INDEX

Explanations

significant events or disclosures that reveal hidden information or secrets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UILDER

-0.14

askets

-0.14

iche

-0.13

æİª

-0.13

QR

-0.13

ØµØ§Øª

-0.13

webdriver

-0.13

essim

-0.13

_LS

-0.12

oto

-0.12

POSITIVE LOGITS

 reveal

0.50

 revealed

0.48

 revealing

0.45

reve

0.45

 expose

0.45

 exposed

0.44

 reveals

0.44

reveal

0.44

 disclosure

0.44

 exposure

0.42

Activations Density 0.484%

significant events or disclosures that reveal hidden information or secrets

No Comments

No Known Activations