INDEX

Explanations

terms and concepts related to equity and equality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

enburg

-0.16

scratch

-0.16

 billig

-0.15

usan

-0.15

tings

-0.15

emez

-0.15

ARGIN

-0.14

ãĥ©ãĥĥãĤ¯

-0.14

casts

-0.14

emies

-0.14

POSITIVE LOGITS

atorial

0.24

Equ

0.21

ipping

0.20

ipped

0.20

ilibrium

0.20

ipment

0.19

equ

0.19

inox

0.17

ator

0.17

anim

0.16

Activations Density 0.025%

terms and concepts related to equity and equality

No Comments

No Known Activations