INDEX

Explanations

phrases related to political discussions and current 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

Cor

-0.65

Cu

-0.64

Ïģ

-0.63

Tag

-0.61

 Elim

-0.58

 Tibet

-0.57

Squ

-0.56

 Sher

-0.56

 Corona

-0.56

Philipp

-0.55

POSITIVE LOGITS

racuse

0.88

hedral

0.75

nces

0.70

STON

0.66

abase

0.64

angan

0.64

amina

0.63

soDeliveryDate

0.63

aminer

0.63

htaking

0.62

Activations Density 28.237%

phrases related to political discussions and current events

No Comments

No Known Activations