INDEX

Explanations

phrases related to perspectives or views within discussing complex concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

UrlEncoded

-0.48

getMenuInflater

-0.43

 Evid

-0.41

zoned

-0.40

ContentAsync

-0.40

ChildScrollView

-0.39

BrowserModule

-0.39

二是

-0.39

acuzzi

-0.39

UnusedPrivate

-0.39

POSITIVE LOGITS

how

0.81

 what

0.76

 cómo

0.74

why

0.73

 jspb

0.66

 Ganzen

0.64

 ceea

0.64

 neler

0.63

 quienes

0.63

hva

0.61

Activations Density 0.733%

phrases related to perspectives or views within discussing complex concepts

No Comments

No Known Activations