INDEX

Explanations

instances of numbers and statistical data being reported

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

numbers in text

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

 propOrder

-0.45

 /\.(

-0.45

 dAtA

-0.44

 crushing

-0.43

 billions

-0.42

ThemeOverlay

-0.42

 GoogleFonts

-0.42

 Deli

-0.42

FontOfSize

-0.41

 NSError

-0.40

POSITIVE LOGITS

RegressionTest

0.49

rungsseite

0.46

ListItemIcon

0.44

IUrlHelper

0.42

 AssemblyCompany

0.41

 disambiguazione

0.39

MLLoader

0.39

 aislados

0.39

verifyException

0.39

yks

0.38

Activations Density 0.208%

instances of numbers and statistical data being reported

numbers in text

No Comments

No Known Activations