INDEX

Explanations

technical terms and parameters related to programming and data structures.

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

technical terms and identifiers

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

IntoConstraints

-0.73

 فريبيس

-0.68

AndEndTag

-0.67

 '\\;'

-0.62

 estekak

-0.62

GEBURTSDATUM

-0.59

المشاركات

-0.59

msgTypes

-0.59

__':

-0.58

 otomatig

-0.58

POSITIVE LOGITS

 commandement

0.37

 capteur

0.35

 atualizado

0.34

 asuntos

0.33

 malheur

0.33

麻煩

0.33

 courriel

0.32

 comarca

0.32

 cabina

0.31

 partenariat

0.31

Activations Density 0.071%

technical terms and parameters related to programming and data structures.

technical terms and identifiers

No Comments

No Known Activations