INDEX

Explanations

phrases related to putting something on hold

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the status of items or initiatives being put on hold

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

Untitled

-0.80

SourceFile

-0.78

xs

-0.71

enegger

-0.66

conom

-0.62

 Rooms

-0.61

 symp

-0.60

omez

-0.59

eni

-0.59

iber

-0.57

POSITIVE LOGITS

 pedest

1.04

 hiatus

1.00

 cooldown

0.96

 hold

0.94

 display

0.93

 indefinite

0.89

oult

0.88

 lockdown

0.88

 pause

0.87

 autop

0.85

Activations Density 0.062%

phrases related to putting something on hold

phrases indicating the status of items or initiatives being put on hold

No Comments

No Known Activations