INDEX

Explanations

terms associated with crime and violence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

olet

-0.17

azer

-0.16

ahren

-0.16

ppard

-0.16

uish

-0.15

quisition

-0.15

Î»Î¿Î³

-0.15

allery

-0.15

Ð»ÐµÐº

-0.15

pany

-0.14

POSITIVE LOGITS

ç´

0.14

emen

0.14

åıĸ

0.14

AndView

0.14

igh

0.13

èĢĲ

0.13

hab

0.13

 Schul

0.13

.compiler

0.13

 synthetic

0.13

Activations Density 0.256%

terms associated with crime and violence

No Comments

No Known Activations