INDEX

Explanations

tubal, tube, tutor

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

cylindrical objects like tubes, tubules, and tutors/tutoring.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The neuron consistently fires on the character sequence “tu” (especially “tub-” and “tube” substrings) wherever it appears in words like tubulin, microtubule, tube, tutor, etc.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

囖

-2.98

 剧照

-2.98



-2.67

 iſt

-2.63

 bunda

-2.61

Ꮉ

-2.59

 umbro

-2.55

踅

-2.53

 новых

-2.45

 individuelles

-2.45

POSITIVE LOGITS

’

2.58

 collectively

2.52

<i>

2.52

 uninterrupted

2.47

section

2.42

 especially

2.31

be

2.28

of

2.25

we

2.23

 extensively

2.19

Activations Density 0.031%

tubal, tube, tutor

cylindrical objects like tubes, tubules, and tutors/tutoring.

The neuron consistently fires on the character sequence “tu” (especially “tub-” and “tube” substrings) wherever it appears in words like tubulin, microtubule, tube, tutor, etc.

No Comments

No Known Activations

tubal, tube, tutor

cylindrical objects like tubes, tubules, and tutors/tutoring.

The neuron consistently fires on the character sequence “tu” (especially “tub-” and “tube” substrings) wherever it appears in words like tubulin, microtubule, tube, tutor, etc.

No Comments

No Known Activations