INDEX

Explanations

proper nouns, particularly names associated with individuals and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

PreExecute

-0.54

Personendaten

-0.41

 Taktlose

-0.40

ỡng

-0.40

diyesi

-0.38

 вікі

-0.38

ladiator

-0.38

 validamos

-0.38

jooq

-0.38

colhead

-0.37

POSITIVE LOGITS

berg

1.33

BERG

1.29

ber

1.14

BER

1.10

берг

1.07

 Berry

1.06

bergs

1.04

 berry

1.02

berry

0.97

burg

0.96

Activations Density 1.136%

proper nouns, particularly names associated with individuals and places

No Comments

No Known Activations

proper nouns, particularly names associated with individuals and places

No Comments

No Known Activations