INDEX

Explanations

topics related to safety and regulatory compliance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Anti

-0.14

alian

-0.14

Dav

-0.14

ä¸Ī

-0.14

egin

-0.13

lub

-0.13

Edu

-0.13

 save

-0.13

 Planet

-0.13

pay

-0.13

POSITIVE LOGITS

ampions

0.17

classpath

0.16

adu

0.15

uisse

0.15

agos

0.14

AMED

0.14

IENT

0.14

ÐµÐºÑĤÐ¾ÑĢ

0.14

.onView

0.14

æĺĩ

0.14

Activations Density 0.129%

topics related to safety and regulatory compliance

No Comments

No Known Activations