INDEX

Explanations

names and proper nouns, particularly related to individuals and their affiliations or contributions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

-0.47

Portály

-0.47

letek

-0.46

 recht

-0.44

ことができました

-0.43

 resa

-0.43

 shell

-0.43

wami

-0.42

 scale

-0.42

degard

-0.42

POSITIVE LOGITS

 تضيفلها

1.00

$_"

0.80

DockStyle

0.79

 estekak

0.76

ंदीखरीदारी

0.75

 defaultstate

0.74

"])

0.73

astify

0.72

dafx

0.71

 دیکھیے

0.69

Activations Density 0.333%

names and proper nouns, particularly related to individuals and their affiliations or contributions

No Comments

No Known Activations