INDEX

Explanations

numbers or quantitative data

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

numbers and symbols

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

 Lightboxes

-0.52

IsContent

-0.50

LookAnd

-0.44

fortunes

-0.44

 ſche

-0.44

 romántico

-0.43

 perman

-0.43

 cherchés

-0.43

 Perman

-0.43

 myſelf

-0.43

POSITIVE LOGITS

 يتيمه

0.51

Erstellt

0.42

complexContent

0.41

August

0.41

Dorothy

0.41

fromLTRB

0.39

 disambiguazione

0.39

[]:

0.39

Zu

0.38

Abstract

0.38

Activations Density 0.004%

numbers or quantitative data

numbers and symbols

No Comments

No Known Activations