INDEX

Explanations

references to popular entertainment figures and media events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 transférez

-0.47

 trám

-0.45

StoryboardSegue

-0.43

RenderAtEndOf

-0.42

dAtA

-0.41

новништво

-0.41

 TestBed

-0.40

NameInMap

-0.40

 humain

-0.38

 épis

-0.38

POSITIVE LOGITS

 EconPapers

0.73

󠁮

0.45

wards

0.43

 affic

0.42

glTexCoord

0.42

Filmografia

0.41

 contestant

0.40

kward

0.40

 eslint

0.40

 lawmaker

0.40

Activations Density 0.180%

references to popular entertainment figures and media events

No Comments

No Known Activations