INDEX

Explanations

preparing to adapt and change

The neuron strongly activates on verbs that denote dynamic actions of change, adjustment, or response (e.g. adapt, recharge, recover, diversify).

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

把它

-0.94

 claridad

-0.91

na

-0.88

 construit

-0.86

因为它

-0.85

ut

-0.84

Anche

-0.84

在你

-0.84

𝅥

-0.83

这也是

-0.83

POSITIVE LOGITS

and

1.30

 oneself

1.26

 yourself

1.25

in

1.17

 ourselves

1.15

themselves

1.03

 their

1.02

 with

1.02

 themselves

1.01

 himself

0.99

Activations Density 0.088%