INDEX

Explanations

physical attributes and accessories commonly associated with characters or fashion

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

hats, caps, and aprons

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

новништво

-0.77

ParallelGroup

-0.56

ﬅ

-0.54

Personendaten

-0.48

 iprot

-0.47

енча

-0.46

 nicolas

-0.46

runApp

-0.46

 newOwner

-0.46

ReusableCell

-0.45

POSITIVE LOGITS

 AssemblyTitle

0.41

 glit

0.40

Demografia

0.40

 WebDriverWait

0.35

хівовано

0.35

 lenker

0.35

 bells

0.34

 smog

0.34

 tremors

0.34

 tiara

0.33

Activations Density 0.042%

physical attributes and accessories commonly associated with characters or fashion

hats, caps, and aprons

No Comments

No Known Activations