INDEX

Explanations

symbols and fragments indicative of formatting or markup language elements

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

legal and scientific citations

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 planificación

-0.68

 valentín

-0.63

 edades

-0.58

 respiración

-0.58

 perfección

-0.56

 taşlı

-0.56

 ventaja

-0.56

 deudas

-0.56

 clín

-0.55

 prohibido

-0.55

POSITIVE LOGITS

脚注の使い方

0.65

XmlAccessType

0.63

OGND

0.58

 ***!

0.54

rrggbb

0.52

unnamed

0.52

المناصب

0.50

 unnamed

0.48

ilever

0.48

toHaveBeen

0.47

Activations Density 0.162%

symbols and fragments indicative of formatting or markup language elements

legal and scientific citations

No Comments

No Known Activations