INDEX

Explanations

references to the Audi car brand and its specific models

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

untime

-0.17

/MPL

-0.15

 Lyons

-0.15

äºķ

-0.15

.fx

-0.15

atican

-0.14

usz

-0.14

 Katrina

-0.14

Clone

-0.14

ij

-0.14

POSITIVE LOGITS

 Audi

0.41

Aud

0.33

qu

0.32

audi

0.31

 audi

0.31

MMI

0.31

RS

0.30

AUD

0.29

AUD

0.28

Qu

0.27

Activations Density 0.005%

references to the Audi car brand and its specific models

No Comments

No Known Activations