INDEX

Explanations

code delimiters and quotes

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

The neuron fires on tokens that are names of imported modules or symbols in import/require statements.

oai_token-act-pair · o4-mini

Triggered by @jyhe0408

semicolons at the end of import statements in programming code.

oai_token-act-pair · claude-4-5-sonnet

Triggered by @jyhe0408

CamelCase/PascalCase identifiers in source code, especially class, type, and function names.

oai_token-act-pair · gpt-5

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

Negative Logits

ón

-1.20

늠

-1.19

了不少

-1.16

 خمس

-1.13

 "$@"

-1.08

了些

-1.07

 notwendig

-1.06

点点头

-1.05

 MUCH

-1.05

стра

-1.05

POSITIVE LOGITS

 will

1.32

žiť

1.22

it

1.16

 macetas

1.15

 would

1.13

↵↵↵

1.13

 secar

1.13

 estekak

1.11

1.09

'],

1.07

Activations Density 0.027%

code delimiters and quotes

The neuron fires on tokens that are names of imported modules or symbols in import/require statements.

semicolons at the end of import statements in programming code.

CamelCase/PascalCase identifiers in source code, especially class, type, and function names.

No Known Activations

code delimiters and quotes

The neuron fires on tokens that are names of imported modules or symbols in import/require statements.

semicolons at the end of import statements in programming code.

CamelCase/PascalCase identifiers in source code, especially class, type, and function names.

No Known Activations