INDEX

Explanations

phrases indicating processes of planning or development

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

risk model

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

 AssemblyProduct

-0.53

UrlResolution

-0.52

+:+

-0.50

parsedMessage

-0.50

+#+

-0.45

verwijspagina

-0.45

 ujednoznacz

-0.42

 ब्रेकडाउन

-0.42

tanleria

-0.42

ScopeManager

-0.41

POSITIVE LOGITS

 arşivlendi

0.56

TypedDataSet

0.45

SuspendLayout

0.43

 NSCoder

0.40

AppMethodBeat

0.40

istoitu

0.39

Rujuakan

0.39

ngrx

0.38

MLLoader

0.38

Kaynakça

0.37

Activations Density 0.210%

phrases indicating processes of planning or development

risk model

No Comments

No Known Activations