INDEX

Explanations

structured data definitions, such as database schemas or programming constructs

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

struct definitions or declarations

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

viewDidLoad

-0.52

AndEndTag

-0.50

GOTREF

-0.47

RODUCTION

-0.47

hyrchwyd

-0.42

instancetype

-0.42

 imageNamed

-0.42

setVerticalGroup

-0.42

 TAMBÉM

-0.41

RTHOOK

-0.38

POSITIVE LOGITS

 bezoekers

0.44

 cottages

0.44

Toy

0.43

ronpa

0.43

meuble

0.43

icitis

0.42

 Hutchison

0.42

 sẻ

0.42

 käytt

0.42

albero

0.42

Activations Density 0.030%

structured data definitions, such as database schemas or programming constructs

struct definitions or declarations

No Comments

No Known Activations