INDEX

Explanations

words related to tragic and negative events or conditions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iaries

-0.87

etheless

-0.86

izoph

-0.85

ipeg

-0.84

iary

-0.78

ifier

-0.75

SpaceEngineers

-0.75

atural

-0.74

iosity

-0.73

ancial

-0.73

POSITIVE LOGITS

lla

1.68

lli

1.60

cki

1.38

llo

1.37

llan

1.31

ll

1.31

Ã±o

1.29

gger

1.26

lda

1.22

vich

1.20

Activations Density 2.817%

words related to tragic and negative events or conditions

No Comments

No Known Activations