INDEX

Explanations

formal academic terminology and statistical variables

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

column headers in tables

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

ViewFeatures

-0.54

 Wiktionnaire

-0.52

 autorytatywna

-0.50

ंदीखरीदारी

-0.50

ChildScrollView

-0.48

Personensuche

-0.48

 GenerationType

-0.47

utilisons

-0.45

 modelAndView

-0.45

RTEX

-0.45

POSITIVE LOGITS

 nahilalakip

0.50

 Spitze

0.43

 Itália

0.38

autoconfigure

0.37

uxxxx

0.36

rens

0.36

ute

0.36

uatan

0.35

สิน

0.35

 tambahan

0.35

Activations Density 0.025%

formal academic terminology and statistical variables

column headers in tables

No Comments

No Known Activations