INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Recru

-0.15

ynn

-0.15

optera

-0.14

 EntryPoint

-0.14

roe

-0.14

deniz

-0.14

NotificationCenter

-0.14

Ø§Ø«

-0.13

kum

-0.13

arkan

-0.13

POSITIVE LOGITS

 risk

0.28

 internal

0.27

 Internal

0.25

 Risk

0.24

 audit

0.24

audit

0.23

aud

0.22

 Audit

0.21

risk

0.21

 INTERNAL

0.20

Activations Density 0.000%

No Known Activations

This feature has no known activations.