INDEX

Explanations

concepts related to authority and governance

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

 itself

-0.33

 themselves

-0.25

 herself

-0.25

 Himself

-0.23

 himself

-0.23

 ourselves

-0.21

 yourself

-0.21

 oneself

-0.19

 samot

-0.19

 yourselves

-0.19

POSITIVE LOGITS

 separate

0.28

 Separate

0.23

 independent

0.22

çĭ¬ç«ĭ

0.21

 seperate

0.20

 independently

0.19

 independ

0.19

 rather

0.19

Independent

0.18

 Independent

0.18

Activations Density 0.036%

concepts related to authority and governance

No Comments

No Known Activations