INDEX

Explanations

working on a

The neuron detects the phrase “working on” (especially in constructions like “working on a…”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

</h3>

-1.69

 gelegt

-1.61

的食物

-1.58

 összefoglal

-1.55

的眼神

-1.53

vertebrate

-1.52

癱

-1.49

的声音

-1.48

 あります

-1.47

 groupings

-1.45

POSITIVE LOGITS

↵↵

1.74

but

1.67

as

1.61

":

1.52

 also

1.52

 Taking

1.51

1.49

в

1.48

')

1.48

it

1.45

Activations Density 0.028%