INDEX

Explanations

phrases related to political and societal structures or events

oai_token-act-pair · gpt-3.5-turbo

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enance

-0.56

ocracy

-0.53

 killers

-0.51

ById

-0.50

phabet

-0.49

 unfocusedRange

-0.48

rise

-0.48

ylum

-0.48

 cleaners

-0.48

 pronouns

-0.48

POSITIVE LOGITS

fil

0.62

advert

0.61

haps

0.53

 Tours

0.51

JECT

0.50

azed

0.50

jured

0.50

ased

0.50

jected

0.49

active

0.48

Activations Density 8.929%

phrases related to political and societal structures or events

No Comments

No Known Activations