INDEX

Explanations

U followed by 8

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

This neuron fires on uppercase C/C++ macro and attribute tokens (like PROGMEM, U8G2_FONT_SECTION, maybe) used for compiler directives or section annotations.

oai_token-act-pair · o4-mini

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

 whereas

-1.18

調味料

-1.14

 while

-1.13

糧

-1.11

 egiten

-1.08

in

-1.06

 whilst

-1.05

it

-1.05

all

-1.04

POSITIVE LOGITS

 Эти

1.28

 эти

1.19

bedienung

1.15

すみません

1.09

 különböző

1.05

These

1.05

久し

1.04

ミノ

1.04

these

1.02

zusetzen

1.02

Activations Density 0.014%

U followed by 8

This neuron fires on uppercase C/C++ macro and attribute tokens (like PROGMEM, U8G2_FONT_SECTION, maybe) used for compiler directives or section annotations.

No Known Activations

U followed by 8

This neuron fires on uppercase C/C++ macro and attribute tokens (like PROGMEM, U8G2_FONT_SECTION, maybe) used for compiler directives or section annotations.

No Known Activations

U followed by 8

This neuron fires on uppercase C/C++ macro and attribute tokens (like PROGMEM, U8G2_FONT_SECTION, __maybe__) used for compiler directives or section annotations.

No Known Activations

U followed by 8

This neuron fires on uppercase C/C++ macro and attribute tokens (like PROGMEM, U8G2_FONT_SECTION, __maybe__) used for compiler directives or section annotations.

No Known Activations

This neuron fires on uppercase C/C++ macro and attribute tokens (like PROGMEM, U8G2_FONT_SECTION, maybe) used for compiler directives or section annotations.

This neuron fires on uppercase C/C++ macro and attribute tokens (like PROGMEM, U8G2_FONT_SECTION, maybe) used for compiler directives or section annotations.