INDEX

Explanations

places where truth or falsity is being checked, often in the context of code or logic

oai_token-act-pair · gemini-2.0-flash

Code checking for emptiness

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_23/width_16k/average_l0_25

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

verwijspagina

-0.67

awtextra

-0.62

tvguidetime

-0.61

 فريبيس

-0.61

UnusedPrivate

-0.59

 تضيفلها

-0.59

 صوتيه

-0.58

 للاسماء

-0.56

Попис

-0.56

kuuta

-0.55

POSITIVE LOGITS

"".

0.56

''.

0.54

EditorBrowsable

0.52

||

0.52

 onPostExecute

0.50

vegli

0.49

'-

0.46

''

0.46

""

0.46

 vain

0.45

Activations Density 13.862%

places where truth or falsity is being checked, often in the context of code or logic

Code checking for emptiness

No Known Activations