INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 nuest

-0.14

umann

-0.14

isiyle

-0.14

abl

-0.14

ju

-0.13

REDENTIAL

-0.13

 Pron

-0.13

mÃ®

-0.13

ToStr

-0.13

.Localization

-0.13

POSITIVE LOGITS

antha

0.19

 silence

0.15

ÑĥÐ¶Ðµ

0.14

acoes

0.13

mbH

0.13

Refs

0.13

esin

0.13

upply

0.13

uyo

0.13

 Math

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.