INDEX

Explanations

words related to incarceration or imprisonment

oai_token-act-pair · gpt-3.5-turbo

references to "bars"

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

ENG

-0.72

 inherited

-0.70

ASE

-0.68

 Darwin

-0.67

ove

-0.65

 Liang

-0.64

Voy

-0.64

 heir

-0.63

Ily

-0.63

anza

-0.62

POSITIVE LOGITS

 bars

4.28

 Bars

3.11

bars

2.60

bar

2.26

 cafes

1.55

 pubs

1.53

bar

1.53

 bart

1.38

Bar

1.35

Bar

1.34

Activations Density 0.010%

words related to incarceration or imprisonment

references to "bars"

No Comments

No Known Activations