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

ega

-0.18

Ã±

-0.17

rlen

-0.16

Ä©

-0.16

 Desired

-0.15

 iken

-0.14

å½¼

-0.14

iferay

-0.14

unga

-0.14

iani

-0.13

POSITIVE LOGITS

Mu

0.18

weit

0.17

buz

0.15

tuy

0.15

Locator

0.15

yiy

0.14

 relying

0.14

 countless

0.14

ãĤ·ãĥ¥

0.14

ä¸

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.