INDEX

Explanations

sections and structure in a document, likely related to web or document formatting

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

web page structure HTML or Javascript

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

WriteTagHelper

-0.54

enumi

-0.52

########.

-0.49

racked

-0.47

+};

-0.44

togen

-0.44

 pocket

-0.42

 cosmic

-0.41

Datuak

-0.41

ympto

-0.41

POSITIVE LOGITS

styleUrls

0.46

 ProtoMessage

0.42

adaptiveStyles

0.38

invokeLater

0.38

 betweenstory

0.38

setVerticalGroup

0.38

 assignment

0.37

endsection

0.36

 Dedication

0.36

Personensuche

0.35

Activations Density 0.003%

sections and structure in a document, likely related to web or document formatting

web page structure HTML or Javascript

No Comments

No Known Activations