INDEX

Explanations

endings and suffixes that suggest complex or modified meanings

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

er

-0.35

-0.34

Ø±

-0.29

-0.27

-0.25

à¤¨

-0.24

lip

-0.23

hop

-0.23

pa

-0.23

POSITIVE LOGITS

licity

0.20

licate

0.19

ensely

0.18

orarily

0.17

licit

0.17

lica

0.17

lications

0.17

loi

0.16

adoo

0.16

dete

0.16

Activations Density 0.029%

endings and suffixes that suggest complex or modified meanings

No Comments

No Known Activations