INDEX

Explanations

programming-related syntax and elements

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

password followed by of

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

Datuak

-0.85

 Infórmanos

-0.68

ConstraintMaker

-0.67

밌

-0.66

RTLD

-0.64

DoubleQuotes

-0.60

rungsseite

-0.59

 estekak

-0.59

 Italijani

-0.59

RenderAtEndOf

-0.59

POSITIVE LOGITS

piram

0.32

freopen

0.31

↵↵

0.31

This

0.30

Welcome

0.29

<strong>

0.28

 This

0.28

zuführen

0.28

hasMore

0.28

Activations Density 0.582%

programming-related syntax and elements

password followed by of

No Comments

No Known Activations