Downstream task evaluation after post-training stage RLVR1 for 13b_trkd.
Source: source_paper:Table 8, Page 33, After RLVR1
Reported and observed measurements
downstream_accuracy
t8_rlvr1_13b_trkd_gsm8k
Reported 72.4 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_gsm_s
Reported 53.3 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_gsmplus
Reported 47.6 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_bbh
Reported 32.7 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_drop
Reported 35 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_math
Reported 10.4 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_tqa
Reported 54.7 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_nq
Reported 22.2 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_sqa
Reported 8.6 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_mmlu
Reported 36.6 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_mmlu_p
Reported 16.6 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_arc_c
Reported 54.1 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_obqa
Reported 54.6 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_wino
Reported 51.7 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_agi
Reported 34.5 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_13b_trkd_ife
Reported 67.1 percentage_points
Observed — percentage_points
Assessments (0)
No immutable Assessment has been published for this Claim yet.
Runs (0)
No execution has been linked to this Claim yet.