INDEX

Explanations

usersummary<TOP_POSITIVE_LOGITS>Summary</TOP_POSITIVE_LOGITS><TOP_ACTIVATING_TEXTS>Summary</TOP_ACTIVATING_TEXTS><MAX_ACTIVATING_TOKENS>Summary</MAX_ACTIVATING_TOKENS><TOKENS_AFTER_MAX_ACTIVATING_TOKEN>user</TOKENS_AFTER_MAX_ACTIVATING_TOKEN>Explain the neuron above with a word or phrase, not a complete sentence

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

of

1.51

1.35

that

1.20

1.19

1.15

 that

1.09

하다

1.05

1.04

it

1.04

1.03

POSITIVE LOGITS

↵

1.31

ের

1.31

কে

1.29

ın

1.15

ang

1.13

 médicos

1.10

 dernier

1.05

 የተለያዩ

1.05

 zeker

1.04

ஸ்

1.02

Activations Density 0.004%