INDEX

Explanations

references to specific elements, entities, or variables in scientific contexts

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

color, get, NH, phase, U

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

 kasarigan

-0.63

ViewFeatures

-0.60

parsedMessage

-0.60

 betweenstory

-0.57

 disambiguazione

-0.57

 ExecuteAsync

-0.57

Gön

-0.56

 שוליים

-0.56

 Wicidata

-0.55

tagext

-0.54

POSITIVE LOGITS

colorPrimary

0.42

 einzeln

0.40

География

0.38

 esternos

0.37

fahrene

0.36

or

0.36

getLatitude

0.35

gehende

0.35

orgi

0.33

NewUrlParser

0.33

Activations Density 0.107%

references to specific elements, entities, or variables in scientific contexts

color, get, NH, phase, U

No Comments

No Known Activations