INDEX

Explanations

words that express recognition or validation

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

 einz

-0.19

ToSend

-0.17

 Nack

-0.15

guard

-0.14

 Trang

-0.14

asa

-0.14

thed

-0.14

Î³Î¿Ïħ

-0.14

mo

-0.14

tons

-0.14

POSITIVE LOGITS

 kepada

0.66

 unto

0.56

to

0.50

åĪ°

0.49

 åĪ°

0.38

 Ø¥ÙĦÙī

0.38

 Ø¥ÙĦÙĬÙĩ

0.33

aos

0.32

 Äĳáº¿n

0.30

 ÏĥÏĦÎ¿ÏħÏĤ

0.29

Activations Density 0.121%

words that express recognition or validation

No Comments

No Known Activations