INDEX

Explanations

instances of the word "arrest."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

chieht

-0.39

writeField

-0.39

Slf

-0.37

出版年

-0.37

#![

-0.37

 crime

-0.36

yszer

-0.36

醐

-0.35

 schlä

-0.34

tagext

-0.34

POSITIVE LOGITS

 arrest

1.41

 arrested

1.33

 arrests

1.28

 Arrest

1.24

Arrest

1.17

arrest

1.11

 Arrested

1.02

 arresting

1.02

逮捕

0.79

 disappointed

0.76

Activations Density 0.075%

instances of the word "arrest."

No Comments

No Known Activations