INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

sdale

-0.21

iness

-0.17

enheim

-0.15

rophy

-0.15

apa

-0.15

ened

-0.15

çİĩ

-0.15

ably

-0.14

igy

-0.14

POSITIVE LOGITS

ton

0.21

TON

0.18

tons

0.16

 thá»Ŀ

0.15

 bells

0.15

 grounds

0.15

odge

0.15

toi

0.15

Zuk

0.15

 Goldman

0.14

Activations Density 0.017%

No Comments

No Known Activations