INDEX

Explanations

references to political debates and commentaries

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

Ìģ

-0.15

.spotify

-0.15

ÌĢ

-0.14

esda

-0.14

apiro

-0.14

ÑĢÐ¾Ð²Ð¸Ñĩ

-0.14

okrat

-0.13

.github

-0.13

Integer

-0.13

POSITIVE LOGITS

 examiner

0.25

 Examiner

0.22

 Chronicles

0.20

 Around

0.17

ReadWrite

0.17

 Suite

0.17

 Inside

0.16

 Behind

0.16

 Lets

0.16

 About

0.16

Activations Density 0.238%

references to political debates and commentaries

No Comments

No Known Activations