INDEX

Explanations

purified and peeled

The neuron strongly fires on technical terms describing biochemical isolation or purification steps (e.g. solubilization, purification, isolation, elution).

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

</i>

-1.56

en

-1.42

ati

-1.29

 hammers

-1.26

</td>

-1.25

-1.23

還好

-1.22

还好

-1.21

ti

-1.20

 effic

-1.20

POSITIVE LOGITS

 from

1.85

ではでは

1.58

ズム

1.29

というわけで

1.27

Palmar

1.27

nemers

1.25

vlád

1.24

Much

1.23

This

1.20

 ensures

1.20

Activations Density 0.074%