INDEX

Explanations

references to permission or authorization

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

 académica

-0.37

 Universitaria

-0.35

 judiciales

-0.34

 Occidente

-0.34

 Jurí

-0.34

 pacar

-0.33

 Naciones

-0.33

 universitaria

-0.33

 companhia

-0.33

 Justicia

-0.32

POSITIVE LOGITS

 permitted

2.03

permitted

1.63

 prohibited

1.48

 allowed

1.47

 Permit

1.38

 permit

1.37

 permitting

1.36

permit

1.34

 Allowed

1.32

 permits

1.29

Activations Density 0.171%

references to permission or authorization

No Comments

No Known Activations