INDEX

Explanations

navigational prompts related to content browsing or management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

Personendaten

-0.51

TagMode

-0.47

 defaultstate

-0.46

ftagPool

-0.45

ThroughAttribute

-0.43

findpost

-0.42

 ſte

-0.41

HasAnnotation

-0.41

 tranſ

-0.40

UrlResolution

-0.40

POSITIVE LOGITS

cln

0.55

ValueStyle

0.53

multer

0.50

 TAMBÉM

0.47

appen

0.47

transQ

0.45

traction

0.45

 strictest

0.44

GGI

0.44

portation

0.44

Activations Density 0.122%

navigational prompts related to content browsing or management

No Comments

No Known Activations