INDEX

Explanations

specific programming and database-related terms and structures

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

collection or view

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

 sarili

-0.48

 complete

-0.41

 passing

-0.39

 veik

-0.39

 directed

-0.39

 Veranstaltungen

-0.38

 exponer

-0.38

Vertices

-0.37

 ready

-0.36

 nett

-0.36

POSITIVE LOGITS

exitRule

0.59

onViewCreated

0.57

="@+

0.56

ModelAdmin

0.55

parsedMessage

0.54

########.

0.54

 "..\..\..\

0.54

UnitTesting

0.53

UnusedPrivate

0.52

GOTREF

0.52

Activations Density 0.092%

specific programming and database-related terms and structures

collection or view

No Comments

No Known Activations