INDEX

Explanations

concepts related to electoral integrity and fairness in the political process

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

qli

-0.17

endale

-0.15

fil

-0.15

.ribbon

-0.14

ÙĤÙĪÙĦ

-0.14

akah

-0.14

 chÃ¡y

-0.13

Ple

-0.13

_usb

-0.13

asy

-0.13

POSITIVE LOGITS

 fair

0.34

 Fair

0.28

fair

0.27

 fairness

0.26

 faire

0.26

Fair

0.26

 democratic

0.24

 democracy

0.22

 free

0.21

 transparency

0.21

Activations Density 0.058%

concepts related to electoral integrity and fairness in the political process

No Comments

No Known Activations