INDEX

Explanations

elements related to judgments or assessments of quality

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

 accustomed

-1.70

ĳ

-1.57

atan

-1.56

ols

-1.52

ĸ

-1.49

otry

-1.48

ķ

-1.43

Pal

-1.37

 forma

-1.32

 cour

-1.32

POSITIVE LOGITS

dered

1.60

dern

1.54

itely

1.50

onian

1.45

TRODUCTION

1.43

orkshire

1.43

ively

1.42

izumab

1.42

uto

1.39

rd

1.39

Activations Density 3.076%

elements related to judgments or assessments of quality

No Comments

No Known Activations