INDEX

Explanations

references to different geographical or topical areas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

esse

-0.17

weed

-0.16

ess

-0.15

éĴ

-0.14

air

-0.14

 toJSON

-0.14

mith

-0.14

à¸Ľà¸£à¸°à¸Īà¸³

-0.14

oga

-0.14

uya

-0.14

POSITIVE LOGITS

OfWork

0.18

/Area

0.18

OfBirth

0.17

åŁŁ

0.16

.Areas

0.15

olina

0.15

abouts

0.15

deki

0.15

ahat

0.15

Activations Density 0.068%

references to different geographical or topical areas

No Comments

No Known Activations