INDEX

Explanations

punctuation marks, specifically parentheses

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

user turn start

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

ніципалі

-0.46

yarnpkg

-0.42

 Biôgrafia

-0.39

 exigir

-0.38

 asesinado

-0.38

spyOn

-0.37

 Formato

-0.37

össä

-0.37

 discul

-0.37

quelize

-0.36

POSITIVE LOGITS

RegistryLite

0.64

 title

0.52

content

0.50

 content

0.49

title

0.49

 Content

0.47

centralwidget

0.44

CppMethod

0.44

 Title

0.43

Content

0.42

Activations Density 0.000%

punctuation marks, specifically parentheses

user turn start

No Comments

No Known Activations