INDEX

Explanations

occurrences of the word "Red" and its variations in context, particularly in relation to sports or teams

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 younger

-1.94

urally

-1.63

 correct

-1.53

rapeut

-1.53

oglobin

-1.50

prud

-1.50

son

-1.47

PEG

-1.46

OLOG

-1.43

 calm

-1.43

POSITIVE LOGITS

skins

2.23

uced

2.16

wood

1.89

woods

1.89

neck

1.84

leaf

1.77

ford

1.74

emption

1.72

fish

1.67

efined

1.65

Activations Density 0.049%

occurrences of the word "Red" and its variations in context, particularly in relation to sports or teams

No Comments

No Known Activations