INDEX

Explanations

ZIP code identifiers and their relevant numerical data

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

foreign or technical terms

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

httphttps

-0.64

المكان

-0.56

تقاوى

-0.54

 sumpay

-0.51

 electrolux

-0.48

Personendaten

-0.47

 snippetHide

-0.46

 esternos

-0.46

RTLI

-0.46

цездатний

-0.46

POSITIVE LOGITS

 AssemblyCulture

0.46

 chrétien

0.43

Geografi

0.35

jména

0.34

kuuta

0.33

GraphicsUnit

0.32

 LCCN

0.32

fresh

0.31

 sırt

0.31

 sauvage

0.31

Activations Density 0.002%

ZIP code identifiers and their relevant numerical data

foreign or technical terms

No Comments

No Known Activations