INDEX

Explanations

references to battles and conflicts between groups or individuals

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

being destroyed

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

MigrationBuilder

-0.50

ragalactic

-0.49

----</

-0.49

WebServlet

-0.49

devamını

-0.49

NUMX

-0.48

saraba

-0.48

OrBuilder

-0.47

TokenNameLBRACE

-0.47

rollment

-0.46

POSITIVE LOGITS

smiley

0.42

 يتيمه

0.38

acana

0.31

offsetof

0.30

 متعلقه

0.30

ฝ

0.30

MfG

0.28

مص

0.28

arela

0.27

관

0.27

Activations Density 0.116%

references to battles and conflicts between groups or individuals

being destroyed

No Comments

No Known Activations