INDEX

Explanations

references to political figures, particularly connections to controversial associations

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

rale

-0.18

utsch

-0.17

exo

-0.16

unsch

-0.16

sville

-0.15

 Hobby

-0.15

querque

-0.14

 Ivanka

-0.14

gle

-0.14

leigh

-0.14

POSITIVE LOGITS

 McCain

0.29

McC

0.22

Sen

0.19

mcc

0.19

 senator

0.18

POW

0.18

 Maver

0.18

Eag

0.17

McL

0.17

Tro

0.16

Activations Density 0.035%

references to political figures, particularly connections to controversial associations

No Comments

No Known Activations