INDEX

Explanations

the word "its" followed or preceded by other words indicative of spatial relationships or adversarial situations

oai_token-act-pair · gemini-2.0-flash

New Auto-Interp

Configuration

fnlp/Llama-Scope-R1-Distill/400M-Slimpajama-400M-OpenR1-Math-220k/L11R

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.11.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/MPL

-0.07

_hal

-0.07

anna

-0.06

vac

-0.06

 Elli

-0.06

otte

-0.06

ubu

-0.06

ais

-0.05

 bá»ĵi

-0.05

POSITIVE LOGITS

own

0.08

spacer

0.07

rah

0.06

 Leaf

0.06

 Feinstein

0.06

 leaf

0.06

Asian

0.06

Leaf

0.06

ulent

0.06

Ð½Ñİ

0.06

Activations Density 0.072%

the word "its" followed or preceded by other words indicative of spatial relationships or adversarial situations

No Comments

No Known Activations