INDEX

Explanations

patterns related to quantity and comparisons in various contexts

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

scores and amounts

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

GEBURTSDATUM

-0.52

rawDesc

-0.51

 AssemblyCulture

-0.48

 fieldNum

-0.48

contentLoaded

-0.44

 defaultstate

-0.43

Попис

-0.42

 iNdEx

-0.42

 disambiguazione

-0.42

üyada

-0.41

POSITIVE LOGITS

ImageContext

0.42

 αρ

0.41

ysty

0.41

 InputDecoration

0.40

 ilang

0.40

 SEVER

0.39

 almost

0.39

 lots

0.38

 særlig

0.37

 several

0.37

Activations Density 0.388%

patterns related to quantity and comparisons in various contexts

scores and amounts

No Comments

No Known Activations