INDEX

Explanations

bullet point lists or enumerations of tips and advice

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

发表于

-0.40

 utafitiHapana

-0.37

 Taktlose

-0.33

 betweenstory

-0.33

/\.

-0.32

تقاوى

-0.31

nictwa

-0.31

GEBURTSDATUM

-0.31

esModule

-0.29

 tartalomajánló

-0.28

POSITIVE LOGITS

awtextra

0.59

 snippetHide

0.55

tableFuture

0.54

RegistryLite

0.52

EnumMember

0.51

 للاسماء

0.51

DockStyle

0.50

เอง

0.50

tahui

0.50

OrBuilder

0.50

Activations Density 0.017%

bullet point lists or enumerations of tips and advice

No Comments

No Known Activations