INDEX

Explanations

words related to physical locations or spaces

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/4-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.4.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

plicial

-1.67

itime

-1.59

unnumbered

-1.57

izable

-1.56

ikan

-1.53

labelled

-1.52

iative

-1.52

onomic

-1.52

istically

-1.49

undry

-1.48

POSITIVE LOGITS

 syndrome

1.71

 Syndrome

1.54

robe

1.49

 qualify

1.48

 rights

1.41

 topped

1.39

 matches

1.36

 qual

1.34

 Biden

1.34

 match

1.31

Activations Density 1.153%

words related to physical locations or spaces

No Comments

No Known Activations