INDEX

Explanations

high-profile figures and events in news articles

oai_token-act-pair · gpt-3.5-turbo

terms associated with high visibility or prominence, often in public or media contexts

oai_token-act-pair · gpt-3.5-turbo Triggered by @danbraun

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arcity

-0.53

shall

-0.49

iph

-0.48

anus

-0.48

aves

-0.47

rates

-0.47

ensibly

-0.47

ften

-0.47

ope

-0.47

ieves

-0.46

POSITIVE LOGITS

 endorsements

0.63

 bearer

0.63

 casualty

0.59

 celebrity

0.58

 contender

0.56

 endors

0.56

 endorsement

0.53

 engagements

0.52

 accomplishments

0.51

 celebrities

0.50

Activations Density 12.886%

high-profile figures and events in news articles

terms associated with high visibility or prominence, often in public or media contexts

No Comments

No Known Activations