INDEX

Explanations

words inside blocks of code, especially Java

oai_token-act-pair · gemini-2.0-flash

parentheses

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_8/width_16k/average_l0_52

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+#+#

-0.69

 EconPapers

-0.63

farwyddwr

-0.62

énario

-0.59

Demografie

-0.59

 विश्वसनीयता

-0.57

httphttps

-0.55

Билгалдахарш

-0.55

يميديا

-0.54

 Waray

-0.53

POSITIVE LOGITS

<bos>

0.62

الإنجليزية

0.54

SharedCtor

0.53

สร

0.52

Portail

0.51

TabIndex

0.50

 يتيمه

0.49

memberOf

0.46

 pector

0.46

 Huma

0.45

Activations Density 44.336%

words inside blocks of code, especially Java

parentheses

No Comments

No Known Activations