INDEX

Explanations

phrases emphasizing the concept of 'the' as a determiner highlighting specific entities

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Huguen

-1.05

 Mahomet

-1.04

 Phry

-1.03

 Shakspeare

-0.98

 Fascism

-0.97

 Agamemnon

-0.96

 Cæsar

-0.93

 Houſe

-0.93

 Cister

-0.92



-0.91

POSITIVE LOGITS

the

1.69

The

1.19

 same

1.13

 entire

1.13

THE

1.13

The

1.04

 most

0.98

 latter

0.98

new

0.96

enthe

0.96

Activations Density 3.972%

phrases emphasizing the concept of 'the' as a determiner highlighting specific entities

No Comments

No Known Activations