INDEX

Explanations

mentions of instructions or options related to conditions and preferences

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

otherwise if

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

yarnpkg

-0.48

Trayectoria

-0.47

LayoutStyle

-0.46

 Blat

-0.45

gdala

-0.44

 afge

-0.43

RTCF

-0.43

<#

-0.42

compensation

-0.42

 HasFactory

-0.42

POSITIVE LOGITS

oneofs

0.41

PrivateRoute

0.38

хьтан

0.37

 contemporáneo

0.36

toContain

0.34

soort

0.34

 pourriez

0.34

styrelsen

0.33

हरु

0.33

égi

0.33

Activations Density 0.069%

mentions of instructions or options related to conditions and preferences

otherwise if

No Comments

No Known Activations