INDEX

Explanations

prepositions that indicate location or spatial relationships

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

-0.30

The

-0.29

↵

-0.29

 ouvrages

-0.29

and

-0.28

 Romains

-0.28

POSITIVE LOGITS

PreferredItem

1.02

 мәкал

1.01

MENAFN

0.98

majánló

0.96

iſen

0.96

MemoryWarning

0.90

wiliwch

0.86

تقاوى

0.86

citroën

0.86

 Infórmanos

0.85

Activations Density 0.123%

prepositions that indicate location or spatial relationships

No Comments

No Known Activations