INDEX

Explanations

references to significant historical events and figures

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

disaster, conflict, or execution

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

出版年

-0.61

KURZBESCHREIBUNG

-0.52

Viitteet

-0.48

-------------</

-0.43



-0.43

Naissance

-0.42

 Salinity

-0.42

fiddle

-0.42

 ashamed

-0.41

 EconPapers

-0.41

POSITIVE LOGITS

djangoproject

0.40

 ویکی‌پدی

0.40

 behö

0.39

dafx

0.38

<%=

0.38

ględ

0.37

REQUIRES

0.36

MessageTagHelper

0.36

titleMargin

0.35

 MenuView

0.34

Activations Density 0.016%

references to significant historical events and figures

disaster, conflict, or execution

No Comments

No Known Activations