INDEX

Explanations

variations and adaptations of concepts or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

dej

-0.16

 architekt

-0.15

SWG

-0.14

 Simulator

-0.14

 slik

-0.14

/gtest

-0.14

 TestUtils

-0.13

 Mori

-0.13

enti

-0.13

æ¥

-0.13

POSITIVE LOGITS

 version

0.32

 versions

0.30

/version

0.27

version

0.26

of

0.25

 thereof

0.24

-version

0.23

 Versions

0.23

 Version

0.22

Versions

0.21

Activations Density 0.105%

variations and adaptations of concepts or items

No Comments

No Known Activations