INDEX

Explanations

conjunctions and phrases indicating conditions or relationships among items

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

or characters preceding specific code or URLs

np_acts-logits-general · gemini-2.0-flash

used to generate graphs

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

 للمعارف

-0.75

kháu

-0.70

UrlResolution

-0.70

principalColumn

-0.68

 Administrativna

-0.67

DockStyle

-0.66

sizeCache

-0.66

 חיצוניים

-0.65

 '\\;'

-0.62

 ویکی‌پدی

-0.62

POSITIVE LOGITS

ing

0.70

 Amar

0.58

 kang

0.58

ing

0.57

 sing

0.55

 Kang

0.53

Dh

0.53

Ing

0.53

Kang

0.51

ke

0.50

Activations Density 0.042%

conjunctions and phrases indicating conditions or relationships among items

or characters preceding specific code or URLs

used to generate graphs

No Comments

No Known Activations