INDEX

Explanations

words related to ethical evaluation, including concepts such as morality, acceptability, appropriateness, and correctness

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.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

Kings

-0.61

ston

-0.60

quin

-0.58

-0.55

rongh

-0.53

phabet

-0.53

 Ended

-0.52

along

-0.50

bum

-0.50

illions

-0.50

POSITIVE LOGITS

to

1.01

ceivable

0.77

 practise

0.71

to

0.70

for

0.69

 folly

0.68

ifiable

0.65

²¾

0.63

To

0.62

erto

0.58

Activations Density 0.184%

words related to ethical evaluation, including concepts such as morality, acceptability, appropriateness, and correctness

No Comments

No Known Activations