Files
shopmanager-grpo-smoke-l4-v2/metrics.csv
ModelHub XC b8ef3bde9b 初始化项目,由ModelHub XC社区提供模型
Model: hard007ik/shopmanager-grpo-smoke-l4-v2
Source: Original Platform
2026-05-02 12:37:45 +08:00

4.1 KiB

1steplossgrad_normlearning_ratenum_tokenscompletions/mean_lengthcompletions/min_lengthcompletions/max_lengthcompletions/clipped_ratiocompletions/mean_terminated_lengthcompletions/min_terminated_lengthcompletions/max_terminated_lengthrewards/reward_total/meanrewards/reward_total/stdrewards/reward_market/meanrewards/reward_market/stdrewards/reward_warehouse/meanrewards/reward_warehouse/stdrewards/reward_showroom/meanrewards/reward_showroom/stdrewardreward_stdfrac_reward_zero_stdsampling/sampling_logp_difference/meansampling/sampling_logp_difference/maxsampling/importance_sampling_ratio/minsampling/importance_sampling_ratio/meansampling/importance_sampling_ratio/maxentropyclip_ratio/low_meanclip_ratio/low_minclip_ratio/high_meanclip_ratio/high_maxclip_ratio/region_meanstep_timeepochtrain_runtimetrain_samples_per_secondtrain_steps_per_secondtotal_flostrain_loss
21-0.00.00.06155.018.513.024.00.018.513.024.00.87374997138977050.0140714412555098530.60000002384185790.00.200000002980232240.00.073750004172325130.014071426354348660.87374997138977050.0140714421868324280.04.52034139633178727.1742382049560554.035991810979052e-403.371377950408304e-346.742751768219281e-340.18117728829383850.00.00.00.00.04.6183280850018490.0033333333333333335
320.00.05.000000000000001e-0713990.020.513.028.00.020.513.028.00.90200001001358030.040587920695543290.200000002980232240.00.60000002384185790.00.101999998092651370.040587935596704480.90200001001358030.040587920695543290.04.293838500976562526.6465091705322272.0038568039844884e-439.08055335179305e-341.81611067035861e-330.118028290569782260.00.00.00.00.06.26182352700197950.006666666666666667
430.00.01.0000000000000002e-0627189.047.540.055.00.047.540.055.00.8000000119209290.00.200000002980232240.00.60000002384185790.00.00.00.8000000119209290.01.02.25201129913330128.7227287292480470.05.605193857299268e-451.2611686178923354e-440.120176136493682860.00.00.00.00.010.5593015830017980.01
540.00.01.5e-0632537.016.016.016.00.016.016.016.00.87269997596740720.0083438539877533910.200000002980232240.00.60000002384185790.00.072699993848800660.0083438595756888390.87269997596740720.0083438539877533910.06.42499351501464830.2747344970703121.401298464324817e-452.802596928649634e-452.802596928649634e-450.174016386270523070.00.00.00.00.01.95528083399767640.013333333333333334
65-0.00.02.0000000000000003e-0641360.032.012.052.00.032.012.052.00.83799999952316280.053740095347166060.60000002384185790.00.200000002980232240.00.037999998778104780.0537401139736175540.83799999952316280.053740095347166060.03.495557546615600631.435132980346680.02.3879863577791495e-324.775972715558299e-320.105947196483612060.00.00.00.00.08.4673852480009370.016666666666666666
760.00.02.5e-0653669.047.034.060.00.047.034.060.00.8000000119209290.00.200000002980232240.00.60000002384185790.00.00.00.8000000119209290.01.02.415662527084350628.9523735046386720.01.090479293292958e-332.180958586585916e-330.13157893717288970.00.00.00.00.08.834365269998670.02
87-0.00.03e-0662446.026.013.039.00.026.013.039.00.84815001487731930.0246780328452587130.200000002980232240.00.60000002384185790.00.048149999231100080.0246780272573232650.84815001487731930.0246780328452587130.03.940788030624389628.37194824218750.06.8739474988531065e-341.3747894997706213e-330.106197118759155270.00.00.00.00.08.3261336590003340.023333333333333334
980.00.03.5e-0672064.029.519.040.00.029.519.040.00.84385001659393310.062013272196054460.200000002980232240.00.60000002384185790.00.043850000947713850.062013264745473860.84385001659393310.062013272196054460.03.499578237533569330.463426589965827.006492321624085e-458.407790785948902e-459.80908925027372e-450.115790568292140960.00.00.00.00.08.3655062139987420.02666666666666667
1080.0266666666666666791.21820.1750.0880.0-7.903189287664419e-34