INDEX

Explanations

phrases related to political influence and bias

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

circle

-0.16

 Circle

-0.16

 circle

-0.15

ouro

-0.15

Circle

-0.15

 circles

-0.14

_circle

-0.14

veillance

-0.14

igid

-0.14

hone

-0.14

POSITIVE LOGITS

 neutrality

0.28

 political

0.24

 Political

0.24

 partisan

0.23

political

0.22

 politic

0.20

Political

0.20

 neutral

0.18

 pros

0.17

 overt

0.17

Activations Density 0.100%

phrases related to political influence and bias

No Comments

No Known Activations