INDEX

Explanations

references to waterfront locations

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

near water locations

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

 RISERVATA

-0.54

 Cassius

-0.53

%");

-0.50

cepción

-0.50

__':

-0.49

Kjelder

-0.49

 nagging

-0.49

()")

-0.48

[]]

-0.48

Conclusão

-0.48

POSITIVE LOGITS

 beachfront

0.58

 waterfront

0.56

 Waterfront

0.46

 beach

0.40

wapV

0.38

DeleteBehavior

0.38

 oprot

0.36

 Beach

0.36

 seaside

0.34

vician

0.34

Activations Density 0.003%

references to waterfront locations

near water locations

No Comments

No Known Activations