INDEX

Explanations

field names and properties

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

array or dictionary key access operations in code, particularly when enclosed in brackets or quotes.

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

The neuron detects quoted identifiers used as keys inside square‐bracket indexing (e.g. array or object property accesses).

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

-1.55

یت

-1.51

鋮

-1.46

aning

-1.39

...,

-1.37

粪

-1.37

脑海

-1.34

 például

-1.34

 néhány

-1.31

ły

-1.30

POSITIVE LOGITS

 would

1.73

or

1.65

 после

1.60

its

1.60

欷

1.52

see

1.50

 feel

1.50

 científicas

1.46

 Preferencias

1.46

are

1.46

Activations Density 0.004%

field names and properties

array or dictionary key access operations in code, particularly when enclosed in brackets or quotes.

The neuron detects quoted identifiers used as keys inside square‐bracket indexing (e.g. array or object property accesses).

No Comments

No Known Activations

field names and properties

array or dictionary key access operations in code, particularly when enclosed in brackets or quotes.

The neuron detects quoted identifiers used as keys inside square‐bracket indexing (e.g. array or object property accesses).

No Comments

No Known Activations