INDEX

Explanations

references to the origin and quality of products, particularly highlighting the "Made in China" and "Made in Japan" labels

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

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

olet

-0.15

 åŃ

-0.15

idak

-0.14

anax

-0.14

alars

-0.13

.scalablytyped

-0.13

enheim

-0.13

awk

-0.13

ookies

-0.13

aket

-0.13

POSITIVE LOGITS

 Made

0.80

Made

0.78

 made

0.75

made

0.72

 MADE

0.68

-made

0.60

emade

0.47

 manufactured

0.43

 Manufact

0.35

 produced

0.34

Activations Density 0.503%

references to the origin and quality of products, particularly highlighting the "Made in China" and "Made in Japan" labels

No Comments

No Known Activations