INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 Kiss

-0.16

onaut

-0.16

 Sark

-0.15

 Cameron

-0.15

 McKay

-0.14

braco

-0.14

æ¼Ķ

-0.14

reset

-0.13

 MacDonald

-0.13

à¥Ģà¤ļ

-0.13

POSITIVE LOGITS

 Bart

0.38

 bart

0.30

bart

0.26

 Barr

0.21

 Bars

0.19

 bartender

0.19

 barr

0.19

 Barton

0.18

 Barth

0.18

associ

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.