INDEX

Explanations

references to geographical features and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 40-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_40/width_16k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.40.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

 Sligo

-0.43

 Telangana

-0.42

 Sinaloa

-0.42

faden

-0.41

Jîn

-0.40

 Smyrna

-0.37

 GREEK

-0.37

 Winaray

-0.36

تفصیلات

-0.36

 Abia

-0.35

POSITIVE LOGITS

 polar

1.02

 Antarctic

0.95

 Polar

0.94

 Antarctica

0.90

Polar

0.86

 antar

0.85

polar

0.85

Antar

0.80

 Antar

0.76

 polaire

0.73

Activations Density 0.275%

references to geographical features and locations

No Comments

No Known Activations