INDEX

Explanations

mentions of reviews and opinions about products or media

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

start of user input

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

MigrationBuilder

-0.59

yntaxException

-0.59

 surla

-0.58

GOTREF

-0.54

DockStyle

-0.52

 '\\;'

-0.52

sizeCache

-0.50



-0.48

 ब्रेकडाउन

-0.47

 дописавши

-0.46

POSITIVE LOGITS

 चीज़ों

0.41

 recensioni

0.40

éc

0.37

 endregion

0.37

 complaints

0.36

🦲

0.36

audiovisuel

0.35

adple

0.35

 postings

0.34

 lewd

0.34

Activations Density 0.003%

mentions of reviews and opinions about products or media

start of user input

No Comments

No Known Activations