INDEX

Explanations

phrases related to concerns with documents or legal matters

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

 swell

-0.37

bom

-0.33

thur

-0.31

 lang

-0.31

origin

-0.30

 illum

-0.30

Filename

-0.29

 emanc

-0.29

spin

-0.29

blood

-0.28

POSITIVE LOGITS

enza

0.45

ently

0.43

ances

0.42

ency

0.41

ories

0.38

ences

0.38

ory

0.38

eto

0.37

ateurs

0.37

orship

0.35

Activations Density 3.911%

phrases related to concerns with documents or legal matters

No Comments

No Known Activations