INDEX

Explanations

references to specific articles, images, and concepts related to documentation and description

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

legal testimony and cases

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

AndEndTag

-0.70

 noDo

-0.69

AddTagHelper

-0.66

évaluateur

-0.66

 BoxFit

-0.66

 nakalista

-0.62

 resourceCulture

-0.62

ंदीखरीदारी

-0.58

SharedCtor

-0.54

 Lightboxes

-0.53

POSITIVE LOGITS

مسلسل

0.33

InjectMocks

0.33

 DIAL

0.33

 للمعارف

0.33

 policial

0.32

alm

0.32

 tadi

0.32

arkiv

0.31

цездатний

0.30

 dure

0.30

Activations Density 0.178%

references to specific articles, images, and concepts related to documentation and description

legal testimony and cases

No Comments

No Known Activations