INDEX

Explanations

geographic locations and relevant identifiers within a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

 Leuven

-0.63

DotNetBar

-0.54

ulemon

-0.51

 heuer

-0.50

 Eindhoven

-0.50

 Seil

-0.49

 resourceCulture

-0.48

vedra

-0.48

 Alois

-0.46

 Sepp

-0.46

POSITIVE LOGITS

Anhalt

0.70

Berlin

0.52

WithIdentifier

0.49

 Berlin

0.47

flink

0.47

cipar

0.47

//

0.45

حوالہ

0.44

 Baltic

0.42

 BERLIN

0.42

Activations Density 0.051%

geographic locations and relevant identifiers within a document

No Comments

No Known Activations