INDEX

Explanations

indicators of substantial changes in context or emphasis

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 pinulongan

-0.91

ScopeManager

-0.81

+#+#

-0.79

HtmlAttribute

-0.68

Personendaten

-0.67

ValueStyle

-0.67

ContentAsync

-0.66

 الحره

-0.66

 nahilalakip

-0.66

contentLoaded

-0.65

POSITIVE LOGITS

 Polres

0.40

 beliau

0.40

mación

0.34

 Marín

0.33

 autorité

0.32

dė

0.30

 dignité

0.30

 titolata

0.29

 asmen

0.28

cción

0.28

Activations Density 1.363%

indicators of substantial changes in context or emphasis

No Comments

No Known Activations