INDEX

Explanations

references to historical wars and conflicts, particularly related to the Civil War and colonialism

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

bin

-0.16

zon

-0.14

asse

-0.14

 suspect

-0.14

vester

-0.14

izer

-0.14

 Kraft

-0.13

 weapon

-0.13

 float

-0.13

 recept

-0.13

POSITIVE LOGITS

 Civil

0.47

 civil

0.46

Civil

0.42

war

0.38

War

0.36

 wars

0.35

civil

0.35

War

0.31

 Wars

0.30

WAR

0.28

Activations Density 0.118%

references to historical wars and conflicts, particularly related to the Civil War and colonialism

No Comments

No Known Activations