INDEX

Explanations

proper names, particularly related to individuals and their associated attributes

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

names and people

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

aarrggbb

-0.62

principalColumn

-0.59

 فريبيس

-0.59

 jsPsych

-0.58

SequentialGroup

-0.56

 consultato

-0.55

 Administrativna

-0.55

 Italijanski

-0.54

tagHelperRunner

-0.54

ToBounds

-0.54

POSITIVE LOGITS

 enfans

0.43

 garçons

0.43

 alumnos

0.41

 Ajoutez

0.40

 fleuve

0.40

 professeurs

0.39

 Egipto

0.39

 étudiants

0.38

 saudara

0.37

 noël

0.37

Activations Density 0.011%

proper names, particularly related to individuals and their associated attributes

names and people

No Comments

No Known Activations