INDEX

Explanations

remaster, error, import, component, development

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

technical terms related to software errors, remasters, or manufacturing/engineering processes.

oai_token-act-pair · claude-4-5-sonnet Triggered by @bfgcn7frf5

The neuron responds strongly to words exhibiting common English derivational suffixes (e.g. –ing, –ed, –ness, –ance) that turn stems into agent/process or quality nouns.

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

酲

-3.39

牁

-3.20

蟜

-3.06

 implementar

-2.73

 støtte

-2.70

ﻂ

-2.67

 hjemmes

-2.64

 ſta

-2.59

 faſt

-2.55

 prøve

-2.47

POSITIVE LOGITS

3.23

</b>

3.11

or

3.02

 Which

2.88

2.80

2.66

2.52

ly

2.50

Activations Density 0.013%

remaster, error, import, component, development

technical terms related to software errors, remasters, or manufacturing/engineering processes.

The neuron responds strongly to words exhibiting common English derivational suffixes (e.g. –ing, –ed, –ness, –ance) that turn stems into agent/process or quality nouns.

No Comments

No Known Activations

remaster, error, import, component, development

technical terms related to software errors, remasters, or manufacturing/engineering processes.

The neuron responds strongly to words exhibiting common English derivational suffixes (e.g. –ing, –ed, –ness, –ance) that turn stems into agent/process or quality nouns.

No Comments

No Known Activations