INDEX

Explanations

comparative phrases indicating a relationship between two entities

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

gotten

-0.36

aquin

-0.34

fly

-0.34

ohyd

-0.34

 backlog

-0.33

gur

-0.33

hof

-0.33

azar

-0.32

warning

-0.32

gling

-0.32

POSITIVE LOGITS

 versa

0.59

 twins

0.42

 sexes

0.41

forward

0.38

imilar

0.38

ither

0.38

worldly

0.37

 sides

0.37

isSpecialOrderable

0.37

atu

0.37

Activations Density 9.308%

comparative phrases indicating a relationship between two entities

No Comments

No Known Activations