INDEX

Explanations

references to directional locations and orientations

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

 للمعارف

-0.75

 Normdatei

-0.70

󠁴

-0.65

thâu

-0.56

GEBURTSDATUM

-0.54

@[+][

-0.54

IUrlHelper

-0.54

/**

-0.50

dafx

-0.50

 multiplic

-0.50

POSITIVE LOGITS

 gespannt

0.41

 disambiguazione

0.40

fatalError

0.35

 Abteilung

0.35

 convertido

0.35

 embarazada

0.35

 halus

0.35

 Schulter

0.34

}{*}{

0.34

gól

0.33

Activations Density 0.040%

references to directional locations and orientations

No Comments

No Known Activations