INDEX

Explanations

programming and software development-related terms and function calls

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

code comparators and identifiers

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

!*\

-0.41

stdarg

-0.39

RTEX

-0.38

Slf

-0.37

 betweenstory

-0.37

pre

-0.36

/*

-0.36

 plant

-0.36

 tongue

-0.36

HtmlAttribute

-0.35

POSITIVE LOGITS

colgante

0.56

 autorytatywna

0.53

Tikang

0.50

verwijspagina

0.47

 église

0.47

OGND

0.47

曖昧さ回避

0.46

 anniversaire

0.46

árbol

0.46

ThroughAttribute

0.45

Activations Density 0.016%

programming and software development-related terms and function calls

code comparators and identifiers

No Comments

No Known Activations