INDEX

Explanations

references to whistleblowers and related events or concepts

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

leaks and publications

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

setVerticalGroup

-0.62

RectangleBorder

-0.61

LayoutPanel

-0.52

허

-0.47

IsMutable

-0.47

localctx

-0.47

DIPSETTING

-0.46

 propOrder

-0.45

styleable

-0.45

 typelib

-0.45

POSITIVE LOGITS

Leaks

0.62



0.44

CIA

0.44

ข่าว

0.44

 journalistic

0.42

 dénon

0.42

Tikang

0.42

 bombe

0.41

 bombar

0.41

ſelf

0.40

Activations Density 0.021%

references to whistleblowers and related events or concepts

leaks and publications

No Comments

No Known Activations