INDEX

Explanations

references to places or positional phrases

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

the word "in"

np_acts-logits-general · gemini-2.0-flash

in Residence Education

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

󠁢

-0.53

specialchars

-0.50

MessageTagHelper

-0.49

openConnection

-0.48

 callbacks

-0.47

 BoxDecoration

-0.46

Ế

-0.46

isNameExpr

-0.46

 areas

-0.46

ruptedException

-0.46

POSITIVE LOGITS

 kaarangay

0.88

Jereo

0.81

 Wikimedijinoj

0.76

Particular

0.66

 ivelany

0.66

 beginnetje

0.66

tagHelperRunner

0.63

 Particular

0.63

 particuliers

0.63

 whom

0.63

Activations Density 0.348%

references to places or positional phrases

the word "in"

in Residence Education

No Comments

No Known Activations