INDEX

Explanations

references to specific projects or initiatives

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.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

ĥ½

-1.99

oxacin

-1.74

unnumbered

-1.63

¨

-1.60

¬

-1.60

coming

-1.58

abouts

-1.57

thems

-1.56

etheless

-1.53

§

-1.52

POSITIVE LOGITS

ground

1.77

emis

1.76

craft

1.63

master

1.61

 Manager

1.58

smen

1.57

 underway

1.57

manship

1.54

IFIC

1.52

ura

1.51

Activations Density 0.072%

references to specific projects or initiatives

No Comments

No Known Activations