INDEX

Explanations

various programming constructs and statements related to data management and processing in code

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

foreign and programming terms

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

-0.68

MigrationBuilder

-0.66

tonode

-0.51

ksikon

-0.51

UnsafeEnabled

-0.50

dafx

-0.50

wahati

-0.47

 Taktlose

-0.47

InitVars

-0.47

رشف

-0.47

POSITIVE LOGITS

 thereof

0.43

 olmayan

0.40

 Represents

0.40

 Schülerinnen

0.40

fortawesome

0.38

 Mexique

0.38

 AspNetCore

0.38

Ārējās

0.38

 representing

0.38

 respectively

0.38

Activations Density 0.007%

various programming constructs and statements related to data management and processing in code

foreign and programming terms

No Comments

No Known Activations