INDEX

Explanations

references to patents and patent-related literature

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

 batteries

-1.65

 away

-1.54

 follower

-1.49

 departing

-1.48

»

-1.44

uder

-1.44

 recharge

-1.42

 entirety

-1.42

apshot

-1.42

 shoulders

-1.41

POSITIVE LOGITS

ÑĤÐµ

1.83

ÑģÑĤÐ²Ð°

1.66

isan

1.64

ools

1.64

ceedings

1.61

iÃ³n

1.60

fileID

1.58

language

1.56

urities

1.56

ÑģÑĤÐ²Ð¾

1.55

Activations Density 0.038%

references to patents and patent-related literature

No Comments

No Known Activations