INDEX

Explanations

repeated mentions of the term "og"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

Спасылкі

-0.48

épendance

-0.47

 Aufenthal

-0.47

nalpot

-0.44

 aislada

-0.44

 Hildebrand

-0.43

 grueso

-0.43

kwds

-0.42

ApiOperation

-0.42

Portail

-0.42

POSITIVE LOGITS

og

0.56

floating

0.49

 proportional

0.49

 typelib

0.49

 floating

0.47

ance

0.47

 ModelExpression

0.46

 Floating

0.45

 StatelessWidget

0.45

Lak

0.45

Activations Density 0.438%

repeated mentions of the term "og"

No Comments

No Known Activations