INDEX

Explanations

calls to action and promotional language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

ufficio

-0.46

sätzlich

-0.45

følgelig

-0.45

omalainen

-0.42

oughlin

-0.42

 ogóle

-0.41

tamientos

-0.40

ėse

-0.40

Kau

-0.39

vrouw

-0.38

POSITIVE LOGITS

 tartalomajánló

0.80

 متعلقه

0.78

PreferredItem

0.78

KURZBESCHREIBUNG

0.75

 hurry

0.71

BufferException

0.68

anglès

0.66

 مشين

0.65

enterOuterAlt

0.62

 Hurry

0.62

Activations Density 0.083%

calls to action and promotional language

No Comments

No Known Activations