INDEX

Explanations

No Explanations Found

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

undra

-0.18

 vlak

-0.15

intr

-0.15

alc

-0.15

oran

-0.14

vap

-0.14

vell

-0.14

.toJSON

-0.14

enu

-0.14

 realpath

-0.14

POSITIVE LOGITS

0.15

 ëĺĲ

0.15

xed

0.14

ampie

0.14

~=

0.14

~(

0.14

oga

0.14

osten

0.13

kee

0.13

.gg

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.