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

``

-0.19

veau

-0.15

Ð·Ð²

-0.14

 âĢŀ

-0.14

\:

-0.14

 PYTHON

-0.13

-âĢĲ

-0.13

 DOWNLOAD

-0.13

ì¶ľ

-0.12

hev

-0.12

POSITIVE LOGITS

 essay

0.24

ge

0.23

 Mapping

0.23

 pain

0.22

 cart

0.22

 maps

0.22

/maps

0.21

 essays

0.21

 mapping

0.21

 Maps

0.20

Activations Density 0.000%

No Known Activations

This feature has no known activations.