INDEX

Explanations

Markov decision process

The neuron fires most strongly on specialized technical terms and jargon (e.g. “stationary,” “Markov Decision Processes,” “state,” “reinforcement learning,” etc.) in scientific or academic text.

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

𝙃

-0.87

 giocare

-0.84

tagem

-0.81

 preen

-0.79

ACTED

-0.79

hadas

-0.77

ябре

-0.77

tuje

-0.76

neten

-0.76

 Réunion

-0.76

POSITIVE LOGITS

mdp

1.27

 Markov

1.01

 environment

0.97

 reward

0.96

 states

0.94

 transition

0.89

 States

0.88

reward

0.86

 transitions

0.84

Absor

0.84

Activations Density 0.012%