INDEX

Explanations

references to various roles or titles in different contexts

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

ÑıÐ³

-0.17

ÑĤÑı

-0.17

ÑĢÑĸÑĩ

-0.16

ohan

-0.16

éĺħè¯»æ¬¡æķ°

-0.15

Ð¼ÑĸÐ½

-0.15

Ã©ri

-0.14

zbo

-0.14

Ä©nh

-0.14

azing

-0.14

POSITIVE LOGITS

0.32

’

0.26

hip

0.22

hips

0.22

0.20

chaft

0.20

are

0.20

who

0.19

'-

0.19

‘

0.18

Activations Density 0.035%

references to various roles or titles in different contexts

No Comments

No Known Activations