INDEX

Explanations

`sed`, `if`, `torch`, `Warning`

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

 assured

1.17

 slipped

1.16

 sounded

1.09

 শাস

1.07

 retrait

1.05

 coronal

1.05

ल

1.04

 displace

1.03

 discursive

1.02

 coincided

1.01

POSITIVE LOGITS

 لي

1.19

 دان

1.17

げる

1.13

 Runtime

1.10

و

1.10

ilangkan

1.08

 महंगे

1.07

 قيمة

1.06

Big

1.06

‌ها

1.06

Activations Density 0.001%