INDEX

Explanations

references to aesthetics and design elements in spaces

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

oria

-0.18

ollar

-0.16

orer

-0.15

 enumerator

-0.14

 legisl

-0.14

dto

-0.14

uckles

-0.14

ë³¸

-0.14

LEGRO

-0.14

.xaml

-0.14

POSITIVE LOGITS

 exposed

0.22

 industrial

0.20

 Edison

0.20

 reclaimed

0.19

 booths

0.19

 giant

0.18

 hanging

0.18

 distressed

0.18

 dangling

0.17

 Booth

0.17

Activations Density 0.126%

references to aesthetics and design elements in spaces

No Comments

No Known Activations