INDEX

Explanations

words related to actions of undoing or reversing

oai_token-act-pair · gpt-3.5-turbo

prefixes or roots related to negation or inversion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 guessing

-0.74

 characteristic

-0.73

 wont

-0.72

EngineDebug

-0.67

 sticking

-0.66

 marks

-0.64

 transitions

-0.63

 loos

-0.62

 wanting

-0.62

 Moving

-0.62

POSITIVE LOGITS

ify

1.37

create

1.32

imize

1.31

activate

1.29

itate

1.27

idate

1.25

strate

1.24

ulate

1.23

ize

1.22

igrate

1.22

Activations Density 0.256%

words related to actions of undoing or reversing

prefixes or roots related to negation or inversion

No Comments

No Known Activations