INDEX

Explanations

references to Nazi Germany and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

onaut

-0.16

ft

-0.15

imos

-0.15

 Pascal

-0.14

aab

-0.14

icense

-0.14

Ã¦kke

-0.14

ayet

-0.14

ÄĽt

-0.14

utenberg

-0.14

POSITIVE LOGITS

Hit

0.47

 Hitler

0.43

Hit

0.42

.Hit

0.36

HIT

0.36

NS

0.35

 Nazi

0.33

hit

0.32

Naz

0.32

SS

0.32

Activations Density 0.110%

references to Nazi Germany and related concepts

No Comments

No Known Activations