INDEX

Explanations

definitions, acronyms, and classifications

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

abbreviations and their definitions in scientific and technical texts.

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

The neuron strongly activates on technical multi‐word terms that serve as the spelled‐out definitions of acronyms or abbreviations (e.g. “equilibrative nucleoside transporter,” “hypoxia inducible factor 1α,” “blood pressure”).

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



-1.59

 motoru

-1.55

ynchronously

-1.51

 modrá

-1.48

…»

-1.38

 premières

-1.38

 經典

-1.38

...»

-1.38

###########

-1.37

phalt

-1.36

POSITIVE LOGITS

 there

1.67

we

1.56

 will

1.39

if

1.38

in

1.37

 here

1.36

of

1.33

 need

1.31

 cuando

1.30

now

1.29

Activations Density 0.024%

definitions, acronyms, and classifications

abbreviations and their definitions in scientific and technical texts.

The neuron strongly activates on technical multi‐word terms that serve as the spelled‐out definitions of acronyms or abbreviations (e.g. “equilibrative nucleoside transporter,” “hypoxia inducible factor 1α,” “blood pressure”).

No Comments

No Known Activations

definitions, acronyms, and classifications

abbreviations and their definitions in scientific and technical texts.

The neuron strongly activates on technical multi‐word terms that serve as the spelled‐out definitions of acronyms or abbreviations (e.g. “equilibrative nucleoside transporter,” “hypoxia inducible factor 1α,” “blood pressure”).

No Comments

No Known Activations