INDEX

Explanations

instances of quantitative or numerical information

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

readjusting to life

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

estand

-0.48

 surla

-0.43

 Israelis

-0.42

Snapchat

-0.42

bindingNavigator

-0.42

 Canadiens

-0.41

 Snapchat

-0.40

lapsingToolbar

-0.40

 okuyayım

-0.40

 Arhivirano

-0.39

POSITIVE LOGITS

awarkan

0.41

SerializedSize

0.41

WebServlet

0.41

 argint

0.41

+#+

0.41

haviors

0.40

PreferredItem

0.40

 jonge

0.40

GEBURTSDATUM

0.39

OCCURRED

0.39

Activations Density 0.166%

instances of quantitative or numerical information

readjusting to life

No Comments

No Known Activations