INDEX

Explanations

terms related to locking and unlocking mechanisms

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

ccione

-0.17

edeki

-0.17

GER

-0.16

à¸¥à¸ĩ

-0.15

lags

-0.15

éľ²

-0.15

ë¡ľ

-0.15

è´¹

-0.15

ulfilled

-0.14

yb

-0.14

POSITIVE LOGITS

/un

0.24

 doors

0.21

heed

0.21

tober

0.20

busters

0.20

 bootloader

0.18

 door

0.18

able

0.18

edException

0.18

(lock

0.18

Activations Density 0.015%

terms related to locking and unlocking mechanisms

No Comments

No Known Activations