INDEX

Explanations

structured data, particularly in a markup or programming context

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

XML or code strings

np_acts-logits-general · gemini-2.0-flash

foreign identifiers and code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WebServlet

-0.53

 scrum

-0.52

 Rugby

-0.48

 Toul

-0.45

 történ

-0.45

 szerző

-0.43

vans

-0.43

 confinement

-0.42

ITUD

-0.41

gwt

-0.41

POSITIVE LOGITS

windowFixed

0.80

Bioaccumulative

0.74

 BoxFit

0.67

 Bioaccumulative

0.67

colgroup

0.66

 kaarangay

0.65

enumii

0.64

olstadt

0.63

 فريبيس

0.61

 DbSet

0.59

Activations Density 0.057%

structured data, particularly in a markup or programming context

XML or code strings

foreign identifiers and code

No Comments

No Known Activations