INDEX

Explanations

references to the word "New" in various contexts

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

"New" followed by a location

np_acts-logits-general · gemini-2.0-flash

new place names

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/*

-0.62

ymce

-0.57

 Bluffs

-0.55

 aanwezig

-0.54

ftagPool

-0.54

 ModelExpression

-0.53

estors

-0.53

PRD

-0.53

よいよ

-0.51

 civiles

-0.51

POSITIVE LOGITS

tonsoft

1.01

 york

1.00

fang

0.94

 York

0.92

sworth

0.91

york

0.90

 Zealand

0.88

York

0.79

 zealand

0.79

sprint

0.77

Activations Density 0.163%

references to the word "New" in various contexts

"New" followed by a location

new place names

No Comments

No Known Activations