INDEX

Explanations

references to specific quantities or items

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

ending

-1.55

agraph

-1.52

actic

-1.51

thirds

-1.50

ends

-1.49

essential

-1.48

onomy

-1.47

itional

-1.46

ative

-1.43

osity

-1.42

POSITIVE LOGITS

±

2.28

¿

2.19

ĻĤ

2.18

¹

2.03

ī

2.02

½

2.02

«

2.01

Ľ

1.86

»

1.84

´

1.84

Activations Density 0.221%

references to specific quantities or items

No Comments

No Known Activations