INDEX

Explanations

for/to purpose or action

The neuron strongly detects the preposition “for” when it’s used to introduce purpose clauses.

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

by

-1.60

on

-1.34

 soorten

-1.22

 voordelen

-1.12

お疲れ様でした

-1.11

 sebagainya

-1.09

dina

-1.09

zkou

-1.08

ņa

-1.06

 even

-1.05

POSITIVE LOGITS

 which

1.67

 jonka

1.23

来说

1.19

に基

1.19

 ktoré

1.15

其

1.14

其

1.14

 kautta

1.12

vät

1.11

 poderia

1.09

Activations Density 0.194%