INDEX

Explanations

references to brands and brand-related terms

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

"Brand new"

np_acts-logits-general · gemini-2.0-flash

brand new

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Дереккөздер

-0.79

 >=",

-0.76

 تانيه

-0.76

EndProject

-0.73

 Heere

-0.73

出版年

-0.71

 ब्रेकडाउन

-0.70

 referenties

-0.69

 خارجية

-0.69

 noqa

-0.68

POSITIVE LOGITS

 BRAND

0.85

 name

0.84

 brand

0.84

new

0.80

BRAND

0.77

ishing

0.77

 brands

0.77

 Brand

0.76

Brands

0.75

 Brands

0.75

Activations Density 0.057%

references to brands and brand-related terms

"Brand new"

brand new

No Comments

No Known Activations