INDEX

Explanations

instances of the word "is" and its variations in various contexts

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

is the

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

AnimationsModule

-0.55

UseVisualStyle

-0.47

SourceChecksum

-0.47

 protoimpl

-0.45

limus

-0.35

"...

-0.35

enderror

-0.35

ⓧ

-0.34

ImageContext

-0.34

ﬅ

-0.34

POSITIVE LOGITS

 standard

0.53

MENAFN

0.52

 standaard

0.46

 المفت

0.46

NON

0.45

Ārējās

0.45

 abstracto

0.43

后者

0.43

cellaneous

0.43

 cherchés

0.43

Activations Density 0.015%

instances of the word "is" and its variations in various contexts

is the

No Comments

No Known Activations