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

 armor

-0.17

 probe

-0.15

ige

-0.15

 Keto

-0.15

plate

-0.15

armor

-0.15

 probes

-0.14

³

-0.14

 armored

-0.14

 harbor

-0.14

POSITIVE LOGITS

 Iran

0.29

 Iranian

0.28

 Tehran

0.26

 Iranians

0.25

iran

0.24

Iran

0.24

 gays

0.22

gay

0.22

 western

0.20

 Ä°ran

0.20

Activations Density 0.000%

No Known Activations

This feature has no known activations.