INDEX

Explanations

code type declarations

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

the word "Types" in programming code, especially when used in enums, class names, or constant declarations.

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

The neuron strongly activates on occurrences of the exact token “Types.”

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

さんの

-2.67

↵

-2.44

-2.30

薷

-2.11

 Erstellung

-2.11

 resonate

-1.98

 извест

-1.95

蜕

-1.92

増えて

-1.89

-1.88

POSITIVE LOGITS

2.97

 exista

2.69

ၞ

2.63

ﱠ

2.36

In

2.28

"}}>

2.14

 geldi

2.13

 绘画

2.13

坵

2.11

ၜ

2.11

Activations Density 0.011%

code type declarations

the word "Types" in programming code, especially when used in enums, class names, or constant declarations.

The neuron strongly activates on occurrences of the exact token “Types.”

No Comments

No Known Activations

code type declarations

the word "Types" in programming code, especially when used in enums, class names, or constant declarations.

The neuron strongly activates on occurrences of the exact token “Types.”

No Comments

No Known Activations