INDEX

Explanations

references to royalty and kingship

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

fromLTRB

-0.47

 dataSnapshot

-0.43

 Leverkusen

-0.40

 Needham

-0.40

 flung

-0.40

anasia

-0.39

hradun

-0.39

polator

-0.38

ucchini

-0.38

badger

-0.38

POSITIVE LOGITS

 King

0.78

 Kings

0.75

 KING

0.74

 king

0.74

 KINGS

0.73

King

0.73

 Kral

0.73

 prince

0.72

 Queen

0.70

Kings

0.70

Activations Density 0.248%

references to royalty and kingship

No Comments

No Known Activations