INDEX

Explanations

discussions and questions, often containing logic or conditions, within mathematical or game-like problems.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Hzfinfdu/SlimPajama-3B and open-r1/OpenR1-Math-220k

Features

32,768

Data Type

float32

Hook Name

blocks.17.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ø§Ø±Ú©

-0.06

Nes

-0.06

wil

-0.06

 Selected

-0.06

 closest

-0.06

uitka

-0.06

Subtitle

-0.06

ansen

-0.05

_CLAMP

-0.05

lou

-0.05

POSITIVE LOGITS

0.07

only

0.06

ØªØ§

0.06

æĸ°å¢ŀ

0.06

 WaitFor

0.06

Activations Density 0.047%