INDEX

Explanations

specific years, particularly those related to significant historical events

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

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

 suspicion

-1.71

ird

-1.50

 behold

-1.45

ĻĤ

-1.40

 dissent

-1.39

gmail

-1.38

 realm

-1.37

 privileges

-1.37

yon

-1.36

arts

-1.34

POSITIVE LOGITS

icolor

1.50

 etched

1.49

endants

1.48

 anniversary

1.46

okia

1.41

icol

1.41

ienn

1.38

uary

1.36

^[@

1.35

erals

1.34

Activations Density 0.017%

specific years, particularly those related to significant historical events

No Comments

No Known Activations