INDEX

Explanations

information related to support for media production and journalism

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

ãĥ¯ãĥ³

-0.34

ãĤ¡

-0.29

hov

-0.28

idable

-0.28

enting

-0.28

erning

-0.28

arthy

-0.28

itect

-0.28

å°Ĩ

-0.28

idated

-0.28

POSITIVE LOGITS

river

0.45

NESS

0.43

ness

0.39

dies

0.32

bys

0.31

nesses

0.30

ricks

0.30

rogens

0.30

itiz

0.28

gers

0.28

Activations Density 8.250%

information related to support for media production and journalism

No Comments

No Known Activations