INDEX

Explanations

proper nouns related to individuals and their attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ButtonItem

-0.20

ayi

-0.19

jspb

-0.16

ington

-0.16

ewan

-0.16

neau

-0.14

bane

-0.14

uke

-0.14

zier

-0.14

inq

-0.13

POSITIVE LOGITS

of

0.31

 cá»§a

0.24

of

0.22

 mili

0.17

ustil

0.16

 Nicholson

0.15

 hometown

0.15

uner

0.15

á»§a

0.15

icky

0.14

Activations Density 0.054%

proper nouns related to individuals and their attributes

No Comments

No Known Activations