INDEX

Explanations

phrases related to actions or processes

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

ullah

-0.38

ricting

-0.34

ipation

-0.33

ropolitan

-0.33

ussen

-0.32

 elevation

-0.32

ament

-0.31

stru

-0.31

ortium

-0.31

uctor

-0.30

POSITIVE LOGITS

vt

0.47

verning

0.44

Ń·

0.41

©¶æ

0.39

lems

0.38

 limp

0.37

 overboard

0.37

bye

0.36

 forth

0.35

aded

0.34

Activations Density 11.524%

phrases related to actions or processes

No Comments

No Known Activations