INDEX

Explanations

references to physical harm or damage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

lify

-0.17

GI

-0.17

enty

-0.15

shire

-0.15

midi

-0.15

?(:

-0.15

mie

-0.15

Ø®Ø§ÙĨÙĩ

-0.15

NÄ°

-0.14

anness

-0.14

POSITIVE LOGITS

 done

0.29

å®³

0.22

 Done

0.22

done

0.21

 DONE

0.21

 sustained

0.21

Done

0.20

-done

0.18

aceutical

0.18

(done

0.17

Activations Density 0.060%

references to physical harm or damage

No Comments

No Known Activations