INDEX

Explanations

instances of the word "seem."

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

 brush

-1.59

 herself

-1.58

inse

-1.56

---|---|---

-1.55

 rely

-1.46

 story

-1.44

 suit

-1.43

 answer

-1.41

ise

-1.38

 extent

-1.38

POSITIVE LOGITS

manship

1.88

hof

1.60

encial

1.59

hoc

1.56

sterdam

1.55

vu

1.54

ugin

1.51

ingly

1.50

 unable

1.47

dal

1.46

Activations Density 0.334%

instances of the word "seem."

No Comments

No Known Activations