INDEX

Explanations

the concept of evil in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

luvÃŃ

-0.15

INAL

-0.15

posable

-0.15

 Pett

-0.15

esh

-0.14

åĭĻ

-0.14

-issue

-0.14

inal

-0.14

xic

-0.14

 åĵģ

-0.14

POSITIVE LOGITS

ayar

0.16

hardt

0.15

tram

0.15

ously

0.15

 deeds

0.14

 deed

0.14

reu

0.14

ÛĮÙĩ

0.14

MPS

0.14

lesi

0.14

Activations Density 0.011%

the concept of evil in various contexts

No Comments

No Known Activations