INDEX

Explanations

instances of abbreviations or acronyms related to titles and names

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

lington

-0.16

zar

-0.16

acin

-0.15

geh

-0.15

ŀ

-0.14

allee

-0.14

nell

-0.14

ickers

-0.14

combe

-0.14

 poss

-0.14

POSITIVE LOGITS

ublice

0.32

ub

0.20

ublic

0.20

Ã¼b

0.19

ubi

0.19

hub

0.19

reek

0.18

uber

0.18

 hubs

0.18

agina

0.18

Activations Density 0.006%

instances of abbreviations or acronyms related to titles and names

No Comments

No Known Activations

instances of abbreviations or acronyms related to titles and names

No Comments

No Known Activations