INDEX

Explanations

phrases related to numerical values and groups

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

atown

-1.12

biz

-1.05

osate

-1.01

bye

-1.01

tch

-1.00

ĺħ

-0.98

uterte

-0.95

zilla

-0.94

fw

-0.93

isoft

-0.93

POSITIVE LOGITS

teenth

1.32

 digits

1.24

 sides

1.23

ieth

1.16

teen

1.15

 facets

1.14

 phases

1.14

 iterations

1.13

 branches

1.12

 pillars

1.12

Activations Density 0.939%

phrases related to numerical values and groups

No Comments

No Known Activations