INDEX

Explanations

questions or statements relating to sharing information or knowledge

oai_token-act-pair · gpt-3.5-turbo

inquiries that start with "Do you know" or "Did you know," suggesting a focus on interesting or intriguing information

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

yssey

-0.69

imil

-0.68

 Detail

-0.67

sav

-0.66

href

-0.65

abilia

-0.64

bery

-0.64

nature

-0.63

ixtape

-0.62

inia

-0.62

POSITIVE LOGITS

?:

1.22

?'

1.18

...?

1.17

1.15

?"

1.13

?".

1.06

?)

1.05

?).

1.05

?!

1.05

?'"

1.05

Activations Density 0.119%

questions or statements relating to sharing information or knowledge

inquiries that start with "Do you know" or "Did you know," suggesting a focus on interesting or intriguing information

No Comments

No Known Activations