INDEX

Explanations

years mentioned within the range of 2005 to 2010

oai_token-act-pair · gpt-3.5-turbo

references to the year 2005

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

mia

-0.75

ebus

-0.74

swick

-0.73

eer

-0.72

yang

-0.71

icz

-0.70

worldly

-0.69

oking

-0.69

MSN

-0.68

eye

-0.67

POSITIVE LOGITS

 ILCS

1.26

ERSON

0.80

rak

0.79

%:

0.76

Thirty

0.75

astern

0.72

ASE

0.72

 Lakes

0.71

uate

0.70

tein

0.69

Activations Density 0.057%

years mentioned within the range of 2005 to 2010

references to the year 2005

No Comments

No Known Activations