### gemma2-9b · medial · joint
   idiom config:    {'model': 'google/gemma-2-9b', 'reduction': 'joint', 'medial_only': True, 'dtype': 'bfloat16', 'dataset': '/home/prada/PID_evaluation/data/dataset.tsv', 'num_idioms': 18, 'syn_reg_eps': 0.01}
   nonidiom config: {'model': 'google/gemma-2-9b', 'reduction': 'joint', 'medial_only': True, 'dtype': 'bfloat16', 'dataset': '/home/prada/PID_evaluation/data/nonidioms_dataset.tsv', 'num_idioms': 18, 'syn_reg_eps': 0.01}

== idioms :: ratio_u_idiom ==  (N=18 phrases)
     mean     median                    95% CI
   1.1730     1.1629  [  1.1392,  1.2086]

== non-idioms :: ratio_u_idiom ==  (N=18 phrases)
     mean     median                    95% CI
   1.0413     1.0294  [  1.0263,  1.0579]
  cross-dataset ratio_u_idiom: idioms - nonidioms  Δ=+0.1317  CI=[+0.0945,+0.1712]  *

== idioms :: ratio_s_idiom ==  (N=10 phrases)  (8 non-finite dropped)
     mean     median                    95% CI
   1.0004     1.0000  [  1.0000,  1.0008]

== non-idioms :: ratio_s_idiom ==  (N=1 phrases)  (17 non-finite dropped)
     mean     median                    95% CI
   1.0002     1.0002  [  1.0002,  1.0002]
  cross-dataset ratio_s_idiom: idioms - nonidioms  Δ=+0.0002  CI=[-0.0002,+0.0006]