INDEX

Explanations

auxiliary verbs and infinitives

oai_token-act-pair · gemini-2.0-flash

be/being

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Havolalar

-0.52

FormTagHelper

-0.52

Autoritní

-0.47

SuspendLayout

-0.47

inaison

-0.46

GEBURTSDATUM

-0.46

TestBed

-0.44

rug

-0.41

ben

-0.41

 étant

-0.41

POSITIVE LOGITS

脚注の使い方

0.83

 cared

0.78

 itſelf

0.69

 AssemblyTitle

0.68

 uplifted

0.67

 entertained

0.66

ergies

0.66

 underestimated

0.66

}{*}{

0.65

 tolerated

0.65

Activations Density 11.341%

auxiliary verbs and infinitives

be/being

No Comments

No Known Activations