INDEX

Explanations

references to kings or royalty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hysema

-0.46

epam

-0.46

utsch

-0.43

CPE

-0.42

 Hewitt

-0.42

__':

-0.41

).]

-0.41



-0.41

($.

-0.41

Tämä

-0.40

POSITIVE LOGITS

 King

1.08

King

1.08

 KING

0.95

 king

0.94

 Kings

0.93

Kings

0.92

 Krá

0.88

 KINGS

0.88

 kings

0.83

 Kral

0.75

Activations Density 0.011%

references to kings or royalty

No Comments

No Known Activations