INDEX

Explanations

phrases or words related to denouncing or criticizing

oai_token-act-pair · gpt-3.5-turbo

references to the term "den," particularly in contexts related to groups or individuals

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

OTA

-0.71

OSE

-0.69

NetMessage

-0.67

Touch

-0.67

 Mobility

-0.66

 epilepsy

-0.64

Interstitial

-0.64

ruits

-0.64

Luck

-0.63

ragon

-0.63

POSITIVE LOGITS

den

4.05

Den

1.92

den

1.88

Den

1.75

 denial

1.41

der

1.34

DEN

1.27

dem

1.19

 unden

1.18

 disav

1.15

Activations Density 0.014%

phrases or words related to denouncing or criticizing

references to the term "den," particularly in contexts related to groups or individuals

No Comments

No Known Activations