INDEX

Explanations

references to false or misleading information

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

GroupBox

-0.15

//**↵

-0.15

aylight

-0.14

rade

-0.14

 Dysfunction

-0.14

#__

-0.13

PIX

-0.13

$__

-0.13

vrier

-0.13

olley

-0.13

POSITIVE LOGITS

/false

0.20

 claims

0.18

 evidence

0.18

 story

0.16

pret

0.16

-bal

0.16

ument

0.16

 claim

0.16

 pretended

0.16

ëī´ìĬ¤

0.16

Activations Density 0.078%

references to false or misleading information

No Comments

No Known Activations