### gemma2-9b · full · joint
   idiom config:    {'model': 'google/gemma-2-9b', 'reduction': 'joint', 'medial_only': False, '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': False, '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
   1.1299     1.1224  [  1.1080,  1.1514]

== non-idioms :: ratio_u_idiom ==  (N=18 phrases)
     mean     median                    95% CI
   1.0268     1.0200  [  1.0171,  1.0378]
  cross-dataset ratio_u_idiom: idioms - nonidioms  Δ=+0.1031  CI=[+0.0787,+0.1273]  *

== idioms :: ratio_s_idiom ==  (N=3 phrases)  (15 non-finite dropped)
     mean     median                    95% CI
   1.0002     1.0002  [  1.0001,  1.0003]

== non-idioms :: ratio_s_idiom ==  (N=1 phrases)  (17 non-finite dropped)
     mean     median                    95% CI
   1.0009     1.0009  [  1.0009,  1.0009]
  cross-dataset ratio_s_idiom: idioms - nonidioms  Δ=-0.0007  CI=[-0.0008,-0.0006]  *