INDEX

Explanations

the word "Made" 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

sing

-1.86

verted

-1.59

 maternal

-1.49

gered

-1.43

mit

-1.42

asing

-1.42

ased

-1.41

 feet

-1.40

 reflection

-1.40

aught

-1.40

POSITIVE LOGITS

leine

2.20

holm

1.74

iera

1.74

ira

1.71

àµį

1.71

áŁ

1.69

heets

1.69

etable

1.66

aggio

1.66

heet

1.64

Activations Density 0.066%

the word "Made" in various contexts

No Comments

No Known Activations