INDEX

Explanations

No Explanations Found

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

Ã©is

-0.15

 Sanat

-0.15

Doctrine

-0.15

 ðŁĴ

-0.14

âĿ

-0.14

onet

-0.14

Ø§Ø´

-0.14

ÅĽnie

-0.13

 ðŁĳ

-0.13

POSITIVE LOGITS

 Mass

0.24

Mass

0.23

 MASS

0.21

 sort

0.20

.mass

0.19

 Massive

0.18

 mass

0.18

 fucking

0.17

 Naples

0.17

 fiction

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.