INDEX

Explanations

references to notable historical figures and events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 dominant

-1.33

enance

-1.29

eds

-1.28

mente

-1.26

'"

-1.26

linger

-1.24

elp

-1.24

(%

-1.24

ool

-1.23

lication

-1.23

POSITIVE LOGITS

²

1.65

 marks

1.63

âĢħ

1.43

ĻĤ

1.40

dered

1.35

itos

1.33

 [âĢ¦]

1.32

oths

1.31

 thousands

1.24

 hundreds

1.23

Activations Density 4.437%

references to notable historical figures and events

No Comments

No Known Activations