INDEX

Explanations

references to hate crimes and violence against marginalized communities

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

abad

-0.19

 Ross

-0.14

ãĤ¹ãĤ³

-0.14

moc

-0.14

Ø§Ø¯Ùħ

-0.13

Anc

-0.13

-Smith

-0.13

ROSS

-0.13

 Cannon

-0.13

eneg

-0.13

POSITIVE LOGITS

toi

0.17

ä»

0.16

æ®

0.15

 Journalism

0.15

 hate

0.15

Acts

0.15

 incel

0.14

 Ð°ÐºÑĤÐ¸

0.14

tainment

0.14

finity

0.14

Activations Density 0.061%

references to hate crimes and violence against marginalized communities

No Comments

No Known Activations