INDEX

Explanations

terms related to usernames and passwords

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 tepi

-0.42

WebServlet

-0.39

,:);

-0.39

あえず

-0.39

PageRoute

-0.38

]));

-0.37

])[

-0.36

iyle

-0.36

Rohy

-0.36

BeginContext

-0.36

POSITIVE LOGITS

 password

3.75

password

3.38

 Password

3.36

Password

3.28

 PASSWORD

2.86

 passwords

2.75

PASSWORD

2.75

パスワード

2.28

密码

2.27

getPassword

2.23

Activations Density 0.049%

terms related to usernames and passwords

No Comments

No Known Activations