64060.521518.102953.9844589.674531.3857.9676268.29310.0666.222148.4281312.3997361.1152907.815429.6382312.6205510.83520.00.00276148845318330.5245901639344263320.032786885245901642.00.016393442622950821.00.065573770491803284.0
150.006451252.640600.96888230.67388129.84518310.34973171116991.44813035363.0120.00.91950.00.00.08050.4414-0.00330.17370.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.25470.00.01.00.00.0180434320.0307367215104.00.00.91460.0250.60940.441418.325254.3188587.734931.5857.9676270.23270.06697.103564.8035361.7218426.7919985.029526.1894361.9211524.940.00.00286249645714350.6129032258064516380.064516129032258064.00.0161290322580645161.00.006810.7011
160.006451252.500.90627213.07815913.25998248.37193043516338.43082580776.0107.00.94130.00.00.05870.44340.00330.1204-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2448-0.00.01.00.00.09582464.014558073856.00.00.90690.02480.57810.443418.251654.2496589.289931.3857.9676268.67770.06356.120848.3907308.2879356.9432993.477226.9393308.4887603.46870.00.00296350445219500.6507936507936508410.0317460317460317442.00.1904761904761904712.0
170.006451252.578100.96888266.92977126.32859041.04263172217055.3142992963.0128.00.9070.00.00.0930.4043-0.0020.11060.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.24810.00.01.00.00.000115394298.023996018688.00.00.8820.0270.51560.404318.327354.4241588.107831.5857.9676269.85980.06657.084357.0285349.7239407.04395.83691127.204433.3088349.9485679.76250.00.00306451245624480.59375380.1258.00.031252.00.00859.01950.1258.0
180.006451262.500.93758044.50596225.66839200.89143158917349.79262838452.0111.00.94760.00.00.05240.3887-0.00330.08030.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.23780.00.01.00.00.03174915.753706918656.00.00.88350.02090.48440.388718.459454.4977586.766731.6857.9676271.20090.0677.054259.8986360.1246420.28681049.731528.3635360.3208594.02230.00.00316350443825620.8253968253968254520.063492063492063494.00.0476190476190476163.00.11111111111111117.0
190.006451252.484400.96887292.73056960.70777496.97163178816536.46492648742.0118.00.90960.00.00.09040.38090.00070.13590.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.23710.00.01.00.00.027203108.026022187008.00.00010.86150.02610.50.380918.631954.7449585.625231.7857.9676272.34240.6567.036350.7801323.9554375.76731175.335428.3488324.3308764.17850.00.00326048042519510.8666666666666667520.066666666666666674.00.0166666666666666661.00.00848.45470.083333333333333335.0
200.006451252.203100.93757018.68556590.84457390.31393175816465.5762569695.0102.00.9120.00.00.0880.4648-0.0090.12320.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.23550.00.01.00.00.012588754.012803294208.00.00.86960.02470.59380.464818.750254.8053583.854431.9857.9676274.11320.06576.247551.6752311.2723363.25731139.404624.9403311.5161744.95420.00.00336350444524590.8571428571428571540.0158730158730158721.00.0476190476190476163.00.0158730158730158721.0
210.006451242.218800.93757944.64657031.37068398.6143176317242.362917562.0142.00.92030.00.00.07970.332-0.00550.08970.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.25290.00.01.00.00.012217008.014082661376.00.00.87460.02740.45310.33218.682754.7606583.578132.0857.9676274.38950.07397.079559.8281341.7272401.8631286.672826.0248341.9606851.70040.00.00346249643321610.8709677419354839540.032258064516129032.00.032258064516129032.00.00829.44420.112903225806451617.0
220.006451242.390600.96887951.1277135.5978478.20583177917010.16252935582.0120.00.92940.00.00.07060.3926-0.00620.10020.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.25380.00.01.00.00.032398294.039496724480.00.00.86340.02640.53120.392618.745854.8953582.151232.1857.9676275.81630.06726.467652.9555342.0534395.28736.89511082.554726.2807342.2634654.51410.00.00356350445719380.5714285714285714360.142857142857142859.00.0158730158730158721.00.0158730158730158721.0
230.006451252.296900.92198858.2097790.5049877.01153176517479.49743181656.0101.00.94910.00.00.05090.48830.00930.0996-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2622-0.00.01.00.00.038615248.038171488256.00.00.90160.02050.60940.488318.974155.2587638.413325.6857.9676219.55430.06816.570261.7736377.7247439.75871257.025727.3802377.9165783.31120.00.00366249645418420.6290322580645161390.0161290322580645161.00.00898.70460.032258064516129032.0
240.006451242.234400.96888080.98056786.61849120.11623175716787.56472910077.0121.00.92580.00.00.07420.44530.00690.1142-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2538-0.00.01.00.00.04683183.07682820608.00.00.9090.02450.57810.445319.485755.752633.447726.2857.9676224.51990.06726.427456.9733340.1355397.3946917.778526.8049340.3536487.1460.00.00376350446316400.5079365079365079320.0158730158730158721.00.0317460317460317442.00.095238095238095236.0
250.006451252.12500.96887409.48835886.21729069.54293163016564.962655325.0102.00.91610.00.00.08390.5215-0.00250.17410.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.23820.00.01.00.00.03716242.03137961728.00.00.91240.02160.73440.521519.523655.8805629.250526.7857.9676228.71710.06596.231252.0521326.6665378.9787983.783822.9394326.8603575.62920.00.00386350446217420.5555555555555556350.0098.3140.11111111111111117.0
260.006451252.484400.95318120.85557171.62288882.91553179117231.14132876492.0106.00.93310.00.00.06690.4453-0.00280.08470.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.23640.00.01.00.00.01753330.252051548160.00.00.90390.01920.53120.445319.773756.1929637.375825.7857.9676220.59180.06717.000758.6425357.7414416.66781108.141931.2828357.9578653.18540.00.00396249643121550.8709677419354839540.1612903225806451610.00.032258064516129032.00.048387096774193553.0
270.006451252.265600.96887503.59186671.77828231.80953182417059.04042737800.0123.00.92370.00.00.07630.4668-0.0220.16240.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2380.00.01.00.00.092425392.0144606150656.00.00.85540.02580.57810.466820.015156.3186632.60126.3857.9676225.36660.06636.230956.3286339.6449396.25875.87951095.749625.5044339.8635667.75020.00.00406249644421500.6935483870967742430.032258064516129032.00.080645161290322585.00.00828.8630.048387096774193553.0
280.006451252.093800.93758071.56846913.94169238.27453171917359.75092946770.0109.00.91070.00.00.08930.5020.00480.1319-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2513-0.00.01.00.00.06029641.06124079104.00.00.88910.02160.64060.50220.025856.3937628.889326.7857.9676229.07820.06886.443959.0139360.4215419.69211160.784233.1277360.6089701.51580.00.00416350444122550.7619047619047619480.126984126984126988.00.063492063492063494.00.095238095238095236.0
290.006451252.312500.95317437.36916752.18397966.07613173916667.77752722043.098.00.93690.00.00.06310.4355-0.00220.08250.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.24910.00.01.00.00.000159223816.097439277056.00.00.8640.02080.51560.435520.195756.5093625.445727.1857.9676232.52190.06617.193849.8498317.7805367.92071078.472228.393318.0044674.95980.00.00426249646412250.3064516129032258190.112903225806451617.00.00838.580.09677419354838716.0
300.006451252.156200.92197688.14066817.62348450.24183159616504.82312791232.090.00.94020.00.00.05980.46680.00780.1114-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2533-0.00.01.00.00.0111623.37572708800.00.00.92390.02190.57810.466820.238856.6951622.350827.5857.9676235.61680.06696.254953.6138319.5558373.44691029.592223.8621319.7659626.0230.00.00436249643125560.8870967741935484550.145161290322580669.00.0161290322580645161.00.09677419354838716.0
310.006451252.390600.98447184.62116020.98977900.41963171616772.19732615563.0102.00.94990.00.00.05010.3809-0.00740.06640.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.24050.00.01.00.00.019368192.020019861504.00.00.85860.02030.45310.380920.342556.8174619.828427.8857.9676238.13920.06597.090451.7311335.1143387.1258951.192826.4368335.3279530.53480.00.00446148842819590.9344262295081968570.016393442622950821.00.00829.05020.016393442622950821.00.065573770491803284.0
320.006451252.515600.95318093.89068013.67438153.37073171817129.37832990112.0138.00.91550.00.00.08450.42580.00110.1129-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.259-0.00.01.00.00.01438056.52143342848.00.00.86550.03250.56250.425820.418256.8773616.348328.2857.9676241.61930.06716.921359.1004336.107395.48486.1381940.096426.733336.3164510.95230.00.00456451246817440.546875350.031252.00.0468753.00.0156251.0
330.006451252.421900.93757612.40236184.25559256.57143174416842.65452767039.0107.00.92530.00.00.07470.5352-0.01390.10840.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.24230.00.01.00.00.05549523.58176544256.00.00.91180.02240.67190.535220.779157.1281614.04728.4857.9676243.92060.06686.464555.1899344.4617399.93571020.383125.1067344.6787588.87040.00.00466249644720470.7258064516129032450.032258064516129032.00.0161290322580645161.00.0088.49040.0161290322580645161.0
340.006451252.687500.95318559.84777688.65969298.94223158517207.45413128900.0132.00.90910.00.00.09090.459-0.00.11750.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.25820.00.01.00.00.010542674.016010543104.00.00.89240.03020.59380.45921.114457.4342612.583528.6857.9676245.38410.06726.562858.5747358.2761417.14361061.911125.8273358.4837612.37240.00.00476148842523520.8524590163934426520.1803278688524590211.00.049180327868852463.00.049180327868852463.0
350.006451252.421900.98447159.81845588.43738819.55023179117521.55852605239.0104.00.93260.00.00.06740.5137-0.00380.08190.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2350.00.01.00.00.0273707936.0439460003840.00.00.85770.02140.56250.513721.151757.5652610.780228.8857.9676247.18740.06516.331852.7936334.7332387.8037887.226125.9772334.9446467.10730.00.00486249643321631.0161290322580645630.00848.4740.032258064516129032.0
360.006451252.593800.95318699.76567074.33539507.72813173917577.46843196996.0133.00.93040.00.00.06960.332-0.00130.12120.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.25190.00.01.00.00.016002687.014805905408.00.00.86260.02760.43750.33221.665658.0778608.462729.1857.9676249.50490.06856.598160.3184374.8155435.42961247.893723.6818375.0424782.17870.00.00496148844216460.6229508196721312380.032786885245901642.00.065573770491803284.00.114754098360655747.0
370.006451252.359400.95318679.3776815.98959788.12153171917833.25533205510.0117.00.93890.00.00.06110.373-0.01950.09030.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.25270.00.01.00.00.012993111.014640867328.00.00.89150.02520.46880.37321.662158.1664607.040429.2857.9676250.92720.06777.307765.0076380.9533446.22535.98361232.95827.6876381.1498751.73220.00.00506249643623540.8064516129032258500.09677419354838716.00.032258064516129032.00.0088.36580.032258064516129032.0
380.006451252.468800.93758470.80866011.43889996.24053163717435.94723096734.0140.00.90840.00.00.09160.38280.00420.0756-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.2552-0.00.01.00.00.02417077.03815741696.00.00.87280.02570.48440.382821.972558.5468605.281829.5857.9676252.68580.06826.736761.7314352.1131414.11211069.221931.0897352.3121617.2780.00.00513124820910391.1290322580645162350.129032258064516134.00.225806451612903227.0