INDEX

Explanations

mentions of fees and purchasing associated with vanity awards and related scams

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

fees, purchases, and licenses

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

fromnode

-0.62

Personendaten

-0.57

 transfieras

-0.54

AddTagHelper

-0.54

ViewImports

-0.52

Hentet

-0.50

rungsseite

-0.49

曖昧さ回避

-0.48

FormTagHelper

-0.45

 jurado

-0.44

POSITIVE LOGITS

usercontent

0.44

resave

0.44

OwnProperty

0.42

Билгалдахарш

0.41

 سرعت

0.41

 purchase

0.40

systemctl

0.39

Pratique

0.39

tip

0.38

 tickets

0.38

Activations Density 0.046%

mentions of fees and purchasing associated with vanity awards and related scams

fees, purchases, and licenses

No Comments

No Known Activations