INDEX

Explanations

instances of various forms of the word "write."

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

Following the word "write" or "writing"

np_acts-logits-general · gemini-2.0-flash

writing or authoring

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 resourceCulture

-0.74

contentLoaded

-0.67

 חיצוניים

-0.64

 تضيفلها

-0.61

uxxxx

-0.61

 Produits

-0.59

 pinch

-0.55

URLException

-0.54

 bailey

-0.53

oraz

-0.53

POSITIVE LOGITS

 writers

0.69

 Writes

0.68

 write

0.67

 writes

0.66

 down

0.65

 Write

0.62

write

0.61

Writing

0.61

 ulang

0.60

writers

0.60

Activations Density 0.150%

instances of various forms of the word "write."

Following the word "write" or "writing"

writing or authoring

No Comments

No Known Activations