INDEX

Explanations

verbs related to movement and action

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

making

-0.59

yet

-0.56

 possessions

-0.56

dylib

-0.56

worth

-0.55

 Maxim

-0.55

 attire

-0.53

 mosquito

-0.52

MPG

-0.51

stones

-0.51

POSITIVE LOGITS

 recons

0.86

Äį

0.73

 helicop

0.69

 conclud

0.69

 guiActiveUn

0.67

ividual

0.67

 abruptly

0.65

 proceeded

0.63

DragonMagazine

0.63

anyl

0.63

Activations Density 19.550%

verbs related to movement and action

No Comments

No Known Activations