INDEX

Explanations

disclaimers and information regarding product evaluation and user responsibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

vrier

-0.15

lete

-0.15

 CrossRef

-0.13

ivate

-0.13

aptive

-0.13

Ð¼ÐµÑĤÑĮ

-0.12

Keywords

-0.12

ethylene

-0.12

assa

-0.12

 Specify

-0.12

POSITIVE LOGITS

 contents

0.30

 content

0.30

 contained

0.27

contents

0.26

 conten

0.26

content

0.24

contained

0.24

 contenido

0.23

 Content

0.23

.content

0.22

Activations Density 0.115%

disclaimers and information regarding product evaluation and user responsibility

No Comments

No Known Activations