INDEX

Explanations

words and phrases related to conveying strong emotion such as apologies and gratitude

oai_token-act-pair · gemini-2.0-flash

formal correspondence

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

Negative Logits

 ederim

-0.52

UALA

-0.48

reportWebVitals

-0.48

StoredProcedure

-0.47

findpost

-0.47

 كومونز

-0.46

Véxase

-0.46

setVerticalGroup

-0.45

 navegando

-0.45

 dépasse

-0.45

POSITIVE LOGITS

ophones

0.55

ChromeDriver

0.55

 '\\;'

0.54

 complic

0.54

AndroidJUnit

0.52

kees

0.51

‪

0.50

EntityManager

0.50

icto

0.50

Activations Density 0.625%

words and phrases related to conveying strong emotion such as apologies and gratitude

formal correspondence

No Known Activations