INDEX

Explanations

references to white supremacy and racially charged rhetoric

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

ll

-0.14

 Bert

-0.13

å¢ĥ

-0.13

Ð»Ð°Ð²

-0.13

IRST

-0.13

 Socialist

-0.13

cuador

-0.13

 Schultz

-0.13

 progressive

-0.13

 vers

-0.13

POSITIVE LOGITS

 Whites

0.28

jew

0.26

WN

0.25

(((

0.24

Jew

0.23

 racial

0.23

 racially

0.21

racial

0.21

 White

0.20

White

0.20

Activations Density 0.025%

references to white supremacy and racially charged rhetoric

No Comments

No Known Activations

references to white supremacy and racially charged rhetoric

No Comments

No Known Activations