INDEX

Explanations

language related to warfare, weaponry, and combat scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

xit

-0.16

ertz

-0.15

unge

-0.15

ä¸īçº§

-0.15

orida

-0.14

ossal

-0.14

sealed

-0.14

imension

-0.13

iyah

-0.13

 Knife

-0.13

POSITIVE LOGITS

 fire

0.36

-fire

0.30

fire

0.29

 fired

0.28

 firing

0.27

 shells

0.26

Fire

0.26

 Fire

0.26

 gunfire

0.25

çģ«

0.25

Activations Density 0.053%

language related to warfare, weaponry, and combat scenarios

No Comments

No Known Activations