INDEX

Explanations

references to internal structures or systems

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

following "internal"

np_acts-logits-general · gemini-2.0-flash

internal followed by specification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 Parcelable

-0.68

 SwiftUI

-0.60

 AssemblyCulture

-0.60

__':

-0.55

 ControllerBase

-0.55

 internazionali

-0.54

 navideñas

-0.51

CodeAttribute

-0.50

★★★★★

-0.49

 rayures

-0.49

POSITIVE LOGITS

 workings

0.76

 combustion

0.76

 affairs

0.73

 والخ

0.71

 Affairs

0.63

UpInside

0.63

XmlAccessType

0.62

 organs

0.62

 InputDecoration

0.62

AntiForgeryToken

0.61

Activations Density 0.110%

references to internal structures or systems

following "internal"

internal followed by specification

No Comments

No Known Activations