INDEX

Explanations

terms associated with regulatory compliance and corruption issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

inux

-0.18

Äįas

-0.14

amon

-0.14

ëĨį

-0.14

idel

-0.13

 Filename

-0.13

ollections

-0.13

 Conserv

-0.13

GenerationStrategy

-0.13

Neighbors

-0.13

POSITIVE LOGITS

 bribery

0.44

Bri

0.43

bri

0.39

 brib

0.38

 corruption

0.31

 Corruption

0.31

 corrupt

0.29

 Foreign

0.28

 anti

0.26

 foreign

0.26

Activations Density 0.013%

terms associated with regulatory compliance and corruption issues

No Comments

No Known Activations