INDEX

Explanations

formal academic structures and terminology

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

treatment outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

WireFormatLite

-0.60

 newVal

-0.57

chartInstance

-0.57

;->

-0.56

γρά

-0.54

Jîn

-0.53

ceous

-0.53

})`

-0.52

TLE

-0.52

NewUrlParser

-0.51

POSITIVE LOGITS

الإنجليزية

0.85

Personensuche

0.76

TestingModule

0.68

izr

0.60

orianCalendar

0.59

WebServlet

0.59

 "..\..\

0.59

InjectAttribute

0.54

 <=",

0.53

loài

0.51

Activations Density 1.761%

formal academic structures and terminology

treatment outcomes

No Comments

No Known Activations