INDEX

Explanations

references to men and male-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 FactoryBot

-0.17

Sad

-0.17

ernote

-0.16

imler

-0.15

atik

-0.15

piring

-0.15

AML

-0.14

lassen

-0.14

Ley

-0.14

rÃ¡t

-0.14

POSITIVE LOGITS

opause

0.32

endez

0.26

orca

0.25

ager

0.23

ubar

0.23

cken

0.23

udo

0.23

elik

0.22

ninger

0.22

acing

0.22

Activations Density 0.014%

references to men and male-related terms

No Comments

No Known Activations