INDEX

Explanations

instances of the word "display" and related terms in the context of showcasing or presenting items

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

items on display

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 avoient

-0.53

 mourut

-0.48

 kvinder

-0.46

 almofada

-0.46

 Goddard

-0.44

تقاوى

-0.43

 meia

-0.43

 preghiera

-0.42

 promesse

-0.42

 Manajemen

-0.41

POSITIVE LOGITS

 AVAILABLE

0.50

vailable

0.50

parsedMessage

0.50

 noqa

0.49

ViewChild

0.49

 snork

0.49

 Available

0.48

available

0.48

 available

0.47

copyOf

0.47

Activations Density 0.011%

instances of the word "display" and related terms in the context of showcasing or presenting items

items on display

No Comments

No Known Activations