INDEX

Explanations

numerical values and their corresponding context

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

icus

-0.06

 Andrews

-0.06

FieldType

-0.06

edn

-0.06

/qu

-0.06

.Cmd

-0.05

burger

-0.05

AUD

-0.05

Defaults

-0.05

POSITIVE LOGITS

 Chain

0.13

Chain

0.11

chain

0.10

 Chains

0.09

 chain

0.09

Pac

0.09

 Pist

0.09

pac

0.09

-chain

0.08

_chain

0.08

Activations Density 0.006%