INDEX

Explanations

warnings and safety concerns regarding children and small objects

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

Guy

-0.16

èĤ©

-0.16

Leaks

-0.15

vir

-0.15

 Shack

-0.15

ilon

-0.14

vasion

-0.14

Ã±ana

-0.14

reon

-0.14

emand

-0.14

POSITIVE LOGITS

 dangerous

0.22

 unsafe

0.21

Unsafe

0.21

 danger

0.20

unsafe

0.19

 children

0.19

 safety

0.18

 Dangerous

0.18

 safer

0.17

 dangers

0.17

Activations Density 0.050%

warnings and safety concerns regarding children and small objects

No Comments

No Known Activations