INDEX

Explanations

phrases indicative of confusion or misunderstanding in a scenario

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

 However

-0.35

But

-0.34

Yet

-0.30

 Siempre

-0.30

 особо

-0.29

but

-0.29

AUTH

-0.28

✭✭

-0.28

 Always

-0.27

 Budaya

-0.27

POSITIVE LOGITS

MigrationBuilder

0.83

thâu

0.76

 CreateTagHelper

0.71

 ModelExpression

0.65

UnsafeEnabled

0.65

featureID

0.65

 StatelessWidget

0.65

 ProtoMessage

0.64

 ternyata

0.63

AntiForgeryToken

0.63

Activations Density 0.334%

phrases indicative of confusion or misunderstanding in a scenario

No Comments

No Known Activations