INDEX

Explanations

instances of the word "awarded" in various contexts

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

omorphisms

-1.78

Argued

-1.75

akin

-1.75

ck

-1.66

 characterized

-1.62

Ń

-1.54

Īĺ

-1.50

uten

-1.50

omorphism

-1.50

ushed

-1.47

POSITIVE LOGITS

giving

1.94

 directions

1.69

 favour

1.61

 ganglia

1.51

 wounds

1.37

 refuge

1.34

 territory

1.31

 range

1.29

 astr

1.29

 seconds

1.28

Activations Density 4.904%

instances of the word "awarded" in various contexts

No Comments

No Known Activations