INDEX

Explanations

double-blind studies

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

technical terms in academic, scientific, or programming contexts.

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

This neuron detects terminology related to study design and statistical reporting (e.g. “error bars,” “cumulatively,” “double-blind,” “randomly assigned,” etc.).

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

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/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

it

-1.66

↵↵

-1.58

 capucha

-1.33

 pizarra

-1.30

始め

-1.27

тент

-1.27

YOND

-1.26

渙

-1.25

 będzie

-1.23

寻求

-1.23

POSITIVE LOGITS

as

1.89

 from

1.66

on

1.52

€“

1.41

matisch

1.38

;%%

1.36

 něj

1.34

to

1.33

gingen

1.31

нии

1.30

Activations Density 0.086%

double-blind studies

technical terms in academic, scientific, or programming contexts.

This neuron detects terminology related to study design and statistical reporting (e.g. “error bars,” “cumulatively,” “double-blind,” “randomly assigned,” etc.).

No Comments

No Known Activations

double-blind studies

technical terms in academic, scientific, or programming contexts.

This neuron detects terminology related to study design and statistical reporting (e.g. “error bars,” “cumulatively,” “double-blind,” “randomly assigned,” etc.).

No Comments

No Known Activations