INDEX

Explanations

phrases indicating events or actions that occur prior to a specified moment or event

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

"before"

np_acts-logits-general · gemini-2.0-flash

before

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

SourceChecksum

-0.44

 Folge

-0.43

 poème

-0.43

queryInterface

-0.41

 rubia

-0.40

complexType

-0.40

IonicModule

-0.40

 indicada

-0.39

 disambiguazione

-0.39

 heureuse

-0.38

POSITIVE LOGITS

codehaus

0.52

 before

0.52

 bevor

0.51

tionary

0.51

 voordat

0.50

 PrintWriter

0.49

Phy

0.47

DotNetBar

0.47

 Triton

0.47

ramine

0.46

Activations Density 0.186%

phrases indicating events or actions that occur prior to a specified moment or event

"before"

before

No Comments

No Known Activations