INDEX

Explanations

specific references to contact information and author contributions in documents

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

disambiguation concepts

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

Tmp

-0.43

uts

-0.35

 Espinosa

-0.34

tí

-0.33

ό

-0.33

Serial

-0.33

പ

-0.32

 surprises

-0.32

 outing

-0.32

 outfits

-0.31

POSITIVE LOGITS

 AssemblyCompany

0.88

 propOrder

0.76

 disambiguazione

0.71

تقاوى

0.68

rungsseite

0.66

 nonUne

0.66

 &___

0.66

verwijspagina

0.66

AndEndTag

0.66

homonymie

0.65

Activations Density 0.000%

specific references to contact information and author contributions in documents

disambiguation concepts

No Comments

No Known Activations