INDEX

Explanations

quantitative data or measurements

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

years, weeks, months, lifespan

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



-0.56

 Italijanski

-0.55

 jsPsych

-0.54

ðsíða

-0.52

 QtWidgets

-0.51

 HFILL

-0.51

 ISNI

-0.51

stdc

-0.50

bibinfo

-0.49

₿

-0.49

POSITIVE LOGITS

цездатний

0.40

 gobernador

0.34

mbggenerated

0.33

uğ

0.30

 llavero

0.30

 diseñadores

0.29

 prisonniers

0.29

 labios

0.29

amitié

0.28

radan

0.28

Activations Density 0.210%

quantitative data or measurements

years, weeks, months, lifespan

No Comments

No Known Activations