INDEX

Explanations

references to quantities or numbers

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

instances of the word "more."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vantage

-0.87

essage

-0.77

atan

-0.74

oin

-0.70

ancial

-0.69

,,,,,,,,

-0.69

=-=-

-0.68

xtap

-0.68

ÃŁ

-0.67

}}}

-0.67

POSITIVE LOGITS

 than

2.06

 Than

1.38

than

1.34

 importantly

0.80

 then

0.78

 stringent

0.76

 mouths

0.70

 modest

0.66

 ambitious

0.63

 people

0.63

Activations Density 0.119%

references to quantities or numbers

instances of the word "more."

No Comments

No Known Activations