INDEX

Explanations

The neuron is tuned to detect mentions of scientific or technical research contexts—words related to experiments, scientists, and experimental procedures.

scientific research and experiments

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



-1.15

がついて

-1.14

 archaeology

-1.10

🦄

-1.10

᾿

-1.09

ſe

-1.09

("&

-1.08

 almohada

-1.08

inerario

-1.06

ᴮ

-1.06

POSITIVE LOGITS

 research

1.70

 experiments

1.65

 experiment

1.56

 create

1.49

 testing

1.46

 attempts

1.38

 test

1.38

 tests

1.30

 attempt

1.30

 planned

1.29

Activations Density 0.026%