INDEX

Explanations

references to specific numerical values or quantities

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

usterity

-1.84

elian

-1.71

uppose

-1.63

asting

-1.54

 yours

-1.53

anning

-1.50

rely

-1.46

 gonna

-1.43

 MERCHANTABILITY

-1.42

asts

-1.41

POSITIVE LOGITS

ÅĽÄĩ

1.89

naire

1.87

sky

1.79

 Minn

1.77

ģ

1.72

ÅĽci

1.69

ska

1.69

®

1.62

strÃ¶m

1.60

face

1.59

Activations Density 0.052%

references to specific numerical values or quantities

No Comments

No Known Activations