INDEX

Explanations

programming constructs and control flow operations in code

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Squ

-0.06

 Squad

-0.06

pur

-0.06

ammen

-0.06

 Sour

-0.05

cog

-0.05

unami

-0.05

sat

-0.05

actus

-0.05

uns

-0.05

POSITIVE LOGITS

 input

0.13

_input

0.12

input

0.12

 userInput

0.12

 Input

0.12

-input

0.12

 INPUT

0.12

=input

0.11

è¾ĵåħ¥

0.11

INPUT

0.11

Activations Density 0.047%