INDEX

Explanations

phrases that indicate presence or involvement

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

 Grüsse

-0.51

acidade

-0.47

gdy

-0.47

 domenica

-0.46

 estreia

-0.42

 quaisquer

-0.41

mazioni

-0.40

 respecto

-0.40

 tatuagens

-0.40

เมื่อ

-0.39

POSITIVE LOGITS

 Houſe

0.61

addGap

0.61

 Majefty

0.57

awaiter

0.54

 useContext

0.54

 involved

0.52

 useAuth

0.52

ंदीखरीदारी

0.51

メルカリ

0.51

addPreferredGap

0.50

Activations Density 0.234%

phrases that indicate presence or involvement

No Comments

No Known Activations