INDEX

Explanations

the word "the."

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Ò

-0.95

 besides

-0.85

bg

-0.74

<?

-0.74

estamp

-0.71

ibl

-0.71

 whatsoever

-0.70

 è£ıè

-0.68

puff

-0.68

verage

-0.68

POSITIVE LOGITS

 result

1.22

 centerpiece

1.01

 slightest

0.93

 inevitable

0.91

 sole

0.88

 latter

0.88

 predominant

0.86

 usual

0.86

 tide

0.86

sun

0.84

Activations Density 0.111%

the word "the."

No Comments

No Known Activations