INDEX

Explanations

security-related terms and parameters in technical documents

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

uncher

-1.80

rypt

-1.63

aser

-1.61

ikk

-1.59

ikt

-1.54

gom

-1.52

åĻ¨

-1.52

enario

-1.41

inen

-1.40

ismus

-1.38

POSITIVE LOGITS

leading

1.56

 jurisdiction

1.51

fully

1.50

eful

1.49

 Respondent

1.44

 leads

1.43

Category

1.40

dles

1.36

elves

1.35

asc

1.34

Activations Density 0.107%

security-related terms and parameters in technical documents

No Comments

No Known Activations