INDEX

Explanations

phrases related to taking care of someone or something

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

 disbelief

-0.65

eele

-0.61

Kings

-0.60

onite

-0.60

assic

-0.56

 envy

-0.56

ANS

-0.56

 denial

-0.55

 veins

-0.55

 apologies

-0.55

POSITIVE LOGITS

taker

1.10

tesy

0.99

giving

0.83

tes

0.81

lessly

0.74

maid

0.72

ername

0.71

tta

0.71

fully

0.69

ful

0.68

Activations Density 7.187%

phrases related to taking care of someone or something

No Comments

No Known Activations