INDEX

Explanations

references to key properties or characteristics of various subjects

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

ivated

-1.68

 word

-1.62

 verdict

-1.50

acon

-1.47

respond

-1.47

stance

-1.46

ft

-1.43

 aloud

-1.41

 proceeding

-1.40

ivation

-1.36

POSITIVE LOGITS

à°¿

1.89

à±į

1.63

ĻĤ

1.59

cott

1.58

EGFP

1.58

indows

1.58

BET

1.52

 DAMAGE

1.48

´

1.44

 Redistributions

1.44

Activations Density 0.035%

references to key properties or characteristics of various subjects

No Comments

No Known Activations