Downstream task evaluation after post-training stage RLVR1 for 7b_trkd.
Source: source_paper:Table 8, Page 33, After RLVR1
Reported and observed measurements
downstream_accuracy
t8_rlvr1_7b_trkd_gsm8k
Reported 71 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_gsm_s
Reported 54.6 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_gsmplus
Reported 46.8 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_bbh
Reported 33 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_drop
Reported 37.4 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_math
Reported 11.4 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_tqa
Reported 52.8 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_nq
Reported 22.7 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_sqa
Reported 8 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_mmlu
Reported 40.2 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_mmlu_p
Reported 17.8 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_arc_c
Reported 58.9 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_obqa
Reported 55.6 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_wino
Reported 52 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_agi
Reported 35.9 percentage_points
Observed — percentage_points
downstream_accuracy
t8_rlvr1_7b_trkd_ife
Reported 65.2 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.