INDEX

Explanations

terms related to marking or identifying items

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

katapos

-0.79

<tbody>

-0.70

 Gauthier

-0.67

 Auvergne

-0.67

edicated

-0.66

bibinfo

-0.66

 Montagu

-0.66

anjutnya

-0.65

氓

-0.65

tenham

-0.64

POSITIVE LOGITS

 Marks

2.21

 mark

2.13

 marks

2.13

 Mark

2.06

 MARKS

2.01

Mark

2.00

Marks

2.00

 MARK

1.99

marks

1.97

mark

1.95

Activations Density 0.095%

terms related to marking or identifying items

No Comments

No Known Activations