INDEX

Explanations

institution names followed by for or of

np_acts-logits-general · gemini-2.5-flash-lite

organization names followed by 'of' or 'for'

np_acts-logits-general · gemini-2.5-flash-lite

say "of"

np_max-act-logits · claude-4-5-sonnet Triggered by @sripadkarne

New Auto-Interp

Configuration

google/gemma-scope-2-27b-pt/resid_post/layer_31_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

an

0.47

 제일

0.46

 istem

0.43

 ඉතා

0.43

 मू

0.42

م

0.42

hishek

0.41

ค

0.40

 దా

0.40

meida

0.40

POSITIVE LOGITS

of

0.80

of

0.75

ของการ

0.75

for

0.75

OfDeath

0.71

для

0.70

 ഓഫ്

0.69

 ఆఫ్

0.66

 для

0.65

for

0.64

Activations Density 0.574%

institution names followed by for or of

organization names followed by 'of' or 'for'

say "of"

No Comments

No Known Activations

institution names followed by for or of

organization names followed by 'of' or 'for'

say "of"

No Comments

No Known Activations