INDEX

Explanations

the word "this"

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

 sense

-1.85

zzo

-1.82

ctin

-1.61

ways

-1.57

ald

-1.51

 roles

-1.50

itect

-1.45

 ascribed

-1.40

 role

-1.40

 chance

-1.38

POSITIVE LOGITS

casts

1.49

 Their

1.49

blogspot

1.46

rid

1.43

opus

1.42

↵ÂłÂł

1.41

 Telescope

1.41

lasses

1.40

hor

1.38

Activations Density 0.273%

the word "this"

No Comments

No Known Activations