INDEX

Explanations

proper nouns, specifically names and titles

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

weed

-0.15

umblr

-0.14

idia

-0.14

TAB

-0.14

ining

-0.14

ç

-0.14

-fi

-0.13

 ============================================================================↵

-0.13

owski

-0.13

POSITIVE LOGITS

åŁ

0.17

Ã¤ch

0.16

eteria

0.14

à¥ĩà¤¯

0.14

gorm

0.14

ainen

0.14

 fragile

0.14

ogi

0.14

ÙĪÙĦÙĪØ¬

0.14

lint

0.13

Activations Density 0.026%

proper nouns, specifically names and titles

No Comments

No Known Activations