INDEX

Explanations

technical documentation and code

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

indentation or whitespace at the beginning of lines in source code, particularly where function parameters or arguments span multiple lines.

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

The neuron activates on numeric literals (integer or floating-point constants) in code.

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

 artístico

-1.47

了一会儿

-1.40

一旁

-1.36

 colorear

-1.31

 would

-1.31

 forty

-1.29

nexo

-1.29

lyd

-1.28

 averiguar

-1.28

POSITIVE LOGITS

paedia

1.49

広く

1.41

 Почему

1.40

嗱

1.40

 facie

1.39

 (‘

1.34

 those

1.33

 Their

1.32

wyczaj

1.30

 dewasa

1.29

Activations Density 0.008%

technical documentation and code

indentation or whitespace at the beginning of lines in source code, particularly where function parameters or arguments span multiple lines.

The neuron activates on numeric literals (integer or floating-point constants) in code.

No Comments

No Known Activations

technical documentation and code

indentation or whitespace at the beginning of lines in source code, particularly where function parameters or arguments span multiple lines.

The neuron activates on numeric literals (integer or floating-point constants) in code.

No Comments

No Known Activations