INDEX

Explanations

phrases related to raising questions or concerns

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

New Auto-Interp

Configuration

Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post

Features

65,536

Data Type

float32

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

gated

Context Size

1,024

Dataset

Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

áº¡c

-0.16

licative

-0.15

Shock

-0.15

 tablesp

-0.14

 Complaint

-0.14

 æ±

-0.13

íĿ¬

-0.13

Î»Î¹

-0.13

inclusive

-0.13

 diá»ĩn

-0.13

POSITIVE LOGITS

 questions

0.46

 Questions

0.34

questions

0.32

 question

0.31

 serious

0.30

Questions

0.28

_questions

0.26

 Fragen

0.23

.questions

0.22

 QUEST

0.22

Activations Density 0.033%

phrases related to raising questions or concerns

No Comments

No Known Activations