INDEX

Explanations

phrases indicating cessation or changes in status

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

no longer

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

зульта

-0.34

thenes

-0.33

cocho

-0.33

giri

-0.32

шру

-0.31

史

-0.31

loß

-0.31

very

-0.30

ésultat

-0.29

cula

-0.29

POSITIVE LOGITS

 disambiguazione

0.73

 anymore

0.71

 betweenstory

0.68

 enää

0.65

脚注の使い方

0.56

 endregion

0.56

 endblock

0.56

endpush

0.56

+#+

0.55

 FetchType

0.54

Activations Density 0.067%

phrases indicating cessation or changes in status

no longer

No Comments

No Known Activations