INDEX

Explanations

phrases related to requesting assistance or information

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

untas

-0.17

 Ð¾Ð±Ð¾ÑĢ

-0.15

 setC

-0.15

fea

-0.15

yne

-0.15

defense

-0.14

nelle

-0.14

Äįen

-0.14

ako

-0.14

imson

-0.14

POSITIVE LOGITS

 permission

0.25

 assistance

0.22

 help

0.21

 advice

0.19

permission

0.18

 Permission

0.17

 Assistance

0.16

aid

0.15

 Ð¿Ð¾Ð¼Ð¾ÑīÐ¸

0.15

Permission

0.15

Activations Density 0.127%

phrases related to requesting assistance or information

No Comments

No Known Activations