INDEX

Explanations

references to crime and law enforcement efforts

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

QLD

-0.15

iaux

-0.15

.debian

-0.15

æĻ

-0.15

.misc

-0.15

qv

-0.14

 Queensland

-0.14

Nes

-0.14

alus

-0.14

iales

-0.14

POSITIVE LOGITS

 Detroit

0.62

Detroit

0.54

DET

0.47

DET

0.45

etroit

0.44

det

0.43

 Michigan

0.42

 Pistons

0.38

Det

0.37

 Motor

0.36

Activations Density 0.061%

references to crime and law enforcement efforts

No Comments

No Known Activations

references to crime and law enforcement efforts

No Comments

No Known Activations