INDEX

Explanations

instances of the word "similar" or phrases indicating comparable entities or characteristics

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

eva

-1.88

 smoke

-1.47

ovo

-1.45

ev

-1.45

 Marshal

-1.42

Ã±or

-1.39

piece

-1.38

olip

-1.36

ines

-1.35

 shotgun

-1.34

POSITIVE LOGITS

 liking

1.99

 sized

1.95

ities

1.86

 thereto

1.84

áĢº

1.80

 sexes

1.70

 across

1.63

mentioned

1.62

 sentiments

1.58

 happens

1.55

Activations Density 0.090%

instances of the word "similar" or phrases indicating comparable entities or characteristics

No Comments

No Known Activations