INDEX

Explanations

passages of text where the author is working through a problem by thinking out loud (i.e. using "I", "we", and "you" to refer to the author of the text)

oai_token-act-pair · gemini-2.0-flash

New Auto-Interp

Configuration

fnlp/Llama-Scope-R1-Distill/400M-Slimpajama-400M-OpenR1-Math-220k/L21R

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.21.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

çľĭçľĭ

-0.07

hei

-0.06

asurer

-0.06

Î¹ÏĥÏĦÎ¿

-0.06

eman

-0.06

Know

-0.06

Nab

-0.06

Saw

-0.06

imit

-0.06

 memiliki

-0.06

POSITIVE LOGITS

 supposed

0.11

 usually

0.10

can

0.09

 need

0.09

 typically

0.09

 should

0.09

 needed

0.09

éĢļå¸¸

0.08

usually

0.08

 needs

0.07

Activations Density 0.226%

passages of text where the author is working through a problem by thinking out loud (i.e. using "I", "we", and "you" to refer to the author of the text)

No Known Activations

passages of text where the author is working through a problem by thinking out loud (i.e. using "I", "we", and "you" to refer to the author of the text)

No Known Activations