INDEX

Explanations

phrases related to political processes and transparency

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

stry

-0.14

ä¸ĢåĮº

-0.14

 INTERRU

-0.14

hma

-0.14

 consulate

-0.14

vat

-0.14

anten

-0.14

addin

-0.14

antage

-0.13

rzy

-0.13

POSITIVE LOGITS

 ethics

0.51

 Ethics

0.47

Eth

0.39

 ethical

0.39

 ethic

0.38

eth

0.38

Eth

0.36

ETH

0.35

ETH

0.34

eth

0.33

Activations Density 0.058%

phrases related to political processes and transparency

No Comments

No Known Activations