INDEX

Explanations

proper nouns and noteworthy names in various contexts

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Jîn

-0.54

 GenerationType

-0.54

dsk

-0.48

ffindor

-0.48

tvguidetime

-0.48

ennan

-0.47

lusconi

-0.47

argate

-0.46

nian

-0.46

woofer

-0.45

POSITIVE LOGITS

 AllAfrica

0.50

SaveVideo

0.50

 Wikiseite

0.47

 meurt

0.46

曖昧さ回避

0.44

NewUrlParser

0.43

 jScrollPane

0.42

 joaat

0.42

 disambiguazione

0.42

 évent

0.41

Activations Density 0.857%

proper nouns and noteworthy names in various contexts

No Comments

No Known Activations