INDEX

Explanations

names of individuals

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

Interstitial

-1.13

ãĥ¼ãĥĨãĤ£

-0.95

CLASSIFIED

-0.95

EMENT

-0.93

ãģ®éŃĶ

-0.93

 spirited

-0.88

CEPT

-0.88

ITED

-0.87

ãģ¦

-0.86

LEASE

-0.85

POSITIVE LOGITS

assic

1.46

 Klux

1.25

oser

1.24

itsch

1.18

adium

1.17

utz

1.16

osing

1.15

azo

1.13

appa

1.09

usk

1.09

Activations Density 5.393%

names of individuals

No Comments

No Known Activations