INDEX

Explanations

numerical data and specific measurements related to publications

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

atch

-0.16

awk

-0.15

 Playlist

-0.15

tube

-0.14

Ã¼mÃ¼ÅŁ

-0.14

ãĤ¦ãĥĪ

-0.14

éļ

-0.14

à¸²à¸§

-0.14

 ì½

-0.14

strncmp

-0.13

POSITIVE LOGITS

 sides

0.27

 pages

0.26

pp

0.26

pages

0.25

 leaves

0.24

pg

0.24

pp

0.23

 Seiten

0.21

 sider

0.21

 ìª½

0.21

Activations Density 0.024%

numerical data and specific measurements related to publications

No Comments

No Known Activations