INDEX

Explanations

potential use or proposed action

The neuron strongly fires on past‐participle or adjectival forms followed by a small preposition or the infinitive marker “to” (e.g. “involved in the …,” “known to interact with …,” “used to translate …”).

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

了些

-1.51

了不少

-1.40

TYPE

-1.37

ldots

-1.24

 ۱۵

-1.24

 ۱۱

-1.21

lag

-1.19

nx

-1.16

但却

-1.16

rl

-1.15

POSITIVE LOGITS

 there

1.51

 other

1.30

过来了

1.30

 meditación

1.30

 cirugía

1.26

++;

1.26

 richied

1.22

 другими

1.18

則

1.17

 agarre

1.16

Activations Density 0.215%