Files
ablation-pymethods2test-seq…/training_logs/20260609_214045_metrics_job_653587.csv
ModelHub XC 9f22667c70 初始化项目,由ModelHub XC社区提供模型
Model: laion/ablation-pymethods2test-seqmean-arm0-tis-15-8B
Source: Original Platform
2026-08-09 05:27:18 +08:00

17 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_InvalidChatHistorybatch_errors/total_InvalidChatHistorybatch_errors/avg_DaytonaErrorbatch_errors/total_DaytonaErrorbatch_errors/avg_VerifierTimeoutErrorbatch_errors/total_VerifierTimeoutErrorbatch_errors/avg_AgentTimeoutErrorbatch_errors/total_AgentTimeoutErrortiming/cleanup_old_checkpointstiming/save_checkpointsbatch_errors/avg_DaytonaAuthenticationErrorbatch_errors/total_DaytonaAuthenticationErrorbatch_errors/avg_AgentSetupTimeoutErrorbatch_errors/total_AgentSetupTimeoutError
20.006451230.171900.09385270.01174869.69055658.01542418114003.10291780122.010.00.9870.00.00.0130.49220.00540.0974-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0471-0.00.01.00.00.01642104.16062702316.250.00.95130.01470.60660.492211.263250.6464591.516331.1857.9675266.45120.07175.246236.047201.8628238.22318.06174940.94141.3945202.10374656.07680.00.00355846438922750.0862068965517241451.0172413793103448590.12068965517241387.00.0172413793103448271.00.12068965517241387.0
30.006451211.011.04019.62113506.74334314.72313176114137.7671417266.017.00.95910.00.00.04090.36520.00010.08450.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.04420.00.01.00.00.0192973045248.0849779032064.00.00.85090.01530.48440.365214.051251.1256585.374131.8857.9675272.59340.06746.755429.7924182.7126212.8136608.870636.372182.9532352.92210.00.00366451246417410.07812550.5320.031252.00.1093757.00.031252.011.849121.5894
40.006451221.578100.89064050.47273962.21944155.32052624813351.82071378599.012.00.98740.00.00.01260.5430.00020.10030.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.04290.00010.01.00.00.0164451543132274688.05.2744631064356454e+170.00.91140.01460.6250.54314.052951.2486585.13331.8857.9675272.83450.07635.023227.0398157.8818185.24021221.984832.8949158.1234998.8210.00.00376048042616510.0666666666666666740.8480.066666666666666674.00.053.00.033333333333333332.0
50.006451231.781200.89064166.38093516.15954705.13573175913839.53711482005.013.00.95620.00.00.04380.4531-0.00240.1352-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0404-0.00070.01.00.00.0195167771222540288.06.12676136445621e+180.00.88580.01360.57810.453114.587551.5708583.268232.0857.9675274.69930.08746.166233.6069178.474212.39331845.86536.6567178.69851590.64340.00.00386249642921670.1129032258064516170.8387096774193549520.145161290322580669.00.0161290322580645161.00.00879.8733
60.006451241.953100.89064455.77154539.24184379.77613156814244.58611548541.019.00.97150.00.00.02850.47660.00510.0819-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0407-0.00.01.00.00.0174.4555619643529626e+195.5445188610444524e+220.00.87050.01320.57810.476614.948252.0431582.498732.1857.9675275.46880.08746.755137.2039192.8015230.351216.269535.0045193.0581944.15410.00.00396249642423720.1129032258064516170.7580645161290323470.2258064516129032214.00.064516129032258064.0
70.006451251.890600.89063978.80474426.84943586.56042063013356.53781380244.018.00.97620.00.00.02380.46680.00080.09720.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.03960.00.01.00.00.01518027375480602624.09.356849449931899e+180.00.86350.01520.53120.466814.753551.8218581.713432.2857.9675276.25410.05584.353224.4683152.7384177.48826.73981338.864639.4864152.96361117.53030.00.00406148842518550.0163934426229508210.7049180327868853430.1967213114754098312.00.131147540983606568.00.032786885245901642.00.00738.861
80.006451241.671900.82813751.06643454.83474004.36592889013641.00911296617.012.00.98770.00.00.01230.46090.00220.08-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0406-0.00.01.00.00.01691345461420032.01527438681047040.00.00.87630.01390.57810.460914.970551.9844581.638132.2857.9675276.32940.07975.306630.2736162.8792193.48361768.75931.3917163.12981538.57170.00.00416451244921620.15625100.703125450.0468753.00.0156251.00.06254.0
90.006451241.937500.90624489.39844501.68724481.83913168314385.2931584898.026.00.9390.00.00.0610.38090.01190.06730.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.04340.00.01.00.00.0124602914881536.0717373355065344.00.00.88190.01520.43750.380915.463852.4785580.465832.3857.9675277.50180.06716.195132.6832192.1724225.21521528.471236.6995192.46441260.35680.00.00426249644520450.1451612903225806690.5645161290322581350.09677419354838716.00.0161290322580645161.00.00918.6068
100.006451231.734400.90624579.65434076.38684935.29673179514209.47911582217.011.00.97420.00.00.02580.41410.0030.09330.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.04230.00.01.00.00.01481.2523732566881721e+221.575312744563639e+250.00.90560.01420.50.414115.386852.4961580.984132.3857.9675276.98340.09695.883135.5495203.7713239.68731429.602946.2261204.04031137.79790.00.00435846440416520.0862068965517241450.7068965517241379410.2068965517241379312.00.137931034482758628.00.0344827586206896552.0
110.006451241.703100.8753711.04493450.71543979.63893185813450.52051281843.021.00.95840.00.00.04160.50780.00260.09040.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.03890.00010.01.00.00.018610689648640.012101714182144.00.00.87160.01350.60940.507815.39352.5665580.203132.4857.9675277.76440.06746.939727.3191159.9392187.5421259.623727.2463160.15511037.89050.00.00446451247714350.0937560.25160.1258.00.031252.00.00889.1974
120.006451241.796900.92194316.80474018.29394590.71913180213563.25531489843.018.00.96230.00.00.03770.47850.00110.07850.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.04130.00.01.00.00.01431.399725969946154e+241.7084933114951544e+270.00.94250.01430.53120.478515.408152.5731579.763132.4857.9675278.20450.06816.99633.0651177.8596211.23386.26051402.853432.9762178.10021151.64170.00.00456350445615480.0634920634920634940.5396825396825397340.142857142857142859.00.0158730158730158721.0
130.006451242.12500.96884054.17383689.8884427.09883176713753.67331394041.018.00.96280.00.00.03720.50590.00010.1048-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0372-0.00010.01.00.00.019667826126848.0788490430185472.00.00.91080.01290.57810.505915.49152.6367579.898532.4857.9675278.0690.06776.208529.2944171.0908200.6751235.101635.891171.3125992.32060.00.00466350446516390.0634920634920634940.49206349206349204310.11111111111111117.00.0476190476190476163.00.00858.8618
140.006451242.078100.89064380.81054373.78474388.89923178714245.67961564842.028.00.92360.00.00.07640.53520.00020.11010.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.040.00.01.00.00.01786668615680000.06459941124571136.00.00.9170.01380.6250.535215.55152.6967579.894732.4857.9675278.07280.06835.957730.8067192.315223.441375.874534.2379192.56451112.2330.00.00476048042917510.0530.8166666666666667490.066666666666666674.00.033333333333333332.0
150.006451241.734400.78123799.55663462.69144136.42193166013327.29141342599.018.00.97590.00.00.02410.5-0.01040.06430.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.04010.00.01.00.00.01866386052608.07399555989504.00.00.89170.01340.59380.516.221553.1401577.912832.6857.9675280.05470.08755.647129.4067156.0174185.7121598.919629.9299156.21731377.62240.00.00486350442929670.15873015873015872100.6190476190476191390.269841269841269817.00.126984126984126988.00.063492063492063494.00.00818.3558
160.006451241.812500.85943419.85553529.50673333.8923176413464.68411225041.09.00.96020.00.00.03980.43950.00360.09330.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.03810.00010.01.00.00.019322766618624.022522718322688.00.00.85250.01260.57810.439516.536153.4547577.409132.7857.9675280.55840.0635.987928.4774153.5356182.34681442.621429.7255153.8061224.55620.00.00496350444824490.111111111111111170.5079365079365079320.11111111111111117.00.11111111111111117.0
170.006451241.937500.89064705.5823464.16975295.8793174214707.65281701349.045.00.90290.00.00.09710.32230.00020.0516-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0422-0.00010.01.00.00.0187464861233152.0492378775355392.00.00.90110.01460.39060.322316.545553.464576.993932.7857.9675280.97360.09037.054133.012211.9908245.32935.78841863.317539.9846212.22651570.94360.00.00506249641124780.1129032258064516170.7903225806451613490.2419354838709677515.00.112903225806451617.00.112903225806451617.00.00858.5286
180.006451231.62500.89063573.10743325.93963709.42422169113523.26541255628.017.00.98170.00.00.01830.3555-0.00510.0953-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0379-0.00.01.00.00.0153165468454912.0185397279719424.00.00.83230.01320.48440.355516.330353.2484577.541532.7857.9675280.4260.05745.396629.4475162.4351192.16891459.975733.9624162.66361228.44140.00.00516451242029920.171875110.875560.26562517.00.1406259.0
190.006451241.843800.96883976.17584500.20313661.75942985413987.16471358843.025.00.98480.00.00.01520.3750.00050.099-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0394-0.00.01.00.00.01517481026.57046713344.00.00.82850.01480.46880.37516.674353.6125576.834532.8857.9675281.1330.0845.613329.8419175.3867205.55181339.777438.5075175.62541090.09950.00.00526350443323610.01587301587301587210.49206349206349204310.3809523809523809324.00.063492063492063494.00.095238095238095236.00.00728.61180.095238095238095236.0
200.006451231.781200.90623691.34573547.40333821.37552243613372.43461290968.011.00.98610.00.00.01390.4746-0.00420.07090.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.03810.00.01.00.00.01241.2458670639324969e+231.4152924048125494e+260.00.86360.01320.56250.474616.441753.4303577.742932.7857.9675280.22470.06855.458925.5377153.0849178.94881705.945529.4538153.34211492.07620.00.00536451246319480.062540.5625360.1258.00.0156251.0
210.006451241.718800.90623767.52153727.45933795.15511898612995.15671342725.037.00.9680.00.00.0320.40820.00640.0877-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0444-0.00.01.00.00.0185327695663104.0379507068370944.00.00.90920.01510.48440.408216.346553.3348577.027832.7857.9675280.93970.04443.843525.2314140.1255165.65411372.408333.4557140.37781169.45090.00.00546249642622700.0483870967741935530.9354838709677419580.064516129032258064.00.080645161290322585.00.00879.1348
220.006451242.046900.92194275.68363941.04864465.00313166214375.74591463491.011.00.97140.00.00.02860.3613-0.00460.10220.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.04080.00.01.00.00.02164036787148161024.04.865830434606416e+180.00.86040.01390.46880.361316.603453.5885576.666732.8857.9675281.30080.1576.157833.4267203.4849237.33525.86581613.599628.3574203.75111341.74590.00.00556148842421570.0491803278688524630.819672131147541500.016393442622950821.00.114754098360655747.00.147540983606557379.0
230.006451241.734400.89064390.92773929.15134585.93176714238.37571548471.022.00.94010.00.00.05990.2969-0.0010.0628-0.00010.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.045-0.00420.01.00.00.0158617810003492864.01.0647384730359562e+190.00.87610.0150.3750.296916.706553.8438576.380232.8857.9675281.58730.09076.141130.9645191.9417223.22411633.377135.1631192.16841368.84390.00.00566451247815340.0937560.390625250.0468753.00.00818.2677
240.006451241.546900.82814453.51564019.91414887.11723174014404.6081552609.034.00.91440.00.00.08560.50.00240.1244-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0436-0.00010.01.00.00.01473.5539692025682697e+213.740553229295638e+240.00.93820.01580.60940.516.78753.9058575.740732.9857.9675282.22680.09086.133238.8151208.6843247.80621501.61535.5265208.89971212.14170.00.00576148845511290.0327868852459016420.5081967213114754310.016393442622950821.00.065573770491803284.00.081967213114754095.0
250.006451231.703100.8753884.98634063.31613777.0943184413730.86381405568.016.00.94870.00.00.05130.3770.00850.1314-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0444-0.00.01.00.00.01895.697833777259119e+216.752966200220271e+240.00.9210.01510.50.37717.203254.3492574.893633.0857.9675283.07390.11296.287328.6923178.963207.9881448.736630.4954179.18241203.95680.00.00586451245316530.0312520.671875430.0781255.00.1258.00.00788.58970.0156251.0
260.006451241.593800.8753719.32813822.81423577.51393165213793.5661302863.09.00.94940.00.00.05060.57810.00610.1067-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0407-0.00080.01.00.00.01692.916946366431232e+163.229335359815313e+190.00.88940.01410.68750.578117.376954.5306573.513433.2857.9675284.45410.095.975731.3476170.0586201.78581663.172533.9156170.34781421.48820.00.00596249646015350.0483870967741935530.5310.032258064516129032.00.0161290322580645161.0
270.006451231.718800.8754468.5022738.58715586.54983179914489.81291573926.034.00.93580.00.00.06420.3926-0.01780.0646-0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0453-0.00.01.00.00.01395716752465920.07041144689000448.00.00.93210.01550.46880.392617.302454.4753572.283933.3857.9675285.68360.1086.335635.3958205.7623241.54326.21071507.150125.3052206.0391233.95710.00.00606249643920570.1129032258064516170.6935483870967742430.080645161290322585.00.00798.517