INDEX

Explanations

explicit sexual content and language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 sexuality

-0.19

 sexually

-0.16

 sexual

-0.16

å²³

-0.14

eteria

-0.14

Ã¡Å¡

-0.14

 Sexy

-0.14

ifax

-0.14

æ±Ĺ

-0.14

POSITIVE LOGITS

toy

0.28

 toys

0.26

 vibr

0.25

 vibration

0.24

 vibrations

0.24

 Toys

0.22

toy

0.22

Toy

0.22

 Vibr

0.22

vag

0.20

Activations Density 0.007%

explicit sexual content and language

No Comments

No Known Activations