INDEX

Explanations

geographic locations and notable landmarks

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

places and photography

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

parsedMessage

-0.60

 otomatig

-0.57

 kasarigan

-0.57

unknownFields

-0.54

OGND

-0.50

protoimpl

-0.49

tvguidetime

-0.47

Française

-0.47

 INTERESAR

-0.47

ंदीखरीदारी

-0.46

POSITIVE LOGITS

 Fotograf

0.42

 photographed

0.40

 skyline

0.38

fotograf

0.35

AddTagHelper

0.35

loài

0.34

 scenery

0.34

 fotograf

0.33

FormGroup

0.33

 fotogra

0.32

Activations Density 0.014%

geographic locations and notable landmarks

places and photography

No Comments

No Known Activations