INDEX

Explanations

repeated phrases indicating universality or generality

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

AEL

-0.48

 Ismael

-0.43

BOC

-0.42

ISC

-0.41

kräf

-0.41

 Alderman

-0.40

 Jiao

-0.40

 Slug

-0.39

 Brunch

-0.39

sena

-0.37

POSITIVE LOGITS

every

1.41

 every

1.34

 Every

1.30

Every

1.29

 EVERY

1.25

EVERY

1.16

Ogni

0.98

 Ogni

0.87

 ogni

0.85

ทุก

0.84

Activations Density 0.107%

repeated phrases indicating universality or generality

No Comments

No Known Activations