INDEX

Explanations

expressions of gratitude or appreciation

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 ͜ʖ

-0.39

 Normdatei

-0.34

 suddenly

-0.33

licherweise

-0.32

pAd

-0.32

tighets

-0.32

{?>

-0.32

chieht

-0.31

Населення

-0.31

(`/

-0.30

POSITIVE LOGITS

ocardio

0.63

脚注の使い方

0.62

 kasarigan

0.61

Portail

0.58

 nahilalakip

0.57

 autorytatywna

0.56

 EconPapers

0.56

PreferredItem

0.55

 Wikimedijinoj

0.54

liferay

0.54

Activations Density 0.002%

expressions of gratitude or appreciation

No Known Activations