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

anagan

-0.15

 gpointer

-0.15

ungan

-0.15

ÑĥÐ½Ð´

-0.15

shima

-0.15

icipants

-0.14

 Shelter

-0.14

arkin

-0.14

anford

-0.14

rint

-0.14

POSITIVE LOGITS

 Amazon

0.32

Amazon

0.27

HasBeenSet

0.26

AWS

0.25

 amazon

0.23

aws

0.23

.AWS

0.21

amazon

0.20

.amazonaws

0.20

.aws

0.20

Activations Density 0.000%

No Known Activations

This feature has no known activations.