INDEX

Explanations

phrases indicating requests for assistance or communication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

StructEnd

-0.54

LEncoder

-0.52

 propOrder

-0.46

VYMaps

-0.45

RenderAtEndOf

-0.45

weightedMode

-0.45

rizona

-0.44

RTLR

-0.44

chige

-0.43

Aiheesta

-0.42

POSITIVE LOGITS

my

0.44

vaadin

0.40

让我

0.38

HandlerContext

0.38

讓我

0.38

Synchron

0.36

Ту

0.35

Aff

0.34

presse

0.34

Vac

0.34

Activations Density 0.695%

phrases indicating requests for assistance or communication

No Comments

No Known Activations