INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

edl

-0.15

Ð°Ð»ÑĮ

-0.15

ritt

-0.15

 dist

-0.14

ythe

-0.14

ickle

-0.14

 punch

-0.14

iban

-0.14

aÃ±

-0.13

 Piet

-0.13

POSITIVE LOGITS

hua

0.15

ascal

0.14

à¥ģà¤ģ

0.14

çħ¤

0.14

uzu

0.14

è±

0.14

uniform

0.13

uper

0.13

iser

0.13

igue

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.