INDEX

Explanations

references to personal attributes, specifically names and identifiers

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

urlopen

-0.43

 CreateTagHelper

-0.43

 endregion

-0.42

abetes

-0.40

quelize

-0.40

 tiguan

-0.39

contentLoaded

-0.39

unteer

-0.39

Superview

-0.38

tagHelper

-0.38

POSITIVE LOGITS

 surname

0.77

Personendaten

0.76

 firstName

0.71

 name

0.70

 surnames

0.66

 names

0.66

firstName

0.66

getFirstName

0.64

FirstName

0.64

 firstname

0.64

Activations Density 0.090%

references to personal attributes, specifically names and identifiers

No Comments

No Known Activations