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

ÙĪÙĨØ¯

-0.16

 ÐĵÐ¾Ð»

-0.15

iks

-0.15

toy

-0.14

avl

-0.14

Rol

-0.14

 Jord

-0.14

Doc

-0.13

å¦»

-0.13

 Holl

-0.13

POSITIVE LOGITS

Kou

0.52

kou

0.43

kou

0.37

Hou

0.32

ou

0.30

Tou

0.29

Ou

0.28

OU

0.28

Rou

0.26

cou

0.26

Activations Density 0.000%

No Known Activations

This feature has no known activations.