INDEX

Explanations

references to geopolitical conflicts and military actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

 THROW

-0.16

Pik

-0.16

zik

-0.16

Abe

-0.16

Xi

-0.15

andle

-0.15

çİĦ

-0.15

atal

-0.15

kir

-0.15

-0.14

POSITIVE LOGITS

 Iraqi

0.35

 Saddam

0.35

 Iraq

0.34

 Sadd

0.31

 Baghdad

0.31

 Coalition

0.31

Iraq

0.30

 coalition

0.29

 Ø§ÙĦØ¹Ø±Ø§ÙĤ

0.27

 Ø¹Ø±Ø§ÙĤ

0.25

Activations Density 0.025%

references to geopolitical conflicts and military actions

No Comments

No Known Activations