INDEX

Explanations

technical terms and identifiers related to programming and data processing

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.1.hook_resid_post

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

fileID

-1.91

itives

-1.57

riend

-1.48

usable

-1.48

atives

-1.46

 handful

-1.43

usement

-1.38

 duplicates

-1.38

 identifiable

-1.37

 clue

-1.35

POSITIVE LOGITS

Īĺ

2.03

ilda

1.71

dorff

1.65

¸

1.65

»¿

1.54

uit

1.49

dale

1.48

leigh

1.46

nut

1.42

aber

1.41

Activations Density 1.564%

technical terms and identifiers related to programming and data processing

No Comments

No Known Activations