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

StackSize

-0.17

çıŃ

-0.15

.Localization

-0.15

illard

-0.15

alink

-0.14

ãĥ©ãĥĥãĤ¯

-0.14

hend

-0.14

ÌĨ

-0.14

ãĥŃãĥ¼

-0.14

ojÃŃ

-0.14

POSITIVE LOGITS

 readers

0.27

 reader

0.25

 index

0.23

 pages

0.22

 Readers

0.21

Deletes

0.21

 Vacuum

0.21

 Index

0.20

 Pages

0.20

 indexes

0.20

Activations Density 0.000%

No Known Activations

This feature has no known activations.