INDEX

Explanations

words related to geographical locations

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 obser

-0.76

FTWARE

-0.75

 undermin

-0.73

¥ŀ

-0.72

 absor

-0.70

Sco

-0.67

ĸļ

-0.63

 Witcher

-0.63

 exha

-0.60

 Conquer

-0.60

POSITIVE LOGITS

roit

1.26

ilon

1.05

ernal

1.02

rix

1.01

ted

1.00

rov

0.99

rics

0.99

iquette

0.97

ropolis

0.96

ilde

0.95

Activations Density 0.021%

words related to geographical locations

No Comments

No Known Activations