INDEX

Explanations

abbreviations and acronyms

np_acts-logits-general · gemini-2.5-flash-lite

the Greek letter "Psi" or capitalized abbreviations/acronyms (like DPI, BPS, VN) in technical or formal documents.

oai_token-act-pair · claude-4-5-sonnet Triggered by @bfgcn7frf5

The neuron fires on uppercase “initialisms” or acronyms (often technical names, variable identifiers, or version tags) composed of consecutive capital letters (sometimes with digits).

oai_token-act-pair · o4-mini Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

呔

-2.41

-2.34

 '');

-2.33

鲼

-2.28

 ikke

-2.25

獃

-2.22

嵴

-2.16

」

-2.14

'')

-2.13

POSITIVE LOGITS

 offizielle

2.94

 zusätzliche

2.78

 これ

2.64

 mögliche

2.58

 wahre

2.45

 bestätigen

2.44

 comprehensively

2.41

 größere

2.41

 klassische

2.33

After

2.30

Activations Density 0.002%

abbreviations and acronyms

the Greek letter "Psi" or capitalized abbreviations/acronyms (like DPI, BPS, VN) in technical or formal documents.

The neuron fires on uppercase “initialisms” or acronyms (often technical names, variable identifiers, or version tags) composed of consecutive capital letters (sometimes with digits).

No Comments

No Known Activations

abbreviations and acronyms

the Greek letter "Psi" or capitalized abbreviations/acronyms (like DPI, BPS, VN) in technical or formal documents.

The neuron fires on uppercase “initialisms” or acronyms (often technical names, variable identifiers, or version tags) composed of consecutive capital letters (sometimes with digits).

No Comments

No Known Activations