INDEX

Explanations

terms related to memory and recollection

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

«

-2.29

nered

-1.97

½

-1.92

pired

-1.85

į

-1.70

»¿

-1.67

Ń

-1.65

ainen

-1.65

µ

-1.64

-1.62

POSITIVE LOGITS

 nothing

1.54

 privacy

1.46

 please

1.44

 stolen

1.44

asting

1.42

roviral

1.39

identally

1.38

 gaps

1.37

giving

1.36

ently

1.36

Activations Density 0.024%

terms related to memory and recollection

No Comments

No Known Activations