INDEX

Explanations

the words "can" and "could," sometimes also capturing related words like "making" and "pleasure."

oai_token-act-pair · gemini-2.0-flash

could can

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_2/width_16k/average_l0_49

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

NegativeButton

-0.75

 onCreateView

-0.71

PositiveButton

-0.66

الإنجليزية

-0.64

<bos>

-0.60

oredCriteria

-0.58

ViewFeatures

-0.57

glBegin

-0.57

ImageContext

-0.57

ContentAlignment

-0.57

POSITIVE LOGITS

 للمعارف

0.61

 gekomen

0.55

ailleurs

0.55

Geschiedenis

0.54

Répondre

0.54

 zelfs

0.52

 Selv

0.50

 thèse

0.49

 odeur

0.49

كويكب

0.49

Activations Density 3.850%

the words "can" and "could," sometimes also capturing related words like "making" and "pleasure."

could can

No Known Activations