INDEX

Explanations

code defining classes and implementations in programming

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

programming context or user input

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

 lenker

-0.53

avía

-0.47

 Lightboxes

-0.47

arela

-0.47

twimg

-0.43

 Pounds

-0.43

sima

-0.42

tedrale

-0.42

pgterms

-0.42

getSession

-0.42

POSITIVE LOGITS

setVerticalGroup

0.46

MIDDLEWARE

0.45

providedIn

0.42

 ویکی‌پدیا

0.38

 MainAxisSize

0.38

 ujednoznacz

0.37

MethodManager

0.36

oredCriteria

0.36

ỡng

0.36

istoitu

0.35

Activations Density 0.000%

code defining classes and implementations in programming

programming context or user input

No Comments

No Known Activations