INDEX

Explanations

references to personal information or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

XX

-0.39

request

-0.39

ManyToMany

-0.38

None

-0.36

zu

-0.35

–

-0.34

athon

-0.33

XIX

-0.33

তি

-0.33

||

-0.33

POSITIVE LOGITS

 Personal

1.24

Personal

1.24

 personal

1.23

personal

1.22

 PERSONAL

1.15

PERSONAL

1.11

 personales

1.02

 persönliche

0.88

ersonal

0.85

 persoonlijke

0.82

Activations Density 0.210%

references to personal information or characteristics

No Comments

No Known Activations