INDEX

Explanations

code-related syntax and operations

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

code structure and data insertion

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

NameInMap

-0.55

PropertyChanging

-0.54

encodeWith

-0.49

IEU

-0.46

 invokingState

-0.45

modelBuilder

-0.44

 nahilalakip

-0.44

 <>",

-0.43

lês

-0.42

quité

-0.41

POSITIVE LOGITS

Personendaten

0.40

 mensaj

0.39

 المعيارى

0.38

AsUp

0.37

XmlEnum

0.34

 aguja

0.34

 camada

0.33

 носи

0.33

 Isto

0.33

 randomised

0.33

Activations Density 0.093%

code-related syntax and operations

code structure and data insertion

No Comments

No Known Activations