INDEX

Explanations

instances of mathematical operations and problem-solving within the text

oai_token-act-pair · gpt-4o-mini Triggered by @bot

studying something, following neck, Facing Recall

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

the neuron fires on numeric/math content — tokens that are numbers, math operators/terms, or words introducing calculations (e.g., "Solve", "Convert", numeric values, units).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 otomatig

-0.44

 CreateTagHelper

-0.44

château

-0.44

antaranya

-0.42

WriteBarrier

-0.40

MethodManager

-0.40

Autoritní

-0.40

cherichia

-0.40

 manguera

-0.39

Życiorys

-0.39

POSITIVE LOGITS

 Italijanski

0.49

 الرياضيه

0.48

Fns

0.44

 SDLK

0.42

Kud

0.42

 <<<<<<<<<<<<<<

0.42

enumii

0.39

recycled

0.38

 shove

0.36

 مرئيه

0.36

Activations Density 0.298%

instances of mathematical operations and problem-solving within the text

studying something, following neck, Facing Recall

the neuron fires on numeric/math content — tokens that are numbers, math operators/terms, or words introducing calculations (e.g., "Solve", "Convert", numeric values, units).

No Comments

No Known Activations