INDEX

Explanations

terms related to power dynamics and political authority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

rades

-0.16

æĢĿ

-0.15

finity

-0.15

 Tavern

-0.14

uien

-0.14

_QUAL

-0.14

liers

-0.14

iqueta

-0.13

 vitam

-0.13

èģĮä¸ļ

-0.13

POSITIVE LOGITS

/power

0.33

 power

0.31

power

0.27

 control

0.27

(power

0.26

/control

0.25

 Power

0.24

 authority

0.23

Power

0.23

.Power

0.23

Activations Density 0.118%

terms related to power dynamics and political authority

No Comments

No Known Activations