INDEX

Explanations

instances of vulnerability associated with different subjects or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

Fprintf

-0.48

scription

-0.46

picas

-0.46

tät

-0.46

)$/,

-0.46

olusi

-0.44

tiba

-0.43

fiques

-0.43

korn

-0.42

oray

-0.41

POSITIVE LOGITS

 vulnerability

1.78

 vulnerable

1.74

 Vulnerable

1.54

 vulné

1.52

 weakness

1.48

 vulnerabilities

1.46

 Vulnerability

1.44

 vulner

1.37

vul

1.35

 vulnerables

1.34

Activations Density 0.273%

instances of vulnerability associated with different subjects or contexts

No Comments

No Known Activations