INDEX

Explanations

medical and demographic data related to patients

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

units or abbreviations in parentheses

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

 enfans

-0.45

GOTREF

-0.44

tagHelperRunner

-0.42

AndEndTag

-0.42

 nakalista

-0.42

featureID

-0.40

 Majefty

-0.39

numerusform

-0.37

DoubleQuotes

-0.37

AddTagHelper

-0.35

POSITIVE LOGITS

suya

0.57

rosin

0.51

 nahilalakip

0.49

ҍ

0.48

 disambiguazione

0.48

measurement

0.47

rophobic

0.47

 autorytatywna

0.46

 measurement

0.46

InteropServices

0.46

Activations Density 0.088%

medical and demographic data related to patients

units or abbreviations in parentheses

No Comments

No Known Activations