INDEX

Explanations

the presence of strong warning indicators or important values related to safety or restrictions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

).)

-1.59

 organization

-1.56

 designation

-1.54

idden

-1.46

 aging

-1.41

 measures

-1.35

engers

-1.35

(%)

-1.35

 Newman

-1.34

 Award

-1.33

POSITIVE LOGITS

»

2.98

Ī

2.79

¿

2.78

¿½

2.73

Ŀ

2.53

¢

2.52

¼

2.52

ŀ

2.50

¦

2.47

į

2.45

Activations Density 0.198%

the presence of strong warning indicators or important values related to safety or restrictions

No Comments

No Known Activations