INDEX

Explanations

geographical features and place names

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ujednoznacz

-0.47

BeginContext

-0.42

__':

-0.36

 pié

-0.35

gimento

-0.34

 juger

-0.31

 juez

-0.31

 directiva

-0.31

 sitting

-0.30

 comprensión

-0.30

POSITIVE LOGITS

ConstraintMaker

0.62

 مشين

0.60

UserScript

0.59

CloseOperation

0.55

 sumpay

0.53

 فريبيس

0.51

XmlAccessType

0.50

stdc

0.49

tanleria

0.49

hdashline

0.48

Activations Density 0.037%

geographical features and place names

No Comments

No Known Activations