INDEX

Explanations

phrases related to taking or keeping notes

oai_token-act-pair · gpt-3.5-turbo

the term "notes" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

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

ntil

-0.73

abil

-0.71

arent

-0.70

venge

-0.70

oppable

-0.68

cellence

-0.67

aul

-0.66

angular

-0.65

window

-0.65

Shut

-0.64

POSITIVE LOGITS

 notes

3.89

 note

2.46

notes

2.46

 Notes

2.41

note

1.80

Notes

1.73

 notebooks

1.60

 memos

1.46

 Note

1.45

 remarks

1.43

Activations Density 0.008%

phrases related to taking or keeping notes

the term "notes" in various contexts

No Comments

No Known Activations