INDEX

Explanations

themes related to censorship and historical narrative manipulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ReusableCell

-0.58

WebServlet

-0.57

 propOrder

-0.54

 ujednoznacz

-0.53

RenderAtEndOf

-0.48

Personendaten

-0.46

ContentAsync

-0.46

 dAtA

-0.45

ValueStyle

-0.45

UserScript

-0.45

POSITIVE LOGITS

 conceal

0.57

 ocultar

0.54

 concealment

0.54

 hidden

0.52

 hide

0.51

 hides

0.51

 concealed

0.50

讳

0.50

 deny

0.47

 suppressed

0.47

Activations Density 0.514%

themes related to censorship and historical narrative manipulation

No Comments

No Known Activations