INDEX

Explanations

references to procedural and logistical information

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

 horm

-0.07

Dum

-0.06

ÐµÐ»ÐµÑĦ

-0.06

CU

-0.06

Mahon

-0.06

ANGO

-0.06

 YÄ±l

-0.06

Î¹Ïĩ

-0.06

rote

-0.06

ãĥªãĥ³ãĤ¯

-0.06

POSITIVE LOGITS

 turnaround

0.10

fast

0.09

 fast

0.09

 Fast

0.09

enko

0.09

 fastest

0.09

_FAST

0.09

 speedy

0.09

 delivery

0.08

 exped

0.08

Activations Density 0.046%

No Known Activations