INDEX

Explanations

references to being told or informed about something, particularly in a directive or authoritative manner

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 climbers

-0.40

 intent

-0.40

 vides

-0.40

 Volo

-0.40

 ambient

-0.39

 rast

-0.39

 cosmetics

-0.38

ivant

-0.38

 climber

-0.37

 tull

-0.37

POSITIVE LOGITS

onOptions

0.51

 RemoteException

0.50

 recie

0.47

IBOutlet

0.47

RoutingModule

0.46

лтемелер

0.46

SequentialGroup

0.45

 receive

0.44

 oleh

0.44

enumii

0.44

Activations Density 0.736%

references to being told or informed about something, particularly in a directive or authoritative manner

No Comments

No Known Activations